{"id": "e404291ad74bdaaf1136be51797ed0d76ebcfda2dd2cfc13fba885c1abdcd582", "sources": ["arxiv"], "title": "MemTrace: Probing What Final Accuracy Misses in Long-Term Memory", "abstract": "LLM agents increasingly maintain long-term memory of user facts across sessions. Yet such memory is usually evaluated by aggregating accuracy over question rows or episodes. Because this approach scores question rows independently, even when several questions probe the same fact, it cannot show how that fact behaves as conditions change. We introduce MemTrace, a benchmark whose unit of measurement is the knowledge point: a single typed fact about the user, rather than an individual question. MemTrace probes each fact along three controlled dimensions: memory age, defined by how many sessions ago the fact appeared in the history; question type, covering current state, earlier state, and trajectory of change; and evidence condition, covering present, missing, and contradicted-by-false-premise settings. Evaluating 13 memory-system configurations across four paradigms, we find that similar pooled accuracy hides different failures: recovering a fact's current and earlier states does not imply tracking how it changed, and safe abstention does not imply correcting a false premise. The dominant bottleneck is evidence use, not retrieval: when systems fail, the evidence was retrievable 10 times more often than it was missing. These results suggest that improving long-term memory requires better use of reachable evidence, not simply more storage or retrieval.", "authors": ["Xianxuan Long", "Zhikai Chen", "Shenglai Zeng", "Shouren Wang", "Kai Guo", "Jiliang Tang"], "categories": ["cs.AI"], "fields_of_study": [], "published_date": "2026-06-15", "url": "https://arxiv.org/abs/2606.17328", "pdf_url": "https://arxiv.org/pdf/2606.17328v1", "arxiv_id": "2606.17328", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "1c35f195ee2726a4b934183c9eb8a1bc7960d6d30843e1eed8a99edbfdc0b1d1", "sources": ["arxiv", "semantic_scholar"], "title": "SERF: Spatiotemporal Environment and Robot Feature Map for Long-Horizon Mobile Manipulation", "abstract": "Long-horizon robot mobile manipulation requires continual reasoning about localization, environment changes, and task progress, all of which are challenging to infer from image observations alone. In this paper, we show that conditioning a mobile manipulation policy on a spatiotemporal feature map improves reasoning over long horizons. The map represents the environment and the articulated robot body as neural points in a shared latent space and is updated online from egocentric observations and proprioceptive state. We update the environment neural points using object-level rigid tracking and the robot neural points using forward kinematics. We use our spatiotemporal environment and robot feature (SERF) map as a state input to a vision-language-action (VLA) model by extracting map tokens from multiple reference frames and spatial scales, providing the policy with both local and global context. We demonstrate SERF on BEHAVIOR-1K, a benchmark for long-horizon mobile manipulation in household environments. Experiments show that the SERF VLA policy outperforms image-only baselines, reaches subgoals faster by following more direct trajectories, improves robustness to scene-configuration shifts, and recovers from object-drop failures.", "authors": ["Sunghwan Kim", "Byeonghyun Pak", "Kehan Long", "Yulun Tian", "Nikolay Atanasov"], "categories": ["cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-11", "url": "https://arxiv.org/abs/2606.12956", "pdf_url": "https://arxiv.org/pdf/2606.12956v1", "arxiv_id": "2606.12956", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "0e59872f2a812c9afa077800b0f4f6d236ee0f893871a478b149a89d70a9f1db", "sources": ["arxiv", "semantic_scholar"], "title": "Unified KV Pooling to Accelerate Long-Context LLM Serving", "abstract": "Long-context LLM serving requires offloading KV caches to host-memory and SSDs, but existing mechanisms are not designed for such long contexts. We observe significant inefficiencies in current KV caching in long contexts: high serving latency ~30.7 s, exceeding the typical TTFT requirement of 10 s by more than 3x. Our in-depth analysis explains two major reasons: (1) retrieval is serialized through host-memory and SSD, leaving other host-memory modules and SSDs underutilized, and (2) SSD-based KV retrieval spends 84% of its time in the kernel filesystem rather than actual device access. To address the problems, we propose unified KV pooling, which aggregates multiple host-memory modules and SSDs into a single logical pool and distributes KV caches across devices based on their bandwidth. To eliminate the filesystem overhead, we design KV-passthrough, which bypasses the kernel filesystem and directly accesses SSD-resident KV caches from user space via SPDK. Across evaluations on LLaMA 3.1-8B, GPT-OSS-20B, and Qwen3-30B-A3B, unified KV pooling reduces TTFT in long-contexts ~4.1x over state-of-the-art techniques, all making under 10 s. It also reduces blocked I/O time by up to 23.2x by eliminating filesystem overhead.", "authors": ["Minchul Kang", "Changyong Shin", "Jinwoo Jeong", "Jaerim Park", "Woohyun Kim", "Bonyul Gu", "Dongwoo Kang", "Gyeongsik Yang", "Chuck Yoo"], "categories": ["cs.AR"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-10", "url": "https://arxiv.org/abs/2606.14779", "pdf_url": "https://arxiv.org/pdf/2606.14779v1", "arxiv_id": "2606.14779", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "f237f3032c183164f6e9ef474274e99437fb3d91179fb5ced7d13f71a233a5eb", "sources": ["arxiv", "semantic_scholar"], "title": "AHA-WAM:Asynchronous Horizon-Adaptive World-Action Modeling with Observation-Guided Context Routing", "abstract": "World-action models have emerged as a promising paradigm for robot manipulation, jointly modeling visual scene dynamics and actions to inject physical priors into policy learning. However, existing world-action models couple world prediction and action execution at the same temporal resolution, forcing the world branch to model near-term frame variations that are redundant and weakly informative. We posit that strictly binding world prediction and action execution to the same temporal rhythm may underutilize the potential of the video branch for embodied control. Therefore, we propose AHA-WAM, an Asynchronous Horizon-Adaptive World-Action Model built on a dual Diffusion Transformer (DiT) architecture that reorganizes world-action modeling around this temporal asymmetry. AHA-WAM instantiates the video DiT as a low-frequency world planner that maintains rolling key-value memory over past observations and exposes reusable layerwise latent context encoding long-horizon scene evolution, while a high-frequency action DiT executes short action chunks in closed loop by querying this context through layerwise joint attention. To support asynchronous execution, we introduce horizon-adaptive offset training and Observation-Guided Video-Context Routing (OVCR), which together let the action expert exploit long-horizon world context while remaining responsive to real-time execution state without rerunning the video DiT. Experiments on RoboTwin and real-world manipulation tasks show that AHA-WAM achieves state-of-the-art performance without any robot-data pretraining, attaining 92.80% average success on RoboTwin and 78.3% success across 4 real-world tasks, while reaching 24.17 Hz closed-loop control with a 4.59x speedup over Fast-WAM.", "authors": ["Jisong Cai", "Long Ling", "Shiwei Chu", "Zhongshan Liu", "Jiayue Kang", "Zhixuan Liang", "Wenjie Xu", "Yinan Mao", "Weinan Zhang", "Xiaokang Yang", "Ru Ying", "Ran Zheng", "Yao Mu"], "categories": ["cs.RO", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-08", "url": "https://arxiv.org/abs/2606.09811", "pdf_url": "https://arxiv.org/pdf/2606.09811v1", "arxiv_id": "2606.09811", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "5ea3182923a447700ff7de9a75dc9d912a92dbc16a5d4fdefccb458d42d9e167", "sources": ["arxiv", "semantic_scholar"], "title": "YouZhi: Towards High-Concurrency Financial LLMs via Adaptive GQA-to-MLA Transition", "abstract": "Large language models (LLMs) drive significant financial innovations, yet their high-concurrency deployment is severely bottlenecked by KV cache memory overhead, which inflates infrastructure costs and throttles scalability. To address this, we propose YouZhi-LLM, a highly efficient financial LLM empowered by a comprehensive structural transition and training pipeline natively built on the Huawei Ascend ecosystem. At its algorithmic core, YouZhi-LLM features a layer-adaptive GQA-to-MLA transition framework that dynamically assigns per-layer FreqFold sizes, maximizing KV-cache compression while minimizing perplexity degradation. To recover representation capacity and inject domain expertise, the Ascend-based training pipeline seamlessly integrates generalized knowledge distillation with financial-specific supervised fine-tuning. Evaluations demonstrate the superiority of this systematic approach, with the adaptive transition reducing perplexity degradation by up to 35% over uniform baselines. Crucially, when evaluated on Ascend NPUs via vLLM-Ascend, the massive KV-cache reduction translates directly into deployment efficiency. Compared to their respective base models, YouZhi-7B yields a 12.3% improvement in average financial benchmark score alongside a 2.69$\\times$ increase in maximum concurrency; similarly, YouZhi-14B achieves a 7.0% accuracy gain and a 2.43$\\times$ concurrency boost, establishing a new paradigm for cost-effective, high-throughput financial inference.", "authors": [" PSBC LLM Team", " Huawei LLM Team", "Ruihan Long", "Junjie Wu", "Tianan Zhang", "Duo Zhang", "Yaozong Wu", "Jinbin Fu", "Chang Liu", "Zhentao Tang", "Wenshuang Yang", "Xin Wang", "Zhihao Song", "Ning Huang", "Wenjing Xu", "Shuai Zong", "Shupei Sun", "Sen Wang", "Jing Hu", "Bin Wang", "Xinyu Wang", "Junkui Ju", "Zequn Ding", "Jie Ran", "Man Luo", "Shixiong Kai", "Linkai Hou", "Kaichao Liang", "Hu Zhao", "Yang Zhao", "Shucheng Lin", "Wei Yu", "Chenghan Jiang", "Jingjing Ding", "Jiahui Zhang", "Tian Jin", "Yuhang Zhang", "Dong Guo", "Wei Sun", "Jun Xie", "Jianwei Li", "Lei Cao", "Pei Li", "Jiabin Li", "Jia Yuan", "Rui Yuan", "Jing Zhu", "Mingxuan Yuan", "Zhangcheng Lv", "Xin Jiang", "Xiuhong Fei", "Xiaozhe Ren", "Yulong Li", "Zhipeng Zhang", "Hang Wang", "Zhaohui Xu", "Rui Zhao", "Yibo He", "Xinzhuang Niu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-04", "url": "https://arxiv.org/abs/2606.05868", "pdf_url": "https://arxiv.org/pdf/2606.05868v1", "arxiv_id": "2606.05868", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "a1e2b526ef175f3378540718c9054bdf942b9e4bd6c77f12afb45ee7a5a5fb84", "sources": ["arxiv", "semantic_scholar"], "title": "Dense Contexts Are Hard Contexts: Lexical Density Limits Effective Context in LLMs", "abstract": "Input length and the position of relevant information are widely cited as the primary causes of degraded LLM long-context performance. Here, we study lexical density -- the rate at which a context introduces distinct information -- as a third, largely overlooked factor that systematically reduces the effective context window of LLMs. We quantify the impact of lexical density on open-weight LLMs (9B-685B) using three \"find-the-needle\" style benchmarks with identical length (~12k tokens) and controlled needle position, but increasing density of information. We observe a sharp performance collapse in higher-density benchmarks: models that are near-perfect in sparse contexts drop below 60% retrieval score on denser ones. To rule out task-type confounds, we vary and control the density within each benchmark while keeping all other properties unchanged. Reducing density generally restores performance, especially in the high-density regimes where degradation appears. These results show that effective context capacity is a function of lexical density, with direct implications for real-world LLM systems operating on compact, information-rich inputs.", "authors": ["Giovanni Dettori", "Matteo Boffa", "Danilo Giordano", "Idilio Drago", "Marco Mellia"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-04", "url": "https://arxiv.org/abs/2606.06203", "pdf_url": "https://arxiv.org/pdf/2606.06203v1", "arxiv_id": "2606.06203", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "b17032356cad3b28cfd5ee622d61801a67102ec0cddc842acccc425f30c6d0a4", "sources": ["arxiv", "semantic_scholar"], "title": "TokenMizer: Graph-Structured Session Memory for Long-Horizon LLM Context Management", "abstract": "Large language model (LLM) deployments for long-horizon tasks face a fundamental constraint: context windows are finite while productive work sessions are not. When history exceeds the Maximum Effective Context Window (MECW), critical structured information - architectural decisions, task transitions, file histories - is silently discarded. Existing mitigations treat history as flat text, destroying the relational structure that makes sessions resumable. We present TokenMizer, an open-source proxy system that models LLM session history as a typed knowledge graph. The schema defines 14 node types and 7 edge types. A hybrid extraction pipeline populates the graph incrementally, while a three-tier checkpoint system serializes it into compact resume blocks. An 8-layer compression pipeline reduces context overhead, and a semantic cache reduces repeated-query latency. Evaluated on a controlled benchmark of 21 sessions spanning 5 domains, TokenMizer demonstrates significant token economy. It produces resume blocks averaging 78 tokens (range: 42-124) - 2x smaller than evaluated baselines (159-170 tokens) - while achieving higher decision recall (+9-17 percentage points). Crucially, baselines only preserve that a technology was mentioned; TokenMizer preserves the rationale. Across all sessions, TokenMizer achieves mean task recall 51.0%, decision recall 46.6%, and file recall 58.7%. Variance reflects domain heterogeneity: explicit imperative phrasing (software engineering) scores higher than implicit reasoning (research). Ablation studies show fuzzy label matching is the dominant improvement factor (+33 pp task recall). The heuristic compression achieves 47.3% token reduction with zero external dependencies. TokenMizer provides a queryable alternative to text-retention baselines at half the token cost.", "authors": ["Shweta Mishra"], "categories": ["cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-04", "url": "https://arxiv.org/abs/2606.06337", "pdf_url": "https://arxiv.org/pdf/2606.06337v1", "arxiv_id": "2606.06337", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Shweta-Mishra-ai/tokenmizer", "venue": null, "quality_score": 0.65}
{"id": "3efa0fdc00a0e4150bb839ee3809a71d68049f847528dea13c97074daf4ec915", "sources": ["arxiv", "semantic_scholar"], "title": "Discourse-Role Labels as Presentation-Time Variables for Context Use in Language Models", "abstract": "Context-augmented language model systems often wrap supplied content with labels such as Reference:, Evidence:, Instruction:, Note:, or Example:, but the effect of these labels on reader-model behavior remains underexplored. We introduce a paired fixed-content probe over 500 MMLU-Pro items: each item receives the same misleading answer-bearing assertion under different discourse-role labels, and adoption is measured by whether the model outputs the injected wrong option. Across GPT-5.5, DeepSeek V4 Pro, Llama-3-8B-Instruct, and Qwen2.5-7B-Instruct, Misleading Adoption Rate shifts by 56-84 percentage points. Binding or source-like labels such as Instruction: and Reference: produce high adoption, whereas Example: consistently suppresses it. Paired tests, bootstrap intervals, final-instruction ablations, and Qwen final-step log-probability probes support a label-conditioned candidate preference. Boundary probes show where the effect weakens or persists: arithmetic tasks reduce adoption, passage-shaped external context preserves smaller label gaps, short-answer evaluation rules out option-letter copying, and nested-label conflicts suggest that illustrative framing can delimit adoption scope. A 200-case single-author manual audit confirms that the short-answer contrasts are stable under conservative adjudication. The resulting claim is bounded but practical: context-utilization and reader-side RAG benchmarks should report and control wrapper labels, because presentation choices can change measured reliance on supplied context.", "authors": ["Jianguo Zhu", "Xiangmei Li", "Wenjie Liu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-02", "url": "https://arxiv.org/abs/2606.04109", "pdf_url": "https://arxiv.org/pdf/2606.04109v2", "arxiv_id": "2606.04109", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "6eb6c898f43cade2472b10c0f271b1dde0a8672b452f9dcac14a90e600ec0519", "sources": ["arxiv", "semantic_scholar"], "title": "Rethinking the Role of Positional Encoding: Sliding-Window Transformers without PE Remain Turing Complete", "abstract": "Positional encoding (PE) is widely viewed as necessary for transformers to process ordered sequences: without them, the next-token map appears permutation-invariant in its context tokens. This intuition underlies all prior universality results, which rely on positional information to prove that transformers with chain-of-thought can perform arbitrary computation, i.e., they are Turing complete. We revisit this belief in the regime most relevant to long-form reasoning, where generation proceeds through a finite sliding context window. Our opening perception is that the window mechanism itself (mildly) breaks the permutation symmetry. To distill and precisely capture the degree of this added expressiveness, we introduce an abstract autoregressive model, the HIST model, in which each update depends only on constant-size internal state and the token-count histogram within the current window. We prove that this HIST model is Turing complete by showing that the evolution of the window can reveal the token that has just left the window, which suffices to simulate Turing-complete Post machines. We then construct a sliding-window transformer over a constant-size token alphabet, without PE, and show that it can simulate the HIST model. Our result demonstrates that positional encodings are not indispensable for transformers to perform universal computation: The window sliding itself already breaks permutation symmetry and captures sufficient positional information.", "authors": ["Qian Li", "Xinyu Mao", "Shang-Hua Teng"], "categories": ["cs.LG", "cs.CC"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-01", "url": "https://arxiv.org/abs/2606.01532", "pdf_url": "https://arxiv.org/pdf/2606.01532v2", "arxiv_id": "2606.01532", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "51a418865f6983f13e05e889dc0522a3f96575ff62c0f54b4767b417a2a65337", "sources": ["arxiv", "semantic_scholar"], "title": "Why Do Time Series Models Need Long Context Windows?", "abstract": "Modern deep learning models for forecasting groups of time series rely on increasingly longer observation windows. However, the benefit of increasing the window size is often simply attributed to capturing long-range dependencies, and broader discussion on how global forecasting models leverage input observations has been limited. In this paper, we show that forecasting groups of time series involves two objectives: (i) generative process identification (GPI), i.e., inferring the specific process generating the input sequence, and (ii) conditional forecasting (CF), i.e., predicting future values given input observations. From this perspective, optimal predictions can be interpreted as an average over plausible data-generating processes, weighted by their likelihood given the input window. This suggests another explanation for the benefits of long context windows: they reduce the uncertainty about which specific process is generating the input time series during operation. We prove that even for processes with memory length $P$, an input window size strictly larger than $P$ is necessary to achieve the minimum attainable error. Finally, we show how decoupling GPI and CF can improve computational scalability without compromising accuracy. Experiments on synthetic and real-world data validate our insights and their relevance for designing forecasting architectures.", "authors": ["Luca Butera", "Giovanni De Felice", "Andrea Cini", "Cesare Alippi"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-06-01", "url": "https://arxiv.org/abs/2606.01999", "pdf_url": "https://arxiv.org/pdf/2606.01999v1", "arxiv_id": "2606.01999", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "d5db30a2e4f53b7169428137f37e940e069c21a4932c72a471c7d9f074306915", "sources": ["arxiv", "semantic_scholar"], "title": "LongAttnComp: Cross-Family Context Compression for Long-Context Reasoning", "abstract": "As real-world applications increasingly require processing inputs of 100k+ tokens, the gap between context length and inference efficiency has become a critical bottleneck. Context compression offers a way to reduce prefill costs while preserving task accuracy. However, existing training-free attention-based methods leave substantial gaps in demanding long-context tasks such as code reasoning. We present LongAttnComp, a long-context adaptation of AttnComp that fine-tunes a lightweight cross-attention scoring layer and introduces tokenlevel chunking, a token-budget top-p algorithm, positional reordering, and a formatagnostic query parser. We further design a two-stage fine-tuning recipe for the compressor: Stage 1 builds a general retrieval foundation from NIAH-style data, and Stage 2 extends it with multi-hop and reasoning data for broader long-context task coverage. On InfiniteBench Code-Debug, LongAttnComp matches or exceeds full-context accuracy, substantially outperforms training-free baselines, and transfers across four target models from three families. On LongBench v2, the two-stage recipe largely closes the Stage 1 gap on multi-document reasoning while preserving Code-Debug performance.", "authors": ["Mengmeng Ji", "Ravi Shanker Raju", "Jonathan Lingjie Li", "Chen Wu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-31", "url": "https://arxiv.org/abs/2606.01336", "pdf_url": "https://arxiv.org/pdf/2606.01336v1", "arxiv_id": "2606.01336", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "31e3870c5df401528dc775e238c269af506d00fde96d73c72a564ae634dedcc9", "sources": ["arxiv", "semantic_scholar"], "title": "Periodic RoPE for Infinite Context LLMs", "abstract": "The ability to process ultra-long contexts is crucial for large language models (LLMs) to perform long-horizon tasks. While recent efforts have extended context windows to 1M and beyond, model performance degrades when sequence length exceeds the pre-trained range of positional encodings (e.g., RoPE), i.e., position exhaustion. This fundamental limitation must be overcome to achieve a truly infinite context. To address it, we propose Periodic RoPE (P-RoPE), a positional encoding mechanism designed to circumvent this exhaustion. It operates in conjunction with sliding window attention (SWA) to capture local dependencies and relative positions within each window. This local layer is then complemented by a global attention layer with No Positional Encoding (NoPE), enabling unbounded interaction across the entire sequence without positional constraints. By stacking these two types of layers, the model avoids the need for positional extrapolation to generalize longer and theoretically supports an infinite context window. Empirical results show that our model, MiniWin, outperforms MiniMInd with standard GPT architectures in long-context efficiency and stability. Our work provides a possible pathway toward LLMs with genuine infinite-context understanding. The code is available at \\href{https://github.com/Cominder/miniwin}{https://github.com/Cominder/miniwin}.", "authors": ["Simin Huo"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-27", "url": "https://arxiv.org/abs/2605.27980", "pdf_url": "https://arxiv.org/pdf/2605.27980v1", "arxiv_id": "2605.27980", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Cominder/miniwin}{https://github.com/Cominder/miniwin}", "venue": null, "quality_score": 0.65}
{"id": "c552cf680034fee798c7cabbde55a99f6aa22e969ce04a03069ddb68c7ddc2fa", "sources": ["arxiv", "semantic_scholar"], "title": "BCER Agent: Reliable Long-Horizon MRI Workflow Execution via Compilation, Artifact Binding, and Bounded Local Recovery", "abstract": "Many recent medical VLM and agent studies are benchmarked on 2D images or comparatively short tool-calling exchanges, whereas real MRI analysis typically demands long, interdependent pipelines that operate on 3D/4D volumetric data. Under these conditions, reactive tool-calling agents are prone to cascading breakdowns triggered by faulty intermediate references, mismatched tool arguments, and limited control over cross-step dependencies. To address this, we introduce BCER (Brain-Cerebellum-Extremity-Reflector), a controller architecture aimed at dependable long-horizon MRI workflow execution. BCER decouples high-level planning from execution and provides bounded local recovery. We assess BCER on a multi-organ MRI benchmark covering brain, prostate, and cardiac tasks with both short- and long-chain workflows, using matched task contracts across controller variants and several backbone models. Relative to reactive baselines, BCER yields consistent improvements in end-to-end execution, with the most pronounced gains observed on long-chain workflows. BCER additionally enables auditability by maintaining explicit links between final outputs and intermediate artifacts and measurements. Code and benchmark are released at https://github.com/Albertlongzi/BCER.", "authors": ["Ziyang Long", "Xinqi Li", "Junzhou Chen", "Yifan Gao", "Debiao Li", "Hsin-Jung Yang"], "categories": ["eess.IV"], "fields_of_study": ["Engineering"], "published_date": "2026-05-27", "url": "https://arxiv.org/abs/2605.29163", "pdf_url": "https://arxiv.org/pdf/2605.29163v1", "arxiv_id": "2605.29163", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Albertlongzi/BCER", "venue": null, "quality_score": 0.65}
{"id": "383bb6e4f25169577320c864636bad41346fe4215e0e9e2b12fb2e35fc482fb8", "sources": ["arxiv", "semantic_scholar"], "title": "Device Context Protocol: A Compact, Safety-First Architecture for LLM-Driven Control of Constrained Devices", "abstract": "Large language models are increasingly used as orchestrators of external tools via the Model Context Protocol (MCP), but MCP is built for software services with megabytes of memory and does not descend to the microcontrollers that dominate the long tail of physical devices. Recent work (IoT-MCP) ports MCP to edge gateways at 74 KB peak memory; this still excludes the smallest commodity MCUs and, critically, does not address the safety problem of giving an unreliable caller (an LLM that may hallucinate or be prompt-injected) direct control of physical hardware. We present the Device Context Protocol (DCP): a sub-50-byte typical frame (6-byte header + CBOR payload + optional 16-byte HMAC), a manifest schema in which capability scoping, range and type checks, dry-run evaluation, and units-as-types are protocol-layer primitives, and a host-side Bridge that rejects malformed or hallucinated calls before any byte reaches the device. Reference firmware measures 27.6 KB flash / 0.6 KB RAM on ESP32; the Python Bridge, ESP32 firmware, and a language-neutral conformance suite are MIT-licensed and public. An empirical study -- 675 tool calls produced by five LLMs across four vendors (DeepSeek, Alibaba, Zhipu, MiniMax) against six categories of adversarial prompts, with the injection category instantiating AgentDojo's attack templates -- shows DCP rejects 100% of capability-escalation attempts and 78% of prompt-injection attempts, versus 0--1% for Raw MCP and IoT-MCP, matching the expressiveness of a well-formed OpenAPI 3 schema at three orders of magnitude less firmware footprint. We position DCP as the missing layer between MCP (which is moving toward enterprise SaaS connectivity) and the physical devices it does not reach.", "authors": ["Dongxu Yang"], "categories": ["cs.NI", "cs.CR", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-24", "url": "https://arxiv.org/abs/2605.26159", "pdf_url": "https://arxiv.org/pdf/2605.26159v1", "arxiv_id": "2605.26159", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/device-context-protocol/dcp", "venue": null, "quality_score": 0.65}
{"id": "93d694deb30347c0e02071d7b97e1948a4736d9baf3948cd96747eac67dca9e4", "sources": ["arxiv", "semantic_scholar"], "title": "Positional Failures in Long-Context LLMs: A Blind Spot in Reasoning Benchmarks", "abstract": "Position-controlled evaluation is standard for retrieval tasks such as Needle-in-a-Haystack and RULER, but mainstream reasoning benchmarks do not control positional placement of target tasks in long contexts. We audit 11 long-context benchmarks and find none jointly controls task position, filler content, and context length for reasoning. An audit of four flagship long-context releases finds no main result-table entry for NIAH, RULER, or LongBench-family benchmarks, while agentic and coding benchmarks appear in main result-tables across all four. We propose Context Rot Evaluation (CRE), a controlled framework varying all three factors, and evaluate nine LLMs on GSM8K and ARC-Challenge across two rounds: an initial five-model set and four newer vendor releases. Models can drop sharply when the target task moves from end to middle, and the drop grows worse with context length for vulnerable models. MiMo-v2-Flash drops 88pp at 64K under with_solutions filler (middle accuracy 8%). Newer releases show smaller drops: at 64K, three of four stay within +/-6pp of end-position accuracy; MiMo-V2.5-Pro narrows the MiMo-v2-Flash 88pp drop to 32pp. Under questions_only_v2 filler, middle-position drops persist across all four (range -16pp to -56pp across 8K, 32K, 64K). At 8K, a diagnostic probe adding a target-task copy at the end brings middle accuracy within +/-4pp of end baseline across all nine models, consistent with a positional explanation. In the initial five-model set, 76% of middle-position errors match surrounding filler text versus 22% at the end position, consistent with filler-answer interference as a dominant error mode. These results expose a structural evaluation gap in current reasoning benchmark design and vendor evaluation practice: positional vulnerabilities that grow with context length cannot be measured when task position is not controlled.", "authors": ["Chuyifei Zhang", "Hongyu Cui", "Xiaowen Huang", "Jitao Sang"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-22", "url": "https://arxiv.org/abs/2605.23170", "pdf_url": "https://arxiv.org/pdf/2605.23170v1", "arxiv_id": "2605.23170", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "497a8671d278d8e204f3a77c46b5eb3d310e3bd7759f29ea6d225f24f1c3ae47", "sources": ["arxiv", "semantic_scholar"], "title": "Parallel Context Compaction for Long-Horizon LLM Agent Serving", "abstract": "Long-horizon LLM agents accumulate growing conversation histories that eventually exceed the model's context window. Context compaction via LLM-based summarization keeps the conversation bounded, but summarization is inherently lossy and the blocking call stalls agent inference for tens of seconds. Moreover, the operator has no fine-grained control over summary volume since prompt instructions are largely ignored, and as context grows, both the amount of output tokens the model produces and the information it retains fluctuate substantially from run to run, making the agent's retained knowledge unpredictable across runs. We introduce \\textbf{parallel compaction} for long-horizon agentic flows and characterize it against the sequential synchronous baseline across four backbones spanning 8B to 120B parameters, mixing dense and MoE architectures with reasoning and non-reasoning models, on the HotpotQA multi-hop QA and LoCoMo long-context dialogue benchmarks. Parallel compaction gives the operator fine-grained, predictable control over summary volume and enables more targeted prompt engineering per block. At matched compaction decode volume, it reduces end-to-end wall time and improves compaction throughput over the sequential baseline.", "authors": ["Musa Cim", "Burak Topcu", "Chita Das", "Mahmut Taylan Kandemir"], "categories": ["cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-22", "url": "https://arxiv.org/abs/2605.23296", "pdf_url": "https://arxiv.org/pdf/2605.23296v1", "arxiv_id": "2605.23296", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "c6f325268f898ab9ea22fa7c100adfb96c955a21b95e699db9d8f109a536d8fe", "sources": ["arxiv", "semantic_scholar"], "title": "More Context, Larger Models, or Moral Knowledge? A Systematic Study of Schwartz Value Detection in Political Texts", "abstract": "Detecting Schwartz values in political text is difficult because implicit cues often depend on surrounding arguments and fine-grained distinctions between neighboring values. We study when context and explicit moral knowledge help sentence-level value detection. Using the ValuesML/Touché ValueEval format, we compare sentence, window, and full-document inputs; no-RAG and retrieval-augmented settings with a curated moral knowledge base; supervised DeBERTa-v3-base/large encoders; and zero-shot LLMs from 12B to 123B parameters. The results show that more context is not uniformly better: full-document context improves supervised DeBERTa encoders by 3.8-4.8 macro-F1 points over sentence-only input, but does not consistently help zero-shot LLMs. Retrieved moral knowledge is more consistently useful in matched comparisons, improving each tested model family and context condition under early fusion. However, scaling from DeBERTa-v3-base to large and from 12B to larger LLMs does not guarantee gains, and simple early fusion outperforms the tested late-fusion and cross-attention RAG variants for encoders. Per-value analyses show that context and retrieval help most for socially situated or conceptually confusable values. These findings suggest that value-sensitive NLP should evaluate context, knowledge, and model family jointly rather than treating longer inputs or larger models as universal improvements.", "authors": ["Víctor Yeste", "Paolo Rosso"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-21", "url": "https://arxiv.org/abs/2605.22641", "pdf_url": "https://arxiv.org/pdf/2605.22641v3", "arxiv_id": "2605.22641", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/VictorMYeste/human-value-detection-context-rag", "venue": null, "quality_score": 0.65}
{"id": "f4bf3018956b0344e469527a33116434e59b5b8640d9e521ada80b107afd24d9", "sources": ["arxiv", "semantic_scholar"], "title": "The Efficiency Frontier: A Unified Framework for Cost-Performance Optimization in LLM Context Management", "abstract": "Large language models (LLMs) increasingly rely on long-context processing, but expanding context windows introduces substantial computational and financial costs. Existing context reduction approaches, including retrieval and memory compression methods, are typically evaluated using performance and efficiency metrics independently, limiting systematic comparison and deployment-aware decision-making. This paper introduces The Efficiency Frontier, a unified framework for cost-performance optimization in LLM context management. The framework models context strategy selection as a deployment-aware optimization problem that jointly accounts for task performance, token cost, and preprocessing reuse through amortized cost modeling. Unlike existing evaluations that compare methods in isolation, the proposed framework enables decision-oriented analysis of when different context management strategies become preferable under varying operational conditions. Evaluated on 5,000 HotpotQA instances, the framework reveals distinct operational regimes and transition boundaries between retrieval-based and preprocessing-based strategies. Results show that deployment-aware optimization reduces effective token usage by approximately 25% at comparable performance ($F1 \\approx 0.78$), while amortized memory compression achieves over 50% lower token cost relative to full-context prompting in higher-performance settings. Overall, the proposed framework provides a principled and practical foundation for evaluating and deploying scalable, efficient, and sustainable LLM systems.", "authors": ["Binqi Shen", "Lier Jin", "Hanyu Cai", "Lan Hu", "Yuting Xin"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-21", "url": "https://arxiv.org/abs/2605.23071", "pdf_url": "https://arxiv.org/pdf/2605.23071v1", "arxiv_id": "2605.23071", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "5816c4e746ef05b96975cf5b07890fab2f6065bd090728644507a8557ae7adad", "sources": ["arxiv", "semantic_scholar"], "title": "PEEK: Context Map as an Orientation Cache for Long-Context LLM Agents", "abstract": "Large language model (LLM) agents increasingly operate over long and recurring external contexts, like document corpora and code repositories. Across invocations, existing approaches preserve either the agent's trajectory, passive access to raw material, or task-level strategies. None of them preserves what we argue is most needed for repeated same-context workloads: reusable orientation knowledge (e.g., what the context contains, how it is organized, and which entities, constants, and schemas have historically been useful) about the recurring context itself. We introduce PEEK, a system that caches and maintains this orientation knowledge as a context map: a small, constant-sized artifact in the agent's prompt that gives it a persistent peek into the external context. The map is maintained by a programmable cache policy with three modules: a Distiller that extracts transferable knowledge from inference-time signals, a Cartographer that translates it into structured edits, and a priority-based Evictor that enforces a fixed token budget. On long-context reasoning and information aggregation, PEEK improves over strong baselines by 6.3-34.0% while using 93-145 fewer iterations and incurring 1.7-5.8x lower cost than the state-of-the-art prompt-learning framework, ACE. On context learning, PEEK improves solving rate and rubric accuracy by 6.0-14.0% and 7.8-12.1%, respectively, at 1.4x lower cost than ACE. These gains generalize across LMs and agent architectures, including OpenAI Codex, a production-grade coding agent. Together, these results show that a context map helps long-context LLM agents interact with recurring external contexts more accurately and efficiently.", "authors": ["Zhuohan Gu", "Qizheng Zhang", "Omar Khattab", "Samuel Madden"], "categories": ["cs.AI", "cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-19", "url": "https://arxiv.org/abs/2605.19932", "pdf_url": "https://arxiv.org/pdf/2605.19932v1", "arxiv_id": "2605.19932", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "7782628a97eff9a8ec1939ee0a84633ed66e37baf1d2335a457b7ba482af4f3b", "sources": ["arxiv", "semantic_scholar"], "title": "RoPE Distinguishes Neither Positions Nor Tokens in Long Contexts, Provably", "abstract": "We identify intrinsic limitations of Rotary Positional Embeddings (RoPE) in Transformer-based long-context language models. Our theoretical analysis abstracts away from the specific content of the context and depends only on its length. We prove that as context length increases, RoPE-based attention becomes unpredictable and loses two properties that are central to its effectiveness. First, it loses its locality bias: RoPE is no more likely to favor nearer positions than substantially farther ones. Second, it loses consistency in token relevance: a key vector that receives a higher attention score than an alternative at one position may receive a lower score at another. In both cases, the probability of failure approaches 0.5, no better than random guessing. We further prove that the attention score can remain unchanged when a key token is moved to a different position, or even replaced by a different token, indicating a failure to distinguish positions or tokens. Adjusting the RoPE base trades off distinguishing positions against distinguishing tokens but cannot preserve both at the same time. Increasing the RoPE base hyperparameter, a common practice in today's long-context models, helps distinguish different tokens, but inevitably sacrifices the ability to distinguish positions. Our empirical analysis shows that multi-head, multi-layer architectures are insufficient to overcome these limitations. Our findings suggest that fundamentally new mechanisms for encoding position and token order may be needed in future Transformer long-context language models.", "authors": ["Yufeng Du", "Phillip Harris", "Minyang Tian", "Eliu A Huerta", "Srikanth Ronanki", "Subendhu Rongali", "Aram Galstyan", "Hao Peng"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-15", "url": "https://arxiv.org/abs/2605.15514", "pdf_url": "https://arxiv.org/pdf/2605.15514v1", "arxiv_id": "2605.15514", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "ef126b35cc7f6a744dbf0633a98b0bb0545d77b6443e4bd0c9a551bf0ed20ede", "sources": ["arxiv", "semantic_scholar"], "title": "EndPrompt: Efficient Long-Context Extension via Terminal Anchoring", "abstract": "Extending the context window of large language models typically requires training on sequences at the target length, incurring quadratic memory and computational costs that make long-context adaptation expensive and difficult to reproduce. We propose EndPrompt, a method that achieves effective context extension using only short training sequences. The core insight is that exposing a model to long-range relative positional distances does not require constructing full-length inputs: we preserve the original short context as an intact first segment and append a brief terminal prompt as a second segment, assigning it positional indices near the target context length. This two-segment construction introduces both local and long-range relative distances within a short physical sequence while maintaining the semantic continuity of the training text--a property absent in chunk-based simulation approaches that split contiguous context. We provide a theoretical analysis grounded in Rotary Position Embedding and the Bernstein inequality, showing that position interpolation induces a rigorous smoothness constraint over the attention function, with shared Transformer parameters further suppressing unstable extrapolation to unobserved intermediate distances. Applied to LLaMA-family models extending the context window from 8K to 64K, EndPrompt achieves an average RULER score of 76.03 and the highest average on LongBench, surpassing LCEG (72.24), LongLoRA (72.95), and full-length fine-tuning (69.23) while requiring substantially less computation. These results demonstrate that long-context generalization can be induced from sparse positional supervision, challenging the prevailing assumption that dense long-sequence training is necessary for reliable context-window extension. The code is available at https://github.com/clx1415926/EndPrompt.", "authors": ["Han Tian", "Luxuan Chen", "Xinran Chen", "Rui Kong", "Fang Wang", "Jiamin Chen", "Jinman Zhao", "Yuchen Li", "Jiashu Zhao", "Shuaiqiang Wang", "Haoyi Xiong", "Linghe Kong", "Dawei Yin"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-14", "url": "https://arxiv.org/abs/2605.14589", "pdf_url": "https://arxiv.org/pdf/2605.14589v2", "arxiv_id": "2605.14589", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/clx1415926/EndPrompt", "venue": null, "quality_score": 0.65}
{"id": "b86ebb5c5d283a527243e6a91370e150043334d8dd92ccda36d0bf567d8c55be", "sources": ["arxiv", "semantic_scholar"], "title": "MetaBackdoor: Exploiting Positional Encoding as a Backdoor Attack Surface in LLMs", "abstract": "Backdoor attacks pose a serious security threat to large language models (LLMs), which are increasingly deployed as general-purpose assistants in safety- and privacy-critical applications. Existing LLM backdoors rely primarily on content-based triggers, requiring explicit modification of the input text. In this work, we show that this assumption is unnecessary and limiting. We introduce MetaBackdoor, a new class of backdoor attacks that exploits positional information as the trigger, without modifying textual content. Our key insight is that Transformer-based LLMs necessarily encode token positions to process ordered sequences. As a result, length-correlated positional structure is reflected in the model's internal computation and can be used as an effective non-content trigger signal. We demonstrate that even a simple length-based positional trigger is sufficient to activate stealthy backdoors. Unlike prior attacks, MetaBackdoor operates on visibly and semantically clean inputs and enables qualitatively new capabilities. We show that a backdoored LLM can be induced to disclose sensitive internal information, including proprietary system prompts, once a length condition is satisfied. We further demonstrate a self-activation scenario, where normal multi-turn interaction can move the conversation context into the trigger region and induce malicious tool-call behavior without attacker-supplied trigger text. In addition, MetaBackdoor is orthogonal to content-based backdoors and can be composed with them to create more precise and harder-to-detect activation conditions. Our results expand the threat model of LLM backdoors by revealing positional encoding as a previously overlooked attack surface. This challenges defenses that focus on detecting suspicious text and highlights the need for new defense strategies that explicitly account for positional triggers in modern LLM architectures.", "authors": ["Rui Wen", "Mark Russinovich", "Andrew Paverd", "Jun Sakuma", "Ahmed Salem"], "categories": ["cs.CR", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-14", "url": "https://arxiv.org/abs/2605.15172", "pdf_url": "https://arxiv.org/pdf/2605.15172v1", "arxiv_id": "2605.15172", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "f6ac71bf4effc1c63d60fc7d6cebf4345752e7d28243bdab928215bb8ce40d07", "sources": ["arxiv", "semantic_scholar"], "title": "Correctness-Aware Repository Filtering Under Maximum Effective Context Window Constraints", "abstract": "Context window efficiency is a practical constraint in large language model (LLM)-based developer tools. Paulsen [12] shows that all tested models degrade in accuracy well before their advertised context limits the Maximum Effective Context Window (MECW) which makes context construction a quality problem, not just a cost one. Modern software repositories routinely contain large non-code artifacts compiled datasets, binary model weights, minified JavaScript bundles, and gigabyte-scale log files that overflow the context window and push out task-relevant source code. We present a correctness-aware context hygiene framework: a pre-execution, size-based heuristic filter that intercepts repository scans before tokenization, using only OS-level stat() metadata with sub-millisecond overhead. Semantic retrieval approaches such as RepoCoder, GraphRAG, and AST-based chunking require index construction and query-time inference before any filtering decision is reached. Our framework, by contrast, requires no indexing and operates at <0.01 ms per file decision. Across 10 real open-source repositories (22,046 files, 5 languages), the proposed SizeFilter at θ=1 MB achieves 79.6% (\\pm13.2%) mean token reduction at 0.30 ms overhead: the HybridFilter achieves 89.3% (\\pm9.0%) the lowest variance of any filter evaluated. A token-density study across 2,688 files confirms a strong linear correlation (Pearson r=0.997, k=0.250 tokens/byte). A limited-scope evaluation (18 tasks, CodeLlama-7B-Instruct) yields 72% file-level accuracy under filtering versus 25% at baseline; hallucination frequency declines from 61% to 17%. All code and data are released for reproducibility.", "authors": ["Shweta Mishra"], "categories": ["cs.SE", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-14", "url": "https://arxiv.org/abs/2605.14362", "pdf_url": "https://arxiv.org/pdf/2605.14362v1", "arxiv_id": "2605.14362", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": null, "quality_score": 0.65}
{"id": "0874101642345ba3c1df1f096bdd66e9ec846e1e2c2c8529563fa69185497108", "sources": ["arxiv", "semantic_scholar"], "title": "Training Long-Context Vision-Language Models Effectively with Generalization Beyond 128K Context", "abstract": "Long-context modeling is becoming a core capability of modern large vision-language models (LVLMs), enabling sustained context management across long-document understanding, video analysis, and multi-turn tool use in agentic workflows. Yet practical training recipes remain insufficiently explored, particularly for designing and balancing long-context data mixtures. In this work, we present a systematic study of long-context continued pre-training for LVLMs, extending a 7B model from 32K to 128K context with extensive ablations on long-document data. We first show that long-document VQA is substantially more effective than OCR transcription. Building on this observation, our ablations further yield three key findings: i) for sequence-length distribution, balanced data outperforms target-length-focused data (e.g., 128K), suggesting that long-context ability requires generalizable key-information retrieval across various lengths and positions; ii) retrieval remains the primary bottleneck, favoring retrieval-heavy mixtures with modest reasoning data for task diversity; and iii) pure long-document VQA largely preserves short-context capabilities, suggesting that instruction-formatted long data reduces the need for short-data mixing. Based on these findings, we introduce MMProLong, obtained by long-context continued pre-training from Qwen2.5-VL-7B with only a 5B-token budget. MMProLong improves long-document VQA scores by 7.1% and maintains strong performance at 256K and 512K contexts beyond its 128K training window, without additional training. It further generalizes to webpage-based multimodal needle retrieval, long-context vision-text compression, and long-video understanding without task-specific supervision. Overall, our study establishes a practical LongPT recipe and an empirical foundation for advancing long-context vision-language models.", "authors": ["Zhaowei Wang", "Lishu Luo", "Haodong Duan", "Weiwei Liu", "Sijin Wu", "Ji Luo", "Shen Yan", "Shuai Peng", "Sihang Yuan", "Chaoyi Huang", "Yi Lin", "Yangqiu Song"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-13", "url": "https://arxiv.org/abs/2605.13831", "pdf_url": "https://arxiv.org/pdf/2605.13831v1", "arxiv_id": "2605.13831", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "b791e7395a39d5fbe8166baccd4a8b321a96cbae824eace0a34ccbdb03f08d54", "sources": ["arxiv", "semantic_scholar"], "title": "Where Does Long-Context Supervision Actually Go? Effective-Context Exposure Balancing", "abstract": "Long-context adaptation is often viewed as window scaling, but this misses a token-level supervision mismatch: in packed training with document masking, each target token's effective context remains short. We introduce EXACT, a supervision-allocation objective that assigns extra weight to long effective-context targets by inverse frequency within the long tail. Across seven Qwen/LLaMA CPT configurations, EXACT improves all 28 trained/extrapolated NoLiMa and RULER comparisons. On Qwen2.5-0.5B, NoLiMa improves by +10.09 (trained) and +5.34 (extrapolated); RULER by +10.69 and +5.55. On LLaMA-3.2-3B, RULER improves by +17.91 and +16.11. Standard QA/reasoning are preserved (+0.24 macro change across six benchmarks). A distance-resolved probe shows gains arise when evidence is thousands of tokens away, while short cases remain unchanged. Results support a supervision-centric thesis: long-context adaptation depends on how strongly training supervises long-context predictions.", "authors": ["Jinchang Zhu", "Jindong Li", "Chengyu Zou", "Rong Fu", "Chao Wang", "Haowei He", "Menglin Yang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-11", "url": "https://arxiv.org/abs/2605.10544", "pdf_url": "https://arxiv.org/pdf/2605.10544v1", "arxiv_id": "2605.10544", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "22da49e90d9bff6dd3793d4da3ed5bbe0dee0d8d7fd3184aa21f104a0dca4a2f", "sources": ["arxiv", "semantic_scholar"], "title": "Drift is a Sampling Error: SNR-Aware Power Distributions for Long-Horizon Robotic Planning", "abstract": "Despite rapid progress in Vision-Language-Action (VLA) models for robotic control, instruction drift remains a persistent failure mode in long-horizon tasks. This paper reconceptualizes this phenomenon, positing that instruction drift is fundamentally a systematic sampling error: local greedy sampling is prone to collapsing into \"Negative Pivotal Windows\"--irreversible local optima with high local probability that sever global success pathways. To address this, we propose Context-Aware Power Sampling (CAPS), a training-free inference-time computation framework. CAPS leverages power distributions to sharpen global trajectory probabilities, enabling lookahead search over the model's conditional generative trajectory distribution. Furthermore, we introduce a metacognitive control mechanism based on Signal-to-Noise Ratio (SNR). This mechanism triggers adaptive MCMC search solely when drift risk is detected, enabling a dynamic transition from \"intuitive fast thinking\" to \"rational slow search.\" Experiments on RoboTwin, Simpler-WindowX, and Libero-long benchmarks show that CAPS achieves substantial improvements over strong baselines, including OpenVLA and TACO, without parameter updates. These results support the effectiveness of adaptive inference-time computation for improving long-horizon robustness in embodied control.", "authors": ["Kewei Chen", "Yayu Long", "Mingsheng Shang"], "categories": ["cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-10", "url": "https://arxiv.org/abs/2605.09537", "pdf_url": "https://arxiv.org/pdf/2605.09537v1", "arxiv_id": "2605.09537", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "75f53c631b1f93b4fe9dec2699b9f5ad2deb4a720ca7e0165abead8142d3af45", "sources": ["arxiv", "semantic_scholar"], "title": "HexiSeq: Accommodating Long Context Training of LLMs over Heterogeneous Hardware", "abstract": "Long-context training of large language models (LLMs) is commonly distributed with Context Parallelism (CP) and Head Parallelism (HP), but existing training systems largely assume homogeneous GPU meshes. This paper extends CP and HP to heterogeneous GPU clusters with mixed GPU models and non-uniform network bandwidths, a common setting in production training. We introduce HexiSeq, a system that supports fully asymmetric CP--HP partitioning by assigning sequence shards and attention heads according to device compute, memory, and communication capabilities. We formalize heterogeneous CP--HP allocation as a constrained optimization problem and develop an efficient hierarchical scheduler for finding optimal schedules. We evaluate HexiSeq against state-of-the-art CP and HP baselines on both real and simulated heterogeneous clusters. Across models from 3B to 70B parameters and context lengths up to one million tokens, HexiSeq improves throughput by $1.11\\times$ on average and up to $1.19\\times$ on mixed H100--A100 testbeds, and by $1.36\\times$ on average and up to $1.72\\times$ in simulations with 32--128 GPUs spanning up to four GPU models. On FLOP-comparable pairs against homogeneous clusters, HexiSeq reaches throughput close to the strongest homogeneous baseline, showing that heterogeneous clusters can be used efficiently for long-context LLM training.", "authors": ["Yan Liang", "Youhe Jiang", "Ran Yan", "Binhang Yuan", "Wei Wang", "Chuan Wu"], "categories": ["cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-08", "url": "https://arxiv.org/abs/2605.07569", "pdf_url": "https://arxiv.org/pdf/2605.07569v1", "arxiv_id": "2605.07569", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "ae2002bd2c7f42a54fa39bb1c4332334ef111bb2ef4d2488dac3de3a24c4f533", "sources": ["arxiv", "semantic_scholar"], "title": "A$^2$RD: Agentic Autoregressive Diffusion for Long Video Consistency", "abstract": "Synthesizing consistent and coherent long video remains a fundamental challenge. Existing methods suffer from semantic drift and narrative collapse over long horizons. We present A$^2$RD, an Agentic Auto-Regressive Diffusion architecture that decouples creative synthesis from consistency enforcement. A$^2$RD formulates long video synthesis as a closed-loop process that synthesizes and self-improves video segment-by-segment through a Retrieve--Synthesize--Refine--Update cycle. It comprises three core components: (i) Multimodal Video Memory that tracks video progression across modalities; (ii) Adaptive Segment Generation that switches among generation modes for natural progression and visual consistency; and (iii) Hierarchical Test-Time Self-Improvement that self-improves each segment at frame and video levels to prevent error propagation. We further introduce LVBench-C, a challenging benchmark with non-linear entity and environment transitions to stress-test long-horizon consistency. Across public and LVBench-C benchmarks spanning one- to ten-minute videos, A$^2$RD outperforms state-of-the-art baselines by up to 30% in consistency and 20% in narrative coherence. Human evaluations corroborate these gains while also highlighting notable improvements in motion and transition smoothness.", "authors": ["Do Xuan Long", "Yale Song", "Min-Yen Kan", "Tomas Pfister", "Long T. Le"], "categories": ["cs.CV", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-07", "url": "https://arxiv.org/abs/2605.06924", "pdf_url": "https://arxiv.org/pdf/2605.06924v1", "arxiv_id": "2605.06924", "doi": null, "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "2eb4c473a924f72b2870b3edb5140d28111b1a47d0311ee56c4b2e6c0c1f80e3", "sources": ["arxiv", "semantic_scholar"], "title": "A Language for Describing Agentic LLM Contexts", "abstract": "Large language models are increasingly used within larger systems (\"LLM agents\"). These make a sequence of LLM calls, each call providing the LLM with a combination of instructions, observations, and interaction history. The design of the encoded information and its structure play a central role in the quality of the resulting system, leading to efforts spent on context engineering. It is therefore critical to communicate the composition of the LLM context in a system, and how it evolves over time. Yet, no standard exists for doing so: context construction is typically conveyed through informal prose, ad hoc diagrams, or direct inspection of code, none of which precisely capture how a prompt evolves across interaction steps or how two context representation strategies differ. To remedy this, we introduce the Agentic Context Description Language (ACDL), a language for specifying the structure and dynamics of LLM input contexts in a precise, readable, and standard manner, along with visualizations. ACDL provides constructs for specifying context aspects such as role message sequences, dynamic content, time-indexed references, and conditional or iterative structure, capturing the full architecture of a prompt independently of any particular implementation. ACDL diagrams can be hand drawn on a whiteboard, or written in formal language which can then be rendered. We describe the language, demonstrate it by documenting several existing systems and their variants, and encourage the community to adopt it for describing LLM systems context, both in day-to-day communication and in papers. Tooling, examples and documentation are available at www.acdlang.org.", "authors": ["Noga Peleg Pelc", "Gal A. Kaminka", "Yoav Goldberg"], "categories": ["cs.AI", "cs.CL", "cs.MA", "cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-03", "url": "https://arxiv.org/abs/2605.01920", "pdf_url": "https://arxiv.org/pdf/2605.01920v1", "arxiv_id": "2605.01920", "doi": "10.1145/3786335.3813126", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "CAIS '26: ACM Conference on AI and Agentic Systems, May 2026, San Jose, CA, USA", "quality_score": 0.55}
{"id": "f260577b473d5f16dc842cffd962a69558bc9e9013030b84446d01cce6c8b0b9", "sources": ["arxiv", "semantic_scholar"], "title": "Accelerating Long-Tail Generation in Synchronous RLHF Training via Adaptive Tensor Parallelism", "abstract": "Reinforcement Learning from Human Feedback (RLHF) has become a key post-training paradigm for improving model quality. However, the synchronous three-stage RLHF pipeline is often bottlenecked by the generation stage, where response-length skew causes the effective batch size to shrink rapidly during decoding, leaving GPUs underutilized while a few long responses remain unfinished. Mainstream frameworks employ a static tensor parallelism (TP) configuration that cannot adapt to changing batch characteristics, leaving substantial performance headroom unexplored. We propose PAT, an adaptive TP method that dynamically reconfigures TP during the generation stage of each RLHF iteration. PAT introduces two key techniques. First, a predictor-guided online reconfiguration method decides both the reconfiguration point and the target TP configuration based on offline profiling, triggering reconfiguration only when the predicted latency benefit outweighs the reconfiguration overhead. Second, a lightweight online reconfiguration mechanism updates only the states and layouts affected by TP changes: it adapts unfinished decoding states through a cost-model-based choice between KV-cache migration and recomputation, performs in-place weight resharding, and reuses cached communication groups. We implement PAT on top of SGLang and integrate it with the VeRL framework. Evaluations on LLaMA3.1-8B and Qwen3-14B using DeepScaleR show that PAT reduces generation latency by up to 34.6% and end-to-end RLHF training iteration latency by up to 27.2% compared to the original VeRL setup.", "authors": ["Long Zhao", "Qinghe Wang", "Jiaan Zhu", "Youhui Bai", "Zewen Jin", "Chaoyi Ruan", "Shengnan Wang", "Cheng Li"], "categories": ["cs.AI", "cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-03", "url": "https://arxiv.org/abs/2605.23945", "pdf_url": "https://arxiv.org/pdf/2605.23945v1", "arxiv_id": "2605.23945", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "8710db2ab7139228be5e2f7f6c9da5408d83c2883c7f9d698c8bb7e9d8f4ad81", "sources": ["arxiv", "semantic_scholar"], "title": "Beyond Compaction: Structured Context Eviction for Long-Horizon Agents", "abstract": "We present Context Window Lifecycle (CWL), a context-management scheme that gives long-horizon LLM agents an effectively unbounded working horizon. As a session accumulates history, CWL keeps the context within budget through graduated, semantically-aware eviction: the agent annotates its trajectory as typed, dependency-linked episodes as work proceeds, and a deterministic, LLM-free policy evicts content in priority order within that structure when a token budget is exceeded. CWL preserves user turns and the exploratory context the agent is actively reasoning over, while aggressively shedding action episodes whose effects are already persisted in the environment, keeping active context near a stable ceiling that also avoids the performance degradation associated with very large prompts. Compared to summarization-based compaction, CWL avoids four well-known limitations: unpredictable lossiness, destruction of causal structure, blocking model cost, and compression-induced hallucination. Compared to recency truncation, CWL is semantically aware: it drops the oldest-and-most-recoverable content according to the dependency graph rather than oldest-in-time regardless of relevance. We describe the annotation protocol, the episode graph, the eviction policy, and the token-accounting loop, and evaluate CWL on long-horizon agentic benchmarks: a single agent session completing 89 sequential tasks across 80 million tokens with no measurable degradation in task accuracy relative to per-task isolated sessions", "authors": ["Andrew Semenov", "Svyatoslav Dorofeev"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-05-01", "url": "https://arxiv.org/abs/2606.11213", "pdf_url": "https://arxiv.org/pdf/2606.11213v1", "arxiv_id": "2606.11213", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.35}
{"id": "f752c981e458685d634226e67b38880aa4cc6c9e2d40a0e471bd1808a658c423", "sources": ["arxiv", "semantic_scholar"], "title": "AutoSP: Unlocking Long-Context LLM Training Via Compiler-Based Sequence Parallelism", "abstract": "Large-language-models (LLMs) demonstrate enormous utility in long-context tasks which require processing prompts that consist of tens to hundreds of thousands of tokens. However, existing LLM training libraries do not provide easy to use abstractions to optimize for long-context training, instead focusing on optimizations for models with large parameter counts through ZeRO-3/FSDP, Tensor and Pipeline parallelism. This forces users to rewrite LLM training libraries to incorporate compositions of various complex long-context optimizations, such as sequence-parallelism, to training pipelines; a process that requires in-depth expertise, reducing developer productivity. To tackle these challenges, we introduce AutoSP: the first automated solution to automatically optimize LLM training for longer-contexts. AutoSP compiles models and applies a targeted set of optimizations: automated sequence parallelism, and long-context aware activation-checkpointing, to drastically enhance LLM trainability at negligible cost to throughput. Our evaluation demonstrates AutoSP's capability on both NVIDIA and AMD hardware, increasing training contexts by upto 2.7$\\times$ and 2.5$\\times$ respectively over competitive hand-written baseline at negligible cost to runtime performance.", "authors": ["Ahan Gupta", "Zhihao Wang", "Neel Dani", "Masahiro Tanaka", "Olatunji Ruwase", "Minjia Zhang"], "categories": ["cs.LG", "cs.DC", "cs.PF"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-29", "url": "https://arxiv.org/abs/2604.27089", "pdf_url": "https://arxiv.org/pdf/2604.27089v1", "arxiv_id": "2604.27089", "doi": "10.48550/arXiv.2604.27089", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.55}
{"id": "49f072ed4ad7c6e2f1cff23269ee166e31ca501cdc85b10602bbc87a1dc4d823", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Context Aware Upcycling: A New Frontier for Hybrid LLM Scaling", "abstract": "Hybrid sequence models that combine efficient Transformer components with linear sequence modeling blocks are a promising alternative to pure Transformers, but most are still pretrained from scratch and therefore fail to reuse existing Transformer checkpoints. We study upcycling as a practical path to convert pretrained Transformer LLMs into hybrid architectures while preserving short-context quality and improving long-context capability. We call our solution \\emph{HyLo} (HYbrid LOng-context): a long-context upcycling recipe that combines architectural adaptation with efficient Transformer blocks, Multi-Head Latent Attention (MLA), and linear blocks (Mamba2 or Gated DeltaNet), together with staged long-context training and teacher-guided distillation for stable optimization. HyLo extends usable context length by up to $32\\times$ through efficient post-training and reduces KV-cache memory by more than $90\\%$, enabling up to 2M-token prefill and decoding in our \\texttt{vLLM} inference stack, while comparable Llama baselines run out of memory beyond 64K context. Across 1B- and 3B-scale settings (Llama- and Qwen-based variants), HyLo delivers consistently strong short- and long-context performance and significantly outperforms state-of-the-art upcycled hybrid baselines on long-context evaluations such as RULER. Notably, at similar scale, HyLo-Qwen-1.7B trained on only 10B tokens significantly outperforms JetNemotron (trained on 400B tokens) on GSM8K, Lm-Harness common sense reasoning and RULER-64K.", "authors": ["Parsa Ashrafi Fashi", "Utkarsh Saxena", "Mehdi Rezagholizadeh", "Aref Jafari", "Akash Haridas", "Mingyu Yang", "Vansh Bhatia", "Guihong Li", "Vikram Appia", "Emad Barsoum"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-27", "url": "https://arxiv.org/abs/2604.24715", "pdf_url": "https://arxiv.org/pdf/2604.24715v1", "arxiv_id": "2604.24715", "doi": "10.48550/arXiv.2604.24715", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.55}
{"id": "c5de05ed9f2f57d90fe0dfb60123c3371fc59b76550df56b647a09effe58e285", "sources": ["arxiv", "semantic_scholar"], "title": "When Context Sticks: Studying Interference in In-Context Learning", "abstract": "This paper investigates context stickiness in in-context learning (ICL), a phenomenon where earlier examples in a prompt interfere with a transformer's ability to adapt to later tasks. Using synthetic regression tasks over linear and quadratic functions, we examine how models trained under sequential, mixed, and random curricula handle abrupt task switches during inference. By sweeping over structured combinations of misleading linear examples followed by recovery quadratic examples, we quantify how prior context biases prediction error and how quickly models realign. Our results show strong evidence of persistent interference: more preceding linear examples reliably degrade quadratic predictions, while additional quadratic examples reduce error but with diminishing returns. We further find that training curricula significantly modulate resilience, with sequential training on the target function class yielding the fastest recovery, and surprisingly, random training producing the least robust behavior.", "authors": ["Hanna Rød", "Dagny Streit", "Nils Valseth Selte", "Justin Li"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-25", "url": "https://arxiv.org/abs/2604.23371", "pdf_url": "https://arxiv.org/pdf/2604.23371v1", "arxiv_id": "2604.23371", "doi": "10.48550/arXiv.2604.23371", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/nilsvselte/icl-context-stickiness", "venue": "arXiv.org", "quality_score": 0.85}
{"id": "c8463eb25f28d9f99c918065202aa32a3432ce44bb8adc1602cd7e304fb931b8", "sources": ["arxiv", "semantic_scholar"], "title": "Shuffle the Context: RoPE-Perturbed Self-Distillation for Long-Context Adaptation", "abstract": "Large language models (LLMs) increasingly operate in settings that require reliable long-context understanding, such as retrieval-augmented generation and multi-document reasoning. A common strategy is to fine-tune pretrained short-context models at the target sequence length. However, we find that standard long-context adaptation can remain brittle: model accuracy depends strongly on the absolute placement of relevant evidence, exhibiting high positional variance even when controlling for task format and difficulty. We propose RoPE-Perturbed Self-Distillation, a training regularizer that improves positional robustness. The core idea is to form alternative \"views\" of the same training sequence by perturbing its RoPE indices -- effectively moving parts of the context to different positions -- and to train the model to produce consistent predictions across views via self-distillation. This encourages reliance on semantic signals instead of brittle position dependencies. Experiments on long-context adaptation of Llama-3-8B and Qwen-3-4B demonstrate consistent gains on long-context benchmarks, including up to 12.04% improvement on RULER-64K for Llama-3-8B and 2.71% on RULER-256K for Qwen-3-4B after SFT, alongside improved length extrapolation beyond the training context window.", "authors": ["Zichong Li", "Chen Liang", "Liliang Ren", "Tuo Zhao", "Yelong Shen", "Weizhu Chen"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-15", "url": "https://arxiv.org/abs/2604.14339", "pdf_url": "https://arxiv.org/pdf/2604.14339v1", "arxiv_id": "2604.14339", "doi": "10.48550/arXiv.2604.14339", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5443}
{"id": "f9b6961c4ef97aa9000b95c9ad12a43d52b53dc54244eb9ea3ba53c3d6abc383", "sources": ["arxiv", "semantic_scholar"], "title": "Context Kubernetes: Declarative Orchestration of Enterprise Knowledge for Agentic AI Systems", "abstract": "We introduce Context Kubernetes, an architecture for orchestrating enterprise knowledge in agentic AI systems, with a prototype implementation and eight experiments. The core observation is that delivering the right knowledge, to the right agent, with the right permissions, at the right freshness -- across an entire organization -- is structurally analogous to the container orchestration problem Kubernetes solved a decade ago. We formalize six core abstractions, a YAML-based declarative manifest for knowledge-architecture-as-code, a reconciliation loop, and a three-tier agent permission model where agent authority is always a strict subset of human authority. On synthetic seed data, we compare four governance baselines of increasing strength: ungoverned RAG, ACL-filtered retrieval, RBAC-aware routing, and the full architecture. Each layer contributes a different capability: ACL filtering eliminates cross-domain leaks, intent routing reduces noise by 19 percentage points, and only the three-tier model blocks all five tested attack scenarios -- the one attack RBAC misses is an agent sending confidential pricing via email, which RBAC cannot distinguish from ordinary email. TLA+ model-checking verifies safety properties across 4.6 million reachable states with zero violations. A survey of four major platforms (Microsoft, Salesforce, AWS, Google) documents that none architecturally isolates agent approval channels. We identify four properties that make context orchestration harder than container orchestration, and argue these make the solution more valuable.", "authors": ["Charafeddine Mouzouni"], "categories": ["cs.AI", "cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-13", "url": "https://arxiv.org/abs/2604.11623", "pdf_url": "https://arxiv.org/pdf/2604.11623v3", "arxiv_id": "2604.11623", "doi": "10.48550/arXiv.2604.11623", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Cohorte-ai/context-kubernetes", "venue": "arXiv.org", "quality_score": 0.8376}
{"id": "f91f236995a09546fde12f5f6ee1ba0272e69b9aeff68171170891668045e568", "sources": ["arxiv", "semantic_scholar"], "title": "A Decomposition Perspective to Long-context Reasoning for LLMs", "abstract": "Long-context reasoning is essential for complex real-world applications, yet remains a significant challenge for Large Language Models (LLMs). Despite the rapid evolution in long-context reasoning, current research often overlooks the internal complexity of the long-context reasoning task itself. In this paper, we move beyond this holistic view and decompose long-context reasoning into a set of fundamental atomic skills, and we then automatically synthesize a suite of pseudo datasets, each explicitly targeting a specific atomic skill. Our empirical analysis confirms that proficiency in these atomic skills is strongly correlated with general long-text reasoning performance. Building on this insight, we employ reinforcement learning on these pseudo datasets to sharpen the model's atomic skills, in the hope of boosting its general long-context reasoning ability. Extensive experiments across multiple benchmarks demonstrate the effectiveness of our approach: it outperforms a strong baseline by an average margin of 7.7\\% (improving from 46.3\\% to 54.0\\%) across Loogle, Loong, LongBench-v2, BrowscompLong, Ruler-qa2, and MRCR.", "authors": ["Yanling Xiao", "Huaibing Xie", "Guoliang Zhao", "Shihan Dou", "Shaolei Wang", "Yiting Liu", "Nantao Zheng", "Cheng Zhang", "Pluto Zhou", "Zhisong Zhang", "Lemao Liu"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-09", "url": "https://arxiv.org/abs/2604.07981", "pdf_url": "https://arxiv.org/pdf/2604.07981v1", "arxiv_id": "2604.07981", "doi": "10.48550/arXiv.2604.07981", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5374}
{"id": "31071bafec02176b29d2a241d2521acfcf80f81fe9206c562857fbae5a07af21", "sources": ["arxiv", "semantic_scholar"], "title": "Video-guided Machine Translation with Global Video Context", "abstract": "Video-guided Multimodal Translation (VMT) has advanced significantly in recent years. However, most existing methods rely on locally aligned video segments paired one-to-one with subtitles, limiting their ability to capture global narrative context across multiple segments in long videos. To overcome this limitation, we propose a globally video-guided multimodal translation framework that leverages a pretrained semantic encoder and vector database-based subtitle retrieval to construct a context set of video segments closely related to the target subtitle semantics. An attention mechanism is employed to focus on highly relevant visual content, while preserving the remaining video features to retain broader contextual information. Furthermore, we design a region-aware cross-modal attention mechanism to enhance semantic alignment during translation. Experiments on a large-scale documentary translation dataset demonstrate that our method significantly outperforms baseline models, highlighting its effectiveness in long-video scenarios.", "authors": ["Jian Chen", "JinZe Lv", "Zi Long", "XiangHua Fu"], "categories": ["cs.CV", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-08", "url": "https://arxiv.org/abs/2604.06789", "pdf_url": "https://arxiv.org/pdf/2604.06789v1", "arxiv_id": "2604.06789", "doi": "10.48550/arXiv.2604.06789", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5363}
{"id": "45da9b22671cb639e1c1a163e3ce08ec92c3948c9c9a89a668bf4ed448dc80ab", "sources": ["arxiv", "semantic_scholar"], "title": "Short Data, Long Context: Distilling Positional Knowledge in Transformers", "abstract": "Extending the context window of language models typically requires expensive long-context pre-training, posing significant challenges for both training efficiency and data collection. In this paper, we present evidence that long-context retrieval capabilities can be transferred to student models through logit-based knowledge distillation, even when training exclusively on packed short-context samples within a long-context window. We provide comprehensive insights through the lens of Rotary Position Embedding (RoPE) and establish three key findings. First, consistent with prior work, we show that phase-wise RoPE scaling, which maximizes rotational spectrum utilization at each training stage, also achieves the best long-context performance in knowledge distillation setups. Second, we demonstrate that logit-based knowledge distillation can directly enable positional information transfer. Using an experimental setup with packed repeated token sequences, we trace the propagation of positional perturbations from query and key vectors through successive transformer layers to output logits, revealing that positional information systematically influences the teacher's output distribution and, in turn, the distillation signal received by the student model. Third, our analysis uncovers structured update patterns in the query state during long-context extension, with distinct parameter spans exhibiting strong sensitivity to long-context training.", "authors": ["Patrick Huber", "Ernie Chang", "Chinnadhurai Sankar", "Rylan Conway", "Igor Fedorov", "Md Rifat Arefin", "Adithya Sagar"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-07", "url": "https://arxiv.org/abs/2604.06070", "pdf_url": "https://arxiv.org/pdf/2604.06070v1", "arxiv_id": "2604.06070", "doi": "10.48550/arXiv.2604.06070", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5351}
{"id": "4666ed60456a709d99e6cad627d1cf252e251716502708f0d64c9ddd2dbcdc30", "sources": ["arxiv", "semantic_scholar"], "title": "$π^2$: Structure-Originated Reasoning Data Improves Long-Context Reasoning Ability of Large Language Models", "abstract": "We study a pipeline that curates reasoning data from initial structured data for improving long-context reasoning in large language models (LLMs). Our approach, $π^2$, constructs high-quality reasoning data through rigorous QA curation: 1) extracting and expanding tables from Wikipedia, 2) from the collected tables and relevant context, generating realistic and multi-hop analytical reasoning questions whose answers are automatically determined and verified through dual-path code execution, and 3) back-translating step-by-step structured reasoning traces as solutions of QA pairs given realistic web-search context. Supervised fine-tuning with \\textsc{\\small{gpt-oss-20b}} and \\textsc{\\small{Qwen3-4B-Instruct-2507}} on $π^2$ yields consistent improvements across four long-context reasoning benchmarks and our alike $π^2$-Bench, with average absolute accuracy gains of +4.3% and +2.7% respectively. Notably, our dataset facilitates self-distillation, where \\textsc{\\small{gpt-oss-20b}} even improves its average performance by +4.4% with its own reasoning traces, demonstrating $π^2$'s usefulness. Our code, data, and models are open-source at https://github.com/vt-pi-squared/pi-squared.", "authors": ["Quyet V. Do", "Thinh Pham", "Nguyen Nguyen", "Sha Li", "Pratibha Zunjare", "Tu Vu"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-06", "url": "https://arxiv.org/abs/2604.05114", "pdf_url": "https://arxiv.org/pdf/2604.05114v1", "arxiv_id": "2604.05114", "doi": "10.48550/arXiv.2604.05114", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/vt-pi-squared/pi-squared", "venue": "arXiv.org", "quality_score": 0.8252}
{"id": "9962d7de9b2e43fa10f41b4d1fe11e55d64f1408698e58e54c18967d8871e3f2", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Context Reasoning Through Proxy-Based Chain-of-Thought Tuning", "abstract": "Recent large language models support inputs of up to 10 million tokens, yet they perform poorly on long-context tasks that require complex reasoning. Such tasks can be solved using only a subset of the input -- a proxy context -- rather than the full sequence. Despite sharing the same underlying reasoning process, models exhibit a significant performance disparity between proxy and full contexts. To improve long-context reasoning, we propose ProxyCoT, a novel training framework that transfers reasoning capabilities from short proxy contexts to full long contexts. Specifically, we first obtain high-quality chain-of-thought reasoning traces on proxy contexts through reinforcement learning or distillation from a larger teacher model, and then ground the generated traces in full long contexts with supervised fine-tuning. Experiments across different datasets demonstrate that ProxyCoT consistently outperforms strong baselines with reduced computational overhead. Furthermore, models trained with ProxyCoT generalize their long-context reasoning capabilities to out-of-domain tasks.", "authors": ["Miao Li", "Irina Saparina", "Alexander Gurung", "Mirella Lapata"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-04-06", "url": "https://arxiv.org/abs/2605.20201", "pdf_url": "https://arxiv.org/pdf/2605.20201v2", "arxiv_id": "2605.20201", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3398}
{"id": "8842c6f5c655103f16dda518954c1860b810b8b58affdbd3d2a7e35e37ed11d6", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Reach Robotic Manipulation for Assembly and Outfitting of Lunar Structures", "abstract": "Future infrastructure construction on the lunar surface will require semi- or fully-autonomous operation from robots deployed at the build site. In particular, tasks such as electrical outfitting necessitate transport, routing, and fine manipulation of cables across large structures. To address this need, we present a compact and long-reach manipulator incorporating a deployable composite boom, capable of performing manipulation tasks across large structures and workspaces. We characterize the deflection, vibration, and blossoming characteristics inherent to the deployable structure, and present a manipulation control strategy to mitigate these effects. Experiments indicate an average endpoint accuracy error of less than 15 mm for boom lengths up to 1.8 m. We demonstrate the approach with a cable routing task to illustrate the potential for lunar outfitting applications that benefit from long reach.", "authors": ["Stanley Wang", "Venny Kojouharov", "Long Yin Chung", "Daniel Morton", "Mark Cutkosky"], "categories": ["cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-31", "url": "https://arxiv.org/abs/2603.29226", "pdf_url": "https://arxiv.org/pdf/2603.29226v1", "arxiv_id": "2603.29226", "doi": "10.1109/iSpaRo66239.2025.11437024", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3354}
{"id": "31616d6178f023c384fd8fbbd40e3c871003dcee94fd27a7f2ac55f8dc79026d", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Reach Robotic Cleaning for Lunar Solar Arrays", "abstract": "Commercial lunar activity is accelerating the need for reliable surface infrastructure and routine operations to keep it functioning. Maintenance tasks such as inspection, cleaning, dust mitigation, and minor repair are essential to preserve performance and extend system life. A specific application is the cleaning of lunar solar arrays. Solar arrays are expected to provide substantial fraction of lunar surface power and operate for months to years, supplying continuous energy to landers, habitats, and surface assets, making sustained output mission-critical. However, over time lunar dust accumulates on these large solar arrays, which can rapidly degrade panel output and reduce mission lifetime. We propose a small mobile robot equipped with a long-reach, lightweight deployable boom and interchangeable cleaning tool to perform gentle cleaning over meter-scale workspaces with minimal human involvement. Building on prior vision-guided long-reach manipulation, we add a compliant wrist with distal force sensing and a velocity-based admittance controller to regulate stable contact during surface cleaning. In preliminary benchtop experiments on a planar surface, the system maintained approximately 2 N normal force while executing a simple cleaning motion over boom lengths from 0.3 m to 1.0 m, with RMS force error of approximately 0.2 N after initial contact. These early results suggest that deployable long-reach manipulators are a promising architecture for robotic maintenance of lunar infrastructure such as solar arrays, radiators, and optical surfaces.", "authors": ["Stanley Wang", "Velin Kojouharov", "Long Yin Chung", "Daniel Morton", "Mark Cutkosky"], "categories": ["cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-31", "url": "https://arxiv.org/abs/2603.29240", "pdf_url": "https://arxiv.org/pdf/2603.29240v1", "arxiv_id": "2603.29240", "doi": "10.48550/arXiv.2603.29240", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5271}
{"id": "8997a92a1e89e115cc6b4941c5f8a918fad3ce3ffe0381820e4f4d7027fd7fdc", "sources": ["arxiv", "semantic_scholar"], "title": "Developing Adaptive Context Compression Techniques for Large Language Models (LLMs) in Long-Running Interactions", "abstract": "Large Language Models (LLMs) often experience performance degradation during long-running interactions due to increasing context length, memory saturation, and computational overhead. This paper presents an adaptive context compression framework that integrates importance-aware memory selection, coherence-sensitive filtering, and dynamic budget allocation to retain essential conversational information while controlling context growth. The approach is evaluated on LOCOMO, LOCCO, and LongBench benchmarks to assess answer quality, retrieval accuracy, coherence preservation, and efficiency. Experimental results demonstrate that the proposed method achieves consistent improvements in conversational stability and retrieval performance while reducing token usage and inference latency compared with existing memory and compression-based approaches. These findings indicate that adaptive context compression provides an effective balance between long-term memory preservation and computational efficiency in persistent LLM interactions", "authors": ["Payal Fofadiya", "Sunil Tiwari"], "categories": ["cs.CV", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-31", "url": "https://arxiv.org/abs/2603.29193", "pdf_url": "https://arxiv.org/pdf/2603.29193v1", "arxiv_id": "2603.29193", "doi": "10.48550/arXiv.2603.29193", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5271}
{"id": "dd6c0db625bbf56eff3a3d846ec56e4d3177ecd019b45e063828c1f3d6b14568", "sources": ["arxiv", "semantic_scholar"], "title": "MemoryCD: Benchmarking Long-Context User Memory of LLM Agents for Lifelong Cross-Domain Personalization", "abstract": "Recent advancements in Large Language Models (LLMs) have expanded context windows to million-token scales, yet benchmarks for evaluating memory remain limited to short-session synthetic dialogues. We introduce \\textsc{MemoryCD}, the first large-scale, user-centric, cross-domain memory benchmark derived from lifelong real-world behaviors in the Amazon Review dataset. Unlike existing memory datasets that rely on scripted personas to generate synthetic user data, \\textsc{MemoryCD} tracks authentic user interactions across years and multiple domains. We construct a multi-faceted long-context memory evaluation pipeline of 14 state-of-the-art LLM base models with 6 memory method baselines on 4 distinct personalization tasks over 12 diverse domains to evaluate an agent's ability to simulate real user behaviors in both single and cross-domain settings. Our analysis reveals that existing memory methods are far from user satisfaction in various domains, offering the first testbed for cross-domain life-long personalization evaluation.", "authors": ["Weizhi Zhang", "Xiaokai Wei", "Wei-Chieh Huang", "Zheng Hui", "Chen Wang", "Michelle Gong", "Philip S. Yu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-26", "url": "https://arxiv.org/abs/2603.25973", "pdf_url": "https://arxiv.org/pdf/2603.25973v1", "arxiv_id": "2603.25973", "doi": "10.48550/arXiv.2603.25973", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5214}
{"id": "cde7954fa97265f240e722b727f09ea65fcb6a74fe626a36b5ee89ffa55ef0c7", "sources": ["arxiv", "semantic_scholar"], "title": "Reasoner-Executor-Synthesizer: Scalable Agentic Architecture with Static O(1) Context Window", "abstract": "Large Language Models (LLMs) deployed as autonomous agents commonly use Retrieval-Augmented Generation (RAG), feeding retrieved documents into the context window, which creates two problems: the risk of hallucination grows with context length, and token cost scales linearly with dataset size. We propose the Reasoner-Executor-Synthesizer (RES) architecture, a three-layer design that strictly separates intent parsing (Reasoner), deterministic data retrieval and aggregation (Executor), and narrative generation (Synthesizer). The Executor uses zero LLM tokens and passes only fixed-size statistical summaries to the Synthesizer. We formally prove that RES achieves O(1) token complexity with respect to dataset size, and validate this on ScholarSearch, a scholarly research assistant backed by the Crossref API (130M+ articles). Across 100 benchmark runs, RES achieves a mean token cost of 1,574 tokens regardless of whether the dataset contains 42,000 or 16.3 million articles. The architecture eliminates data hallucination by construction: the LLM never sees raw records. KEYWORDS LLM agents; agentic architecture; hallucination elimination; token optimization; context window; retrieval-augmented generation; deterministic execution; scholarly metadata; Crossref API; O(1) complexity.", "authors": ["Ivan Dobrovolskyi"], "categories": ["cs.IR", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-23", "url": "https://arxiv.org/abs/2603.22367", "pdf_url": "https://arxiv.org/pdf/2603.22367v1", "arxiv_id": "2603.22367", "doi": "10.48550/arXiv.2603.22367", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5179}
{"id": "99df631ca195ec785bd83ac377fb879f8a08eb0a8e4615a190ef5ba8244525bd", "sources": ["arxiv", "semantic_scholar"], "title": "Conversation Tree Architecture: A Structured Framework for Context-Aware Multi-Branch LLM Conversations", "abstract": "Large language models (LLMs) are increasingly deployed for extended, multi-topic conversations, yet the flat, append-only structure of current conversation interfaces introduces a fundamental limitation: all context accumulates in a single unbounded window, causing topically distinct threads to bleed into one another and progressively degrade response quality. We term this failure mode logical context poisoning. In this paper, we introduce the Conversation Tree Architecture (CTA), a hierarchical framework that organizes LLM conversations as trees of discrete, context-isolated nodes. Each node maintains its own local context window; structured mechanisms govern how context flows between parent and child nodes, downstream on branch creation and upstream on branch deletion. We additionally introduce volatile nodes, transient branches whose local context must be selectively merged upward or permanently discarded before purging. We formalize the architecture's primitives, characterize the open design problems in context flow, relate our framework to prior work in LLM memory management, and describe a working prototype implementation. The CTA provides a principled foundation for structured conversational context management and extends naturally to multi-agent settings.", "authors": ["Pranav Hemanth", "Sampriti Saha"], "categories": ["cs.CL", "cs.AI", "cs.HC"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-22", "url": "https://arxiv.org/abs/2603.21278", "pdf_url": "https://arxiv.org/pdf/2603.21278v1", "arxiv_id": "2603.21278", "doi": "10.48550/arXiv.2603.21278", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5168}
{"id": "6718abf84cfe2befedec977483c4fd97115bef34ca52f24f73ad93a65967527d", "sources": ["arxiv", "semantic_scholar"], "title": "MKA: Memory-Keyed Attention for Efficient Long-Context Reasoning", "abstract": "As long-context language modeling becomes increasingly important, the cost of maintaining and attending to large Key/Value (KV) caches grows rapidly, becoming a major bottleneck in both training and inference. While prior works such as Multi-Query Attention (MQA) and Multi-Latent Attention (MLA) reduce memory by sharing or compressing KV features, they often trade off representation quality or incur runtime overhead. We propose Memory-Keyed Attention (MKA), a hierarchical attention mechanism that integrates multi-level KV caches (local, session, and long-term) and learns to route attention across them dynamically. We further introduce Route-Fused MKA (FastMKA), a broadcast-routed variant that fuses memory sources before attention computation for improved efficiency. Experiments on different sequence lengths show that FastMKA achieves a favorable accuracy-efficiency trade-off: comparable perplexity to MLA while achieving up to 5x faster training throughput and 1.8x lower evaluation latency. These results highlight MKA as a practical and extensible framework for efficient long-context attention.", "authors": ["Dong Liu", "Yanxuan Yu", "Ben Lengerich", "Ying Nian Wu"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-21", "url": "https://arxiv.org/abs/2603.20586", "pdf_url": "https://arxiv.org/pdf/2603.20586v2", "arxiv_id": "2603.20586", "doi": "10.48550/arXiv.2603.20586", "citation_count": 9, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5156}
{"id": "511989ccf4603e4547100d47c61e64e9297ec8a841b3226201e351751239b4fe", "sources": ["arxiv", "semantic_scholar"], "title": "Coding Agents are Effective Long-Context Processors", "abstract": "Large Language Models (LLMs) have demonstrated remarkable progress in scaling to access massive contexts. However, the access is via the latent and uninterpretable attention mechanisms, and LLMs fail to effective process long context, exhibiting significant performance degradation as context length increases. In this work, we study whether long-context processing can be externalized from latent attention into explicit, executable interactions, by allowing coding agents to organize text in file systems and manipulate it using its native tools. We evaluate off-the-shelf frontier coding agents as the general interface for tasks that require processing long contexts, including long-context reasoning, retrieval-augmented generation, and open-domain question answering with large-scale corpus contains up to three trillion tokens. Across multiple benchmarks, these agents outperform published state-of-the-art by 17.3% on average. We attribute this efficacy to two key factors: native tool proficiency, which enables agents to leverage executable code and terminal commands rather than passive semantic queries, and file system familiarity, which allows them to navigate massive text corpora as directory structures. These findings suggest that delegating long-context processing to coding agents offers an effective alternative to semantic search or context window scaling, opening new directions for long-context processing in LLMs.", "authors": ["Weili Cao", "Xunjian Yin", "Bhuwan Dhingra", "Shuyan Zhou"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-20", "url": "https://arxiv.org/abs/2603.20432", "pdf_url": "https://arxiv.org/pdf/2603.20432v1", "arxiv_id": "2603.20432", "doi": "10.48550/arXiv.2603.20432", "citation_count": 5, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5145}
{"id": "be1dd14fa6bcf09ff067044566d4b9b42556b3b8c0ab9121fa4db1d102889246", "sources": ["arxiv", "semantic_scholar"], "title": "The $\\mathbf{Y}$-Combinator for LLMs: Solving Long-Context Rot with $λ$-Calculus", "abstract": "LLMs are increasingly used as general-purpose reasoners, but long inputs remain bottlenecked by a fixed context window. Recursive Language Models (RLMs) address this by externalising the prompt and recursively solving subproblems. Yet existing RLMs depend on an open-ended read-eval-print loop (REPL) in which the model generates arbitrary control code, making execution difficult to verify, predict, and analyse. We introduce $λ$-RLM, a framework for long-context reasoning that replaces free-form recursive code generation with a typed functional runtime grounded in $λ$-calculus. It executes a compact library of pre-verified combinators and uses neural inference only on bounded leaf subproblems, turning recursive reasoning into a structured functional program with explicit control flow. We show that $λ$-RLM admits formal guarantees absent from standard RLMs, including termination, closed-form cost bounds, controlled accuracy scaling with recursion depth, and an optimal partition rule under a simple cost model. Empirically, across four long-context reasoning tasks and nine base models, $λ$-RLM outperforms standard RLM in 29 of 36 model-task comparisons, improves average accuracy by up to +21.9 points across model tiers, and reduces latency by up to 4.1x. These results show that typed symbolic control yields a more reliable and efficient foundation for long-context reasoning than open-ended recursive code generation. The complete implementation of $λ$-RLM, is open-sourced for the community at: https://github.com/lambda-calculus-LLM/lambda-RLM.", "authors": ["Amartya Roy", "Rasul Tutunov", "Xiaotong Ji", "Matthieu Zimmer", "Haitham Bou-Ammar"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-20", "url": "https://arxiv.org/abs/2603.20105", "pdf_url": "https://arxiv.org/pdf/2603.20105v1", "arxiv_id": "2603.20105", "doi": "10.48550/arXiv.2603.20105", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/lambda-calculus-LLM/lambda-RLM", "venue": "arXiv.org", "quality_score": 0.7951}
{"id": "e1314b19696e108579b7e427d2e0c936f4958d017fa36c278959c440bc036d2e", "sources": ["arxiv", "semantic_scholar"], "title": "UT-ACA: Uncertainty-Triggered Adaptive Context Allocation for Long-Context Inference", "abstract": "Long-context inference remains challenging for large language models due to attention dilution and out-of-distribution degradation. Context selection mitigates this limitation by attending to a subset of key-value cache entries, yet most methods allocate a fixed context budget throughout decoding despite highly non-uniform token-level contextual demands. To address this issue, we propose Uncertainty-Triggered Adaptive Context Allocation (UT-ACA), an inference-time framework that dynamically adjusts the context window based on token-wise uncertainty. UT-ACA learns an uncertainty detector that combines semantic embeddings with logit-based confidence while accounting for uncertainty accumulation across decoding steps. When insufficient evidence is indicated, UT-ACA selectively rolls back, expands the context window, and regenerates the token with additional support. Experiments show that UT-ACA substantially reduces average context usage while preserving generation quality in long-context settings.", "authors": ["Lang Zhou", "Shuxuan Li", "Zhuohao Li", "Shi Liu", "Zhilin Zhao", "Wei-Shi Zheng"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-19", "url": "https://arxiv.org/abs/2603.18446", "pdf_url": "https://arxiv.org/pdf/2603.18446v1", "arxiv_id": "2603.18446", "doi": "10.48550/arXiv.2603.18446", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5133}
{"id": "0009cffae66956b8054a21dea2bb6182cf480edb70affb48c3238ece66368293", "sources": ["arxiv", "semantic_scholar"], "title": "Difference-Based High-Dimensional Long-Run Covariance Matrix Estimation for Mean-shift Time Series", "abstract": "We consider estimation of high-dimensional long-run covariance matrices for time series with nonconstant means, a setting in which conventional estimators can be severely biased. To address this difficulty, we propose a difference-based initial estimator that is robust to a broad class of mean variations, and combine it with hard thresholding, soft thresholding, and tapering to obtain sparse long-run covariance estimators for high-dimensional data. We derive convergence rates for the resulting estimators under general temporal dependence and time-varying mean structures, showing explicitly how the rates depend on covariance sparsity, mean variation, dimension, and sample size. Numerical experiments show that the proposed methods perform favorably in high dimensions, especially when the mean evolves over time.", "authors": ["Yanhong Liu", "Fengyi Song", "Long Feng"], "categories": ["stat.ME"], "fields_of_study": ["Mathematics"], "published_date": "2026-03-18", "url": "https://arxiv.org/abs/2603.17226", "pdf_url": "https://arxiv.org/pdf/2603.17226v1", "arxiv_id": "2603.17226", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3259}
{"id": "6a14dc39117b75261c88b20424b37b4fb359ad2195e959bc43fe2fae9c95be97", "sources": ["arxiv", "semantic_scholar"], "title": "Causal Cellular Context Transfer Learning (C3TL): An Efficient Architecture for Prediction of Unseen Perturbation Effects", "abstract": "Predicting the effects of chemical and genetic perturbations on quantitative cell states is a central challenge in computational biology, molecular medicine and drug discovery. Recent work has leveraged large-scale single-cell data and massive foundation models to address this task. However, such computational resources and extensive datasets are not always accessible in academic or clinical settings, hence limiting utility. Here we propose a lightweight framework for perturbation effect prediction that exploits the structured nature of biological interventions and specific inductive biases/invariances. Our approach leverages available information concerning perturbation effects to allow generalization to novel contexts and requires only widely-available bulk molecular data. Extensive testing, comparing predictions of context-specific perturbation effects against real, large-scale interventional experiments, demonstrates accurate prediction in new contexts. The proposed approach is competitive with SOTA foundation models but requires simpler data, much smaller model sizes and less time. Focusing on robust bulk signals and efficient architectures, we show that accurate prediction of perturbation effects is possible without proprietary hardware or very large models, hence opening up ways to leverage causal learning approaches in biomedicine generally.", "authors": ["Michael Scholkemper", "Sach Mukherjee"], "categories": ["cs.LG", "q-bio.QM"], "fields_of_study": ["Computer Science", "Biology"], "published_date": "2026-03-13", "url": "https://arxiv.org/abs/2603.13051", "pdf_url": "https://arxiv.org/pdf/2603.13051v1", "arxiv_id": "2603.13051", "doi": "10.48550/arXiv.2603.13051", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5065}
{"id": "f5a947fd476a2fd8378bb5d53f7226c4f43226d92a0a76c9279ca73faae6f313", "sources": ["arxiv", "semantic_scholar"], "title": "StatePlane: A Cognitive State Plane for Long-Horizon AI Systems Under Bounded Context", "abstract": "Large language models (LLMs) and small language models (SLMs) operate under strict context window and key-value (KV) cache constraints, fundamentally limiting their ability to reason coherently over long interaction horizons. Existing approaches -- extended context windows, retrieval-augmented generation, summarization, or static documentation -- treat memory as static storage and fail to preserve decision-relevant state under long-running, multi-session tasks. We introduce StatePlane, a model-agnostic cognitive state plane that governs the formation, evolution, retrieval, and decay of episodic, semantic, and procedural state for AI systems operating under bounded context. Grounded in cognitive psychology and systems design, StatePlane formalizes episodic segmentation, selective encoding via information-theoretic constraints, goal-conditioned retrieval with intent routing, reconstructive state synthesis, and adaptive forgetting. We present a formal state model, KV-aware algorithms, security and governance mechanisms including write-path anti-poisoning, enterprise integration pathways, and an evaluation framework with six domain-specific benchmarks. StatePlane demonstrates that long-horizon intelligence can be achieved without expanding context windows or retraining models.", "authors": ["Sasank Annapureddy", "John Mulcahy", "Anjaneya Prasad Thamatani"], "categories": ["cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-13", "url": "https://arxiv.org/abs/2603.13644", "pdf_url": "https://arxiv.org/pdf/2603.13644v1", "arxiv_id": "2603.13644", "doi": "10.48550/arXiv.2603.13644", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5065}
{"id": "16f31e7cd9b7bc7d45d796131b859aed010b69c5e39866d31637c43c364dc2be", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Context Encoder Models for Polish Language Understanding", "abstract": "While decoder-only Large Language Models (LLMs) have recently dominated the NLP landscape, encoder-only architectures remain a cost-effective and parameter-efficient standard for discriminative tasks. However, classic encoders like BERT are limited by a short context window, which is insufficient for processing long documents. In this paper, we address this limitation for the Polish by introducing a high-quality Polish model capable of processing sequences of up to 8192 tokens. The model was developed by employing a two-stage training procedure that involves positional embedding adaptation and full parameter continuous pre-training. Furthermore, we propose compressed model variants trained via knowledge distillation. The models were evaluated on 25 tasks, including the KLEJ benchmark, a newly introduced financial task suite (FinBench), and other classification and regression tasks, specifically those requiring long-document understanding. The results demonstrate that our model achieves the best average performance among Polish and multilingual models, significantly outperforming competitive solutions in long-context tasks while maintaining comparable quality on short texts.", "authors": ["Sławomir Dadas", "Rafał Poświata", "Marek Kozłowski", "Małgorzata Grębowiec", "Michał Perełkiewicz", "Paweł Klimiuk", "Przemysław Boruta"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-12", "url": "https://arxiv.org/abs/2603.12191", "pdf_url": "https://arxiv.org/pdf/2603.12191v1", "arxiv_id": "2603.12191", "doi": "10.48550/arXiv.2603.12191", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5053}
{"id": "d5ce7518569ed8538541977e2ee11fe788fdeecdd4e4a5e1c98712d6de8ad1e8", "sources": ["arxiv", "semantic_scholar"], "title": "The Missing Memory Hierarchy: Demand Paging for LLM Context Windows", "abstract": "The context window of a large language model is not memory. It is L1 cache: a small, fast, expensive resource that the field treats as the entire memory system. There is no L2, no virtual memory, no paging. Every tool definition, every system prompt, and every stale tool result occupies context for the lifetime of the session. The result is measurable: across 857 production sessions and 4.45 million effective input tokens, 21.8% is structural waste. We present Pichay, a demand paging system for LLM context windows. Implemented as a transparent proxy between client and inference API, Pichay interposes on the message stream to evict stale content, detect page faults when the model re-requests evicted material, and pin working-set pages identified by fault history. In offline replay across 1.4 million simulated evictions, the fault rate is 0.0254%. In live production deployment over 681turns, the system reduces context consumption by up to 93% (5,038KB to 339KB); under extreme sustained pressure, the system remains operational but exhibits the expected thrashing pathology, with repeated fault-in of evicted content. The key observation is that the problems the field faces, such as context limits, attention degradation, cost scaling, lost state across sessions, are virtual memory problems wearing different clothes. The solutions exist: working set theory (Denning, 1968), demand paging, fault-driven replacement policies, and memory hierarchies with multiple eviction-managed levels. We describe the architecture of a full memory hierarchy for LLM systems (L1 through persistent storage), report on the first three levels deployed in production use (L1 eviction, L2 fault-driven pinning, L3 model-initiated conversation compaction), and identify cross-session memory as the remaining frontier.", "authors": ["Tony Mason"], "categories": ["cs.OS", "cs.AI", "cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-09", "url": "https://arxiv.org/abs/2603.09023", "pdf_url": "https://arxiv.org/pdf/2603.09023v1", "arxiv_id": "2603.09023", "doi": "10.48550/arXiv.2603.09023", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5019}
{"id": "2b8a21d95c6bee8b94f4ee556b18da0eea35a8008336b3c770963fe98342358a", "sources": ["arxiv", "semantic_scholar"], "title": "Stacked from One: Multi-Scale Self-Injection for Context Window Extension", "abstract": "The limited context window of contemporary large language models (LLMs) remains a primary bottleneck for their broader application across diverse domains. Although continual pre-training on long-context data offers a straightforward solution, it incurs prohibitive data acquisition and computational costs. To address this challenge, we propose~\\modelname, a novel framework based on multi-grained context compression and query-aware information acquisition. SharedLLM comprises two stacked short-context LLMs: a lower model serving as a compressor and an upper model acting as a decoder. The lower model compresses long inputs into compact, multi-grained representations, which are then forwarded to the upper model for context-aware processing. To maximize efficiency, this information transfer occurs exclusively at the lowest layers, bypassing lengthy forward passes and redundant cross-attention operations. This entire process, wherein the upper and lower models are derived from the same underlying LLM layers, is termed~\\textit{self-injection}. To support this architecture, a specialized tree-based data structure enables the efficient encoding and query-aware retrieval of contextual information. Despite being trained on sequences of only 8K tokens, \\modelname~effectively generalizes to inputs exceeding 128K tokens. Across a comprehensive suite of long-context modeling and understanding benchmarks, \\modelname~achieves performance superior or comparable to strong baselines, striking an optimal balance between efficiency and accuracy. Furthermore, these design choices allow \\modelname~to substantially reduce the memory footprint and yield notable inference speedups ($2\\times$ over streaming and $3\\times$ over encoder-decoder architectures).", "authors": ["Wei Han", "Pan Zhou", "Soujanya Poria", "Shuicheng Yan"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-05", "url": "https://arxiv.org/abs/2603.04759", "pdf_url": "https://arxiv.org/pdf/2603.04759v2", "arxiv_id": "2603.04759", "doi": "10.48550/arXiv.2603.04759", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4973}
{"id": "c8e0d3dea9f23866623e571254506854d6667c3ce002b9724f97f79dcd4a56b7", "sources": ["arxiv", "semantic_scholar"], "title": "Beyond the Context Window: A Cost-Performance Analysis of Fact-Based Memory vs. Long-Context LLMs for Persistent Agents", "abstract": "Persistent conversational AI systems face a choice between passing full conversation histories to a long-context large language model (LLM) and maintaining a dedicated memory system that extracts and retrieves structured facts. We compare a fact-based memory system built on the Mem0 framework against long-context LLM inference on three memory-centric benchmarks - LongMemEval, LoCoMo, and PersonaMemv2 - and evaluate both architectures on accuracy and cumulative API cost. Long-context GPT-5-mini achieves higher factual recall on LongMemEval and LoCoMo, while the memory system is competitive on PersonaMemv2, where persona consistency depends on stable, factual attributes suited to flat-typed extraction. We construct a cost model that incorporates prompt caching and show that the two architectures have structurally different cost profiles: long-context inference incurs a per-turn charge that grows with context length even under caching, while the memory system's per-turn read cost remains roughly fixed after a one-time write phase. At a context length of 100k tokens, the memory system becomes cheaper after approximately ten interaction turns, with the break-even point decreasing as context length grows. These results characterize the accuracy-cost trade-off between the two approaches and provide a concrete criterion for selecting between them in production deployments.", "authors": ["Natchanon Pollertlam", "Witchayut Kornsuwannawit"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-05", "url": "https://arxiv.org/abs/2603.04814", "pdf_url": "https://arxiv.org/pdf/2603.04814v1", "arxiv_id": "2603.04814", "doi": "10.48550/arXiv.2603.04814", "citation_count": 4, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4973}
{"id": "51de898e81b52c5238dccc58b5833b1a38699553fc50f6ad7d40f2f9bb076db9", "sources": ["arxiv", "semantic_scholar"], "title": "Engaging students with statistics through choice of real data context on homework", "abstract": "Statistics educators recommend teaching with real data with relevant contexts, but defining relevancy is challenging and varies by student. We investigated whether providing student choice of data context increases engagement through a quasi-experiment in two sections of an introductory probability and statistics course at a large public university (n=65 consenting students). Sections alternated as treatment and control: during their treatment, students chose weekly homework from three similar instructor-provided options varying by data context; during control weeks, they received randomly assigned contexts. We found no significant difference in homework grades between treatment and control conditions. However, thematic analysis revealed students with choice reported enhanced engagement and motivation, greater appreciation for statistics' real-world value, and increased autonomy. Students overwhelmingly preferred contexts relevant to their interests, experiences, daily lives, and career paths-though preferences varied considerably across individuals. Based on these findings, we provide four recommendations for statistics educators: (1) use real data with authentic contexts, (2) select contexts students care about, (3) incorporate variety across data contexts, and (4) consider choice as a pedagogical tool.", "authors": ["Catalina Medina", "Mine Dogucu"], "categories": ["stat.OT"], "fields_of_study": ["Mathematics"], "published_date": "2026-03-04", "url": "https://arxiv.org/abs/2603.04541", "pdf_url": "https://arxiv.org/pdf/2603.04541v1", "arxiv_id": "2603.04541", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/CatalinaMedina/data-context-choice-manuscript", "venue": null, "quality_score": 0.5864}
{"id": "75bddd0f5b9c27d3377c5c56060eed52afdd8611f5f51f95b43ae1f719c3e6a6", "sources": ["arxiv", "semantic_scholar"], "title": "Cross-Family Speculative Prefill: Training-Free Long-Context Compression with Small Draft Models", "abstract": "Prompt length is a major bottleneck in agentic large language model (LLM) workloads, where repeated inference steps and multi-call loops incur substantial prefill cost. Recent work on speculative prefill demonstrates that attention-based token importance estimation can enable training-free prompt compression, but this assumes the existence of a draft model that shares the same tokenizer as the target model. In practice, however, agentic pipelines frequently employ models without any smaller in-family draft model. In this work, we study cross-family speculative prefill, where a lightweight draft model from one model family is used to perform prompt compression for a target model from a different family. Using the same speculative prefill mechanism as prior work, we evaluate a range of cross-family draft-target combinations, including Qwen, LLaMA, and DeepSeek models. Across a broad diversity of tasks, we find that attention-based token importance estimation transfers reliably across different model families despite differences in model architectures and tokenizers between draft and target models. Cross-model prompt compression largely retains 90~100% of full-prompt baseline performance and, in some cases, slightly improves accuracy due to denoising effects, while delivering substantial reductions in time to first token (TTFT). These results suggest that speculative prefill depends mainly on task priors and semantic structure, thus serving as a generalizable prompt compression primitive. We discuss the implications of our findings for agentic systems, where repeated long-context inference and heterogeneous model stacks make cross-model prompt compression both necessary and practical.", "authors": ["Shubhangi Upasani", "Ravi Shanker Raju", "Bo Li", "Mengmeng Ji", "John Long", "Chen Wu", "Urmish Thakker", "Guangtao Wang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-03-03", "url": "https://arxiv.org/abs/2603.02631", "pdf_url": "https://arxiv.org/pdf/2603.02631v3", "arxiv_id": "2603.02631", "doi": "10.48550/arXiv.2603.02631", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.495}
{"id": "cda5d857d46364efc3cc7d88650c1060475241a753e44356eea16e5e24d4d82a", "sources": ["arxiv", "semantic_scholar"], "title": "An Evaluation of Context Length Extrapolation in Long Code via Positional Embeddings and Efficient Attention", "abstract": "The rapid advancement of large language models (LLMs) has led to a significant increase in automated tools in the software engineering, capable of performing various code-related tasks such as code generation, completion, and translation. Despite these advancements, its effectiveness is constrained by fixed context lengths, limiting its ability to generalize across long, domain-specific code sequences. To address this challenge, we investigate zero-shot, inference-only methods aimed at improving position encodings and optimizing attention mechanisms. Our goal is to provide a thorough analysis of current approaches that facilitate context length extrapolation in code, particularly in the context of long code completion tasks.", "authors": ["Madhusudan Ghosh", "Rishabh Gupta"], "categories": ["cs.SE", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-25", "url": "https://arxiv.org/abs/2602.21800", "pdf_url": "https://arxiv.org/pdf/2602.21800v1", "arxiv_id": "2602.21800", "doi": "10.48550/arXiv.2602.21800", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4881}
{"id": "413272986dce8c473b1378d4176d8fb74837b5510154f840be758bd58256dab0", "sources": ["arxiv", "semantic_scholar"], "title": "Efficient Scaling of LLM Training with Flexible Context Parallelism", "abstract": "Scaling long-context capabilities is crucial for Large Language Models (LLMs). However, real-world data contain a large number of sequences with heterogeneous lengths. Existing training libraries for LLMs rely on static parallelism strategies, which suffer from severe load imbalance, redundant communication, and suboptimal hardware utilization under data heterogeneity. In this work, we propose Flexible Context Parallelism (FCP), an efficient parallelism strategy that adaptively reconfigures communication groups and context parallelism degrees during LLM training. We generalize more flexible non-power-of-two parallelism degrees and develop a polynomial-time algorithm to generate near-optimal parallelism strategies with only millisecond-level overhead per training batch. FCP is able to maintain high hardware efficiency even under extreme data heterogeneity. Experimental results demonstrate that FCP significantly outperforms Megatron-LM and DeepSpeed in both LLM and MLLM training, achieving up to 1.46x speedup in average throughput while maintaining near-linear scaling efficiency across large-scale clusters. For extremely unbalanced batches, FCP even achieves 2.24x speedup.", "authors": ["Yifan Niu", "Han Xiao", "Dongyi Liu", "Wei Zhou", "Jia Li"], "categories": ["cs.DC", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-25", "url": "https://arxiv.org/abs/2602.21788", "pdf_url": "https://arxiv.org/pdf/2602.21788v2", "arxiv_id": "2602.21788", "doi": null, "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3106}
{"id": "c401b699dfe6ac4dce45d02635aa05f556db1d6c9abb94963c5f540e1f306965", "sources": ["arxiv", "semantic_scholar"], "title": "Codified Context: Infrastructure for AI Agents in a Complex Codebase", "abstract": "LLM-based agentic coding assistants lack persistent memory: they lose coherence across sessions, forget project conventions, and repeat known mistakes. Recent studies characterize how developers configure agents through manifest files, but an open challenge remains how to scale such configurations for large, multi-agent projects. This paper presents a three-component codified context infrastructure developed during construction of a 108,000-line C# distributed system: (1) a hot-memory constitution encoding conventions, retrieval hooks, and orchestration protocols; (2) 19 specialized domain-expert agents; and (3) a cold-memory knowledge base of 34 on-demand specification documents. Quantitative metrics on infrastructure growth and interaction patterns across 283 development sessions are reported alongside four observational case studies illustrating how codified context propagates across sessions to prevent failures and maintain consistency. The framework is published as an open-source companion repository.", "authors": ["Aristidis Vasilopoulos"], "categories": ["cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-24", "url": "https://arxiv.org/abs/2602.20478", "pdf_url": "https://arxiv.org/pdf/2602.20478v1", "arxiv_id": "2602.20478", "doi": "10.48550/arXiv.2602.20478", "citation_count": 5, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/arisvas4/codified-context-infrastructure", "venue": "arXiv.org", "quality_score": 0.7526}
{"id": "84f59786c029fe44b0337c996559b7747914ef397e3cdeee8345f350e4f10618", "sources": ["arxiv", "semantic_scholar"], "title": "CHESS: Context-aware Hierarchical Efficient Semantic Selection for Long-Context LLM Inference", "abstract": "Long-context LLMs demand accurate inference at low latency, yet decoding becomes primarily constrained by KV cache as context grows. Prior pruning methods are largely context-agnostic: their token selection ignores step-wise relevance and local semantics, which undermines quality. Moreover, their irregular accesses and selection overheads yield only limited wall-clock speedups. To address this, we propose \\textbf{CHESS}, an \\textit{algorithm-system co-design} KV-cache management system. Algorithmically, CHESS introduces a context-aware, hierarchical selection policy that dynamically reconstructs a coherent context for the current decoding. System-wise, coarse granularity selection eliminates expensive data movement, fully realizing practical acceleration from theoretical sparsity. Extensive evaluations demonstrate that CHESS surpasses Full-KV quality using only \\textbf{1\\%} of the KV cache, delivers low-latency stable inference with up to \\textbf{4.56$\\times$} higher throughput, and consistently outperforms other strong baselines. Code is available at \\href{https://anonymous.4open.science/r/CHESS-9958/}{https://anonymous.4open.science/r/CHESS/}.", "authors": ["Chao Fei", "Guozhong Li", "Chenxi Liu", "Panos Kalnis"], "categories": ["cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-24", "url": "https://arxiv.org/abs/2602.20732", "pdf_url": "https://arxiv.org/pdf/2602.20732v1", "arxiv_id": "2602.20732", "doi": "10.48550/arXiv.2602.20732", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.7526}
{"id": "7eaae19db333ce5795413102d01660dbc4af6420cef1d14a556d4bc891e4cb60", "sources": ["arxiv", "semantic_scholar"], "title": "FAST-Prefill: FPGA Accelerated Sparse Attention for Long Context LLM Prefill", "abstract": "In long-context large language model (LLM) inference, the prefill stage dominates computation due to self-attention over the complete input context. Sparse attention significantly reduces self-attention computation by limiting each token's interactions to a subset of tokens. The attention sparsity pattern varies across input prompts, and within a prompt, each attention head can follow a distinct pattern. This makes attention sparsity dynamic. The requirement of generating the sparsity pattern, combined with limited data reuse in attention, shifts the prefill compute to being memory-bound. This, in addition to the huge energy requirements for long-context inference on GPU, motivates FPGAs as good candidates for accelerating dynamic long-context inference. To tackle these challenges, we propose FAST-Prefill, the first FPGA accelerator for long-context prefill-stage inference with dynamic sparse attention. To efficiently generate sparse indices, we propose a \\textit{fused pipeline unit with a memory-aware execution order} to reduce large tensors and irregular memory accesses. To reduce off-chip memory traffic for accessing the KV cache, we utilize the memory hierarchy to design a \\textit{liveness-driven, dual-tier cache}. For high-throughput matrix multiplication, we design a \\textit{hybrid Matrix Processing Unit (MPU)} with DSPs and bit-plane decomposition using LUTs. We implement FAST-Prefill on Alveo U280 and evaluate it on the Llama and Qwen models (batch size = 1) for context lengths ranging from 4K to 128K tokens. We demonstrate an average speedup of up to 2.5$\\times$ in TTFT and 4.5$\\times$ improvement in energy efficiency over GPU implementation on Nvidia A5000 GPU.", "authors": ["Rakshith Jayanth", "Viktor Prasanna"], "categories": ["cs.AR"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-24", "url": "https://arxiv.org/abs/2602.20515", "pdf_url": "https://arxiv.org/pdf/2602.20515v1", "arxiv_id": "2602.20515", "doi": "10.1109/FCCM68464.2026.00067", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Symposium on Field-Programmable Custom Computing Machines", "quality_score": 0.487}
{"id": "83bb72a6c671955abcd9c00198968afa356cba61c217aa763d83d05792ce160e", "sources": ["arxiv", "semantic_scholar"], "title": "The Convergence of Schema-Guided Dialogue Systems and the Model Context Protocol", "abstract": "This paper establishes a fundamental convergence: Schema-Guided Dialogue (SGD) and the Model Context Protocol (MCP) represent two manifestations of a unified paradigm for deterministic, auditable LLM-agent interaction. SGD, designed for dialogue-based API discovery (2019), and MCP, now the de facto standard for LLM-tool integration, share the same core insight -- that schemas can encode not just tool signatures but operational constraints and reasoning guidance. By analyzing this convergence, we extract five foundational principles for schema design: (1) Semantic Completeness over Syntactic Precision, (2) Explicit Action Boundaries, (3) Failure Mode Documentation, (4) Progressive Disclosure Compatibility, and (5) Inter-Tool Relationship Declaration. These principles reveal three novel insights: first, SGD's original design was fundamentally sound and should be inherited by MCP; second, both frameworks leave failure modes and inter-tool relationships unexploited -- gaps we identify and resolve; third, progressive disclosure emerges as a critical production-scaling insight under real-world token constraints. We provide concrete design patterns for each principle. These principles position schema-driven governance as a scalable mechanism for AI system oversight without requiring proprietary system inspection -- central to Software 3.0.", "authors": ["Andreas Schlapbach"], "categories": ["cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-21", "url": "https://arxiv.org/abs/2602.18764", "pdf_url": "https://arxiv.org/pdf/2602.18764v2", "arxiv_id": "2602.18764", "doi": "10.48550/arXiv.2602.18764", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4835}
{"id": "f5d78dd048be0bace346ad59c2439ae2d7373214385c3bd2fb4756925718e7c1", "sources": ["arxiv", "semantic_scholar"], "title": "The Limits of Long-Context Reasoning in Automated Bug Fixing", "abstract": "Rapidly increasing context lengths have led to the assumption that large language models (LLMs) can directly reason over entire codebases. Concurrently, recent advances in LLMs have enabled strong performance on software engineering benchmarks, particularly when paired with agentic workflows. In this work, we systematically evaluate whether current LLMs can reliably perform long-context code debugging and patch generation. Using SWE-bench Verified as a controlled experimental setting, we first evaluate state-of-the-art models within an agentic harness (mini-SWE-agent), where performance improves substantially: GPT-5-nano achieves up to a 31\\% resolve rate on 100 samples, and open-source models such as Deepseek-R1-0528 obtain competitive results. However, token-level analysis shows that successful agentic trajectories typically remain under 20k-30k tokens, and that longer accumulated contexts correlate with lower success rates, indicating that agentic success primarily arises from task decomposition into short-context steps rather than effective long-context reasoning. To directly test long-context capability, we construct a data pipeline where we artificially inflate the context length of the input by placing the relevant files into the context (ensuring perfect retrieval recall); we then study single-shot patch generation under genuinely long contexts (64k tokens). Despite this setup, performance degrades sharply: Qwen3-Coder-30B-A3B achieves only a 7\\% resolve rate at 64k context, while GPT-5-nano solves none of the tasks. Qualitative analysis reveals systematic failure modes, including hallucinated diffs, incorrect file targets, and malformed patch headers. Overall, our findings highlight a significant gap between nominal context length and usable context capacity in current LLMs, and suggest that existing agentic coding benchmarks do not meaningfully evaluate long-context reasoning.", "authors": ["Ravi Raju", "Mengmeng Ji", "Shubhangi Upasani", "Bo Li", "Urmish Thakker"], "categories": ["cs.SE", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-17", "url": "https://arxiv.org/abs/2602.16069", "pdf_url": "https://arxiv.org/pdf/2602.16069v2", "arxiv_id": "2602.16069", "doi": "10.48550/arXiv.2602.16069", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.7402}
{"id": "fb4bc9491ead5aeec4e0cc8902c14df21ea19cca373121c598f087ff2a3e752c", "sources": ["arxiv", "semantic_scholar"], "title": "Long Context, Less Focus: A Scaling Gap in LLMs Revealed through Privacy and Personalization", "abstract": "Large language models (LLMs) are increasingly deployed in privacy-critical and personalization-oriented scenarios, yet the role of context length in shaping privacy leakage and personalization effectiveness remains largely unexplored. We introduce a large-scale benchmark, PAPerBench, to systematically study how increasing context length influences both personalization quality and privacy protection in LLMs. The benchmark comprises approximately 29,000 instances with context lengths ranging from 1K to 256K tokens, yielding a total of 377K evaluation questions. It jointly evaluates personalization performance and privacy risks across diverse scenarios, enabling controlled analysis of long-context model behavior. Extensive evaluations across state-of-the-art LLMs reveal consistent performance degradation in both personalization and privacy as context length increases. We further provide a theoretical analysis of attention dilution under context scaling, explaining this behavior as an inherent limitation of soft attention in fixed-capacity Transformers. The empirical and theoretical findings together suggest a general scaling gap in current models -- long context, less focus. We release the benchmark to support reproducible evaluation and future research on scalable privacy and personalization. Code and data are available at https://github.com/SafeRL-Lab/PAPerBench", "authors": ["Shangding Gu"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-16", "url": "https://arxiv.org/abs/2602.15028", "pdf_url": "https://arxiv.org/pdf/2602.15028v1", "arxiv_id": "2602.15028", "doi": "10.48550/arXiv.2602.15028", "citation_count": 5, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/SafeRL-Lab/PAPerBench", "venue": "arXiv.org", "quality_score": 0.7384}
{"id": "ee5b8a1b03844136c9510a199b2b7f0eb5dda020710272f67c8904848ac7bdbf", "sources": ["arxiv", "semantic_scholar"], "title": "GPT-5 vs Other LLMs in Long Short-Context Performance", "abstract": "With the significant expansion of the context window in Large Language Models (LLMs), these models are theoretically capable of processing millions of tokens in a single pass. However, research indicates a significant gap between this theoretical capacity and the practical ability of models to robustly utilize information within long contexts, especially in tasks that require a comprehensive understanding of numerous details. This paper evaluates the performance of four state-of-the-art models (Grok-4, GPT-4, Gemini 2.5, and GPT-5) on long short-context tasks. For this purpose, three datasets were used: two supplementary datasets for retrieving culinary recipes and math problems, and a primary dataset of 20K social media posts for depression detection. The results show that as the input volume on the social media dataset exceeds 5K posts (70K tokens), the performance of all models degrades significantly, with accuracy dropping to around 50-53% for 20K posts. Notably, in the GPT-5 model, despite the sharp decline in accuracy, its precision remained high at approximately 95%, a feature that could be highly effective for sensitive applications like depression detection. This research also indicates that the \"lost in the middle\" problem has been largely resolved in newer models. This study emphasizes the gap between the theoretical capacity and the actual performance of models on complex, high-volume data tasks and highlights the importance of metrics beyond simple accuracy for practical applications.", "authors": ["Nima Esmi", "Maryam Nezhad-Moghaddam", "Fatemeh Borhani", "Asadollah Shahbahrami", "Amin Daemdoost", "Georgi Gaydadjiev"], "categories": ["cs.CL", "cs.AI", "cs.HC"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-15", "url": "https://arxiv.org/abs/2602.14188", "pdf_url": "https://arxiv.org/pdf/2602.14188v1", "arxiv_id": "2602.14188", "doi": "10.1109/FLLM67465.2025.11391194", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3033}
{"id": "14f13d3c68f93ce82c8008f5913dc1a0a762d433c5fe3bfd3d9d2c1d50133fe5", "sources": ["arxiv", "semantic_scholar"], "title": "Rotary Positional Embeddings as Phase Modulation: Theoretical Bounds on the RoPE Base for Long-Context Transformers", "abstract": "Rotary positional embeddings (RoPE) are widely used in large language models to encode token positions through multiplicative rotations, yet their behavior at long context lengths remains poorly characterized. In this work, we reinterpret RoPE as phase modulation applied to a bank of complex oscillators, enabling analysis through classical signal processing theory. Under this formulation, we derive principled lower bounds on the RoPE base parameter that are necessary to preserve positional coherence over a target context length. These include a fundamental aliasing bound, analogous to a Nyquist limit, and a DC-component stability bound that constrains phase drift in low-frequency positional modes. We further extend this analysis to deep transformers, showing that repeated rotary modulation across layers compounds angular misalignment, tightening the base requirement as depth increases. Complementing these results, we derive a precision-dependent upper bound on the RoPE base arising from finite floating-point resolution. Beyond this limit, incremental phase updates become numerically indistinguishable, leading to positional erasure even in the absence of aliasing. Together, the lower and upper bounds define a precision- and depth-dependent feasibility region a Goldilocks zone for long-context transformers. We validate the framework through a comprehensive case study of state-of-the-art models, including LLaMA, Mistral, and DeepSeek variants, showing that observed successes, failures, and community retrofits align closely with the predicted bounds. Notably, models that violate the stability bound exhibit attention collapse and long-range degradation, while attempts to scale beyond one million tokens encounter a hard precision wall independent of architecture or training.", "authors": ["Feilong Liu"], "categories": ["cs.LG", "cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-11", "url": "https://arxiv.org/abs/2602.10959", "pdf_url": "https://arxiv.org/pdf/2602.10959v1", "arxiv_id": "2602.10959", "doi": "10.48550/arXiv.2602.10959", "citation_count": 2, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4721}
{"id": "e1b6d5d4c993ed34494e78635a871f27d614c55c4cfd5cb6b15716b3e3ea34f3", "sources": ["arxiv"], "title": "When Less is More: The LLM Scaling Paradox in Context Compression", "abstract": "Scaling up model parameters has long been a prevalent training paradigm driven by the assumption that larger models yield superior generation capabilities. However, under lossy context compression in a compressor--decoder setup, we find a \\textbf{\\textit{Size-Fidelity Paradox}}: increasing compressor size can lessen the faithfulness of reconstructed contexts though reconstruction error decreases. Across 27 compressor setups spanning model families, scales, and compression rates, we coin this paradox arising from two dominant factors: 1) \\textit{knowledge overwriting}: larger models increasingly replace source facts with their own prior beliefs, \\textit{e.g.}, ``the white strawberry`` $\\to$ ``the red strawberry``; and 2) \\textit{semantic drift}: larger models tend to paraphrase or restructure content instead of reproducing it verbatim, \\textit{e.g.}, ``Alice hit Bob`` $\\to$ ``Bob hit Alice``. Interestingly, this paradox persists across varied settings, with mid-sized compressors often outperforming larger ones in faithful recovery. By analyzing the compressed memory via embedding geometry and reconstruction determinacy, we further reveal that compressors tend to organize memory across broader semantic subspaces, yielding more ambiguous representations prone to overwriting, drift, and weakened recovery. These findings complement existing evaluations of context compression and expose a breakdown of scaling laws when the objective shifts from plausible generation to faithful preservation.", "authors": ["Ruishan Guo", "Yibing Liu", "Guoxin Ma", "Yan Wang", "Yueyang Zhang", "Long Xia", "Kecheng Chen", "Zhiyuan Sun", "Daiting Shi"], "categories": ["cs.LG"], "fields_of_study": [], "published_date": "2026-02-10", "url": "https://arxiv.org/abs/2602.09789", "pdf_url": "https://arxiv.org/pdf/2602.09789v3", "arxiv_id": "2602.09789", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2997}
{"id": "27e7fa1e4a41edec0b76c3ce0179d191cd76c6ad98095c9c9eb432271f187c42", "sources": ["arxiv", "semantic_scholar"], "title": "Do Reasoning LLMs Refuse What They Infer in Long Contexts?", "abstract": "Long-context LLMs can infer objectives that are not stated explicitly. This capability is useful for reasoning over documents, code, retrieved evidence, and tool traces, but it also creates a safety risk: harmful intent can be distributed across a context and become visible only after the model composes the relevant pieces. Existing safety evaluations mostly test explicit harmful requests, and therefore miss this failure mode. We introduce compositional reasoning attacks, a long-context threat model in which harmful requests are decomposed into semantically incomplete fragments and embedded in long contexts. The final query is neutral; the harmful objective emerges only if the model retrieves the fragments, composes them, and infers the implied goal. We instantiate this setting using AdvBench requests, varying the required reasoning from Direct Retrieval to Single-hop Aggregation, Chain Reasoning, and Multi-hop Deductive Reasoning, and evaluate 15 frontier LLMs on contexts up to 64k tokens. Models usually refuse harmful requests when they are directly retrievable. However, refusal rates drop sharply when the same objectives must be reconstructed compositionally, often with larger failures in longer contexts. Benign reconstruction and fragment-position analyses indicate that these failures are not mainly retrieval errors: models often infer the harmful objective and then comply. Increasing inference-time reasoning improves refusal but remains incomplete and costly. Our results reveal a long-context safety gap: current models are better at refusing harmful requests they see than harmful objectives they infer.", "authors": ["Yu Fu", "Haz Sameen Shahgir", "Huanli Gong", "Zhipeng Wei", "N. Benjamin Erichson", "Yue Dong"], "categories": ["cs.CL", "cs.CR"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-09", "url": "https://arxiv.org/abs/2602.08874", "pdf_url": "https://arxiv.org/pdf/2602.08874v2", "arxiv_id": "2602.08874", "doi": null, "citation_count": 3, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.299}
{"id": "8ea0865f89f2e692b94f0e8602c5fa84d8369634ae8a9835b5f3844bdb6c15f8", "sources": ["arxiv", "semantic_scholar"], "title": "Context Forcing: Consistent Autoregressive Video Generation with Long Context", "abstract": "Recent approaches to real-time long video generation typically employ streaming tuning strategies, attempting to train a long-context student using a short-context (memoryless) teacher. In these frameworks, the student performs long rollouts but receives supervision from a teacher limited to short 5-second windows. This structural discrepancy creates a critical \\textbf{student-teacher mismatch}: the teacher's inability to access long-term history prevents it from guiding the student on global temporal dependencies, effectively capping the student's context length. To resolve this, we propose \\textbf{Context Forcing}, a novel framework that trains a long-context student via a long-context teacher. By ensuring the teacher is aware of the full generation history, we eliminate the supervision mismatch, enabling the robust training of models capable of long-term consistency. To make this computationally feasible for extreme durations (e.g., 2 minutes), we introduce a context management system that transforms the linearly growing context into a \\textbf{Slow-Fast Memory} architecture, significantly reducing visual redundancy. Extensive results demonstrate that our method enables effective context lengths exceeding 20 seconds -- 2 to 10 times longer than state-of-the-art methods like LongLive and Infinite-RoPE. By leveraging this extended context, Context Forcing preserves superior consistency across long durations, surpassing state-of-the-art baselines on various long video evaluation metrics.", "authors": ["Shuo Chen", "Cong Wei", "Sun Sun", "Ping Nie", "Kai Zhou", "Ge Zhang", "Ming-Hsuan Yang", "Wenhu Chen"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-05", "url": "https://arxiv.org/abs/2602.06028", "pdf_url": "https://arxiv.org/pdf/2602.06028v1", "arxiv_id": "2602.06028", "doi": "10.48550/arXiv.2602.06028", "citation_count": 24, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4652}
{"id": "0b5a70fb0622d4e0debeb8451906e5578f9cb5f2a6eef83a90433750838cd8af", "sources": ["arxiv", "semantic_scholar"], "title": "Simulated Adoption: Decoupling Magnitude and Direction in LLM In-Context Conflict Resolution", "abstract": "Large Language Models (LLMs) frequently prioritize conflicting in-context information over pre-existing parametric memory, a phenomenon often termed sycophancy or compliance. However, the mechanistic realization of this behavior remains obscure, specifically how the model resolves these knowledge conflicts through compliance, and whether this suppression arises from signal magnitude dilution or directional geometric alteration within the residual stream. To resolve this, we conducted a layer-wise geometric analysis across Qwen-3-4B, Llama-3.1-8B, and GLM-4-9B, decomposing the residual stream updates induced by counter-factual contexts into radial (norm-based) and angular (cosine-based) components. Our empirical results reject the universality of the \"Manifold Dilution\" hypothesis, as two of the three architectures maintained stable residual norms despite exhibiting significant performance degradation on factual queries. Instead, we observed that compliance is consistently characterized by \"Orthogonal Interference,\" where the conflicting context injects a steering vector that is quasi-orthogonal to the ground-truth direction, effectively rotating the hidden state representation. This suggests that models do not \"unlearn\" or suppress the magnitude of internal truths but rather employ a mechanism of geometric displacement to bypass the correct unembedding vector, effectively simulating adoption while preserving the original structural magnitude. These findings challenge scalar confidence metrics for detecting hallucinations and underscore the necessity of vectorial monitoring to distinguish between genuine knowledge integration and superficial in-context mimicry.", "authors": ["Long Zhang", "Fangwei Lin"], "categories": ["cs.LG", "cs.CL", "cs.CY"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-04", "url": "https://arxiv.org/abs/2602.04918", "pdf_url": "https://arxiv.org/pdf/2602.04918v2", "arxiv_id": "2602.04918", "doi": "10.48550/arXiv.2602.04918", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4641}
{"id": "b0f2c8b875e126495561622b98c904e33b28b54775b4984e682f473f252452aa", "sources": ["arxiv", "semantic_scholar"], "title": "LycheeDecode: Accelerating Long-Context LLM Inference via Hybrid-Head Sparse Decoding", "abstract": "The proliferation of long-context large language models (LLMs) exposes a key bottleneck: the rapidly expanding key-value cache during decoding, which imposes heavy memory and latency costs. While recent approaches attempt to alleviate this by sharing a single set of crucial tokens across layers, such coarse-grained sharing undermines model performance by neglecting the functional diversity of attention heads. To address this, we propose LycheeDecode, an efficient decoding method centered on a fine-grained hybrid-head attention mechanism that employs a hardware-efficient top-k selection strategy. Specifically, the novel HardKuma-based mechanism partitions attention heads into a small subset of retrieval heads that dynamically identify crucial tokens and a majority of sparse heads that reuse them for efficient computation. Through extensive experiments on leading models like Llama3 and Qwen3 across diverse benchmarks for long-context understanding (e.g., LongBench, RULER) and complex reasoning (e.g., AIME24, OlympiadBench), we demonstrate that LycheeDecode achieves generative quality comparable to, and at times surpassing even the full-attention baseline. Crucially, this is accomplished with up to a 2.7x speedup at a 128K context length. By preserving the functional diversity of attention heads, our fine-grained strategy overcomes the performance bottlenecks of existing methods, providing a powerful and validated pathway to both efficient and high-quality long-context LLM inference.", "authors": ["Gang Lin", "Dongfang Li", "Zhuoen Chen", "Yukun Shi", "Xuhui Chen", "Baotian Hu", "Min Zhang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-04", "url": "https://arxiv.org/abs/2602.04541", "pdf_url": "https://arxiv.org/pdf/2602.04541v1", "arxiv_id": "2602.04541", "doi": "10.48550/arXiv.2602.04541", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4641}
{"id": "807269a479aa415a2eee04b41e12407345a60c7fcfa41bfd848d5b3269655a63", "sources": ["arxiv", "semantic_scholar"], "title": "ATACompressor: Adaptive Task-Aware Compression for Efficient Long-Context Processing in LLMs", "abstract": "Long-context inputs in large language models (LLMs) often suffer from the \"lost in the middle\" problem, where critical information becomes diluted or ignored due to excessive length. Context compression methods aim to address this by reducing input size, but existing approaches struggle with balancing information preservation and compression efficiency. We propose Adaptive Task-Aware Compressor (ATACompressor), which dynamically adjusts compression based on the specific requirements of the task. ATACompressor employs a selective encoder that compresses only the task-relevant portions of long contexts, ensuring that essential information is preserved while reducing unnecessary content. Its adaptive allocation controller perceives the length of relevant content and adjusts the compression rate accordingly, optimizing resource utilization. We evaluate ATACompressor on three QA datasets: HotpotQA, MSMARCO, and SQUAD-showing that it outperforms existing methods in terms of both compression efficiency and task performance. Our approach provides a scalable solution for long-context processing in LLMs. Furthermore, we perform a range of ablation studies and analysis experiments to gain deeper insights into the key components of ATACompressor.", "authors": ["Xuancheng Li", "Haitao Li", "Yujia Zhou", "Qingyao Ai", "Yiqun Liu"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-03", "url": "https://arxiv.org/abs/2602.03226", "pdf_url": "https://arxiv.org/pdf/2602.03226v1", "arxiv_id": "2602.03226", "doi": "10.1145/3767695.3769499", "citation_count": 5, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2946}
{"id": "4cafd1d6d3c3c34db8799a0eff28e85946ea39a61f07d4838832a51ba32abf3d", "sources": ["arxiv", "semantic_scholar"], "title": "Focus-dLLM: Accelerating Long-Context Diffusion LLM Inference via Confidence-Guided Context Focusing", "abstract": "Diffusion Large Language Models (dLLMs) deliver strong long-context processing capability in a non-autoregressive decoding paradigm. However, the considerable computational cost of bidirectional full attention limits the inference efficiency. Although sparse attention is promising, existing methods remain ineffective. This stems from the need to estimate attention importance for tokens yet to be decoded, while the unmasked token positions are unknown during diffusion. In this paper, we present Focus-dLLM, a novel training-free attention sparsification framework tailored for accurate and efficient long-context dLLM inference. Based on the finding that token confidence strongly correlates across adjacent steps, we first design a past confidence-guided indicator to predict unmasked regions. Built upon this, we propose a sink-aware pruning strategy to accurately estimate and remove redundant attention computation, while preserving highly influential attention sinks. To further reduce overhead, this strategy reuses identified sink locations across layers, leveraging the observed cross-layer consistency. Experimental results show that our method offers more than $29\\times$ lossless speedup under $32K$ context length. The code is publicly available at: https://github.com/Longxmas/Focus-dLLM", "authors": ["Lingkun Long", "Yushi Huang", "Shihao Bai", "Ruihao Gong", "Jun Zhang", "Ao Zhou", "Jianlei Yang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-02-02", "url": "https://arxiv.org/abs/2602.02159", "pdf_url": "https://arxiv.org/pdf/2602.02159v1", "arxiv_id": "2602.02159", "doi": "10.48550/arXiv.2602.02159", "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Longxmas/Focus-dLLM", "venue": "arXiv.org", "quality_score": 0.7136}
{"id": "46df33c3d74aa5f184d48530720b01ae1d1d36000a78bcde895c9c0c638059c3", "sources": ["arxiv", "semantic_scholar"], "title": "Latent Context Compilation: Distilling Long Context into Compact Portable Memory", "abstract": "Efficient long-context LLM deployment is stalled by a dichotomy between amortized compression, which struggles with out-of-distribution generalization, and Test-Time Training, which incurs prohibitive synthetic data costs and requires modifying model weights, creating stateful parameters that complicate concurrent serving. We propose Latent Context Compilation, a framework that fundamentally shifts context processing from adaptation to compilation. By utilizing a disposable LoRA module as a compiler, we distill long contexts into compact buffer tokens -- stateless, portable memory artifacts that are plug-and-play compatible with frozen base models. Crucially, we introduce a self-aligned optimization strategy that eliminates the need for synthetic context-relevant QA pairs. By regularizing context reconstruction task with context-agnostic random queries, we force compressed tokens to reside within the model's existing instruction-following manifold. Experiments with Llama-3.1-8B demonstrate that Latent Context Compilation preserves fine-grained details and reasoning capabilities where prior methods falter, effectively decoupling memory density from model parameters even at a 16x compression ratio.", "authors": ["Zeju Li", "Yizhou Zhou", "Qiang Xu"], "categories": ["cs.LG", "cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-01-31", "url": "https://arxiv.org/abs/2602.21221", "pdf_url": "https://arxiv.org/pdf/2602.21221v1", "arxiv_id": "2602.21221", "doi": "10.48550/arXiv.2602.21221", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4595}
{"id": "d804ea50b350643e4331b8dbe9a9d18bbd90dee658af47b9cc53c02386cc2b54", "sources": ["arxiv", "semantic_scholar"], "title": "Epistemic Context Learning: Building Trust the Right Way in LLM-Based Multi-Agent Systems", "abstract": "Individual agents in multi-agent (MA) systems often lack robustness, tending to blindly conform to misleading peers. We show this weakness stems from both sycophancy and inadequate ability to evaluate peer reliability. To address this, we first formalize the learning problem of history-aware reference, introducing the historical interactions of peers as additional input, so that agents can estimate peer reliability and learn from trustworthy peers when uncertain. This shifts the task from evaluating peer reasoning quality to estimating peer reliability based on interaction history. We then develop Epistemic Context Learning (ECL): a reasoning framework that conditions predictions on explicitly-built peer profiles from history. We further optimize ECL by reinforcement learning using auxiliary rewards. Our experiments reveal that our ECL enables small models like Qwen 3-4B to outperform a history-agnostic baseline 8x its size (Qwen 3-30B) by accurately identifying reliable peers. ECL also boosts frontier models to near-perfect (100%) performance. We show that ECL generalizes well to various MA configurations and we find that trust is modeled well by LLMs, revealing a strong correlation in trust modeling accuracy and final answer quality.", "authors": ["Ruiwen Zhou", "Maojia Song", "Xiaobao Wu", "Sitao Cheng", "Xunjian Yin", "Yuxi Xie", "Zhuoqun Hao", "Wenyue Hua", "Liangming Pan", "Soujanya Poria", "Min-Yen Kan"], "categories": ["cs.AI", "cs.CL", "cs.MA"], "fields_of_study": ["Computer Science"], "published_date": "2026-01-29", "url": "https://arxiv.org/abs/2601.21742", "pdf_url": "https://arxiv.org/pdf/2601.21742v1", "arxiv_id": "2601.21742", "doi": "10.48550/arXiv.2601.21742", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/skyriver-2000/epistemic-context-learning", "venue": "arXiv.org", "quality_score": 0.7066}
{"id": "5c961e467799ace511d902bebffa8ef48bfc543c3ecc142fcc08bffeafaa805e", "sources": ["arxiv", "semantic_scholar"], "title": "Self-Manager: Parallel Agent Loop for Long-form Deep Research", "abstract": "Long-form deep research requires multi-faceted investigations over extended horizons to get a comprehensive report. When handling such complex tasks, existing agents manage context at the subtask level to overcome linear context accumulation and information loss. However, they still adhere to a single context window and sequential execution paradigm, which results in mutual interference and blocking behavior, restricting scalability and adaptability. To address this issue, this paper introduces Self-Manager, a parallel agent loop that enables asynchronous and concurrent execution. The main thread can create multiple subthreads, each with its own isolated context, and manage them iteratively through Thread Control Blocks, allowing for more focused and flexible parallel agent execution. To assess its effectiveness, we benchmark Self-Manager on DeepResearch Bench, where it consistently outperforms existing single-agent loop baselines across all metrics. Furthermore, we conduct extensive analytical experiments to demonstrate the necessity of Self-Manager's design choices, as well as its advantages in contextual capacity, efficiency, and generalization.", "authors": ["Yilong Xu", "Zhi Zheng", "Xiang Long", "Yujun Cai", "Yiwei Wang"], "categories": ["cs.CL", "cs.AI", "cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2026-01-25", "url": "https://arxiv.org/abs/2601.17879", "pdf_url": "https://arxiv.org/pdf/2601.17879v1", "arxiv_id": "2601.17879", "doi": "10.48550/arXiv.2601.17879", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4526}
{"id": "7278e81ef9098d766f6d3acb7d2db59e9d11c200a2689c31cedf8960958d2031", "sources": ["arxiv", "semantic_scholar"], "title": "Gated Sparse Attention: Combining Computational Efficiency with Training Stability for Long-Context Language Models", "abstract": "The computational burden of attention in long-context language models has motivated two largely independent lines of work: sparse attention mechanisms that reduce complexity by attending to selected tokens, and gated attention variants that improve training sta-bility while mitigating the attention sink phenomenon. We observe that these approaches address complementary weaknesses and propose Gated Sparse Attention (GSA), an architecture that realizes the benefits of both. GSA incorporates a gated lightning indexer with sigmoid activations that produce bounded, interpretable selection scores, an adaptive sparsity controller that modulates the number of attended tokens based on local uncertainty, and dual gating at the value and output stages. We establish theoretical foundations for the approach, including complexity analysis, expressiveness results, and convergence guarantees. In experiments with 1.7B parameter models trained on 400B tokens, GSA matches the efficiency of sparse-only baselines (12-16x speedup at 128K context) while achieving the quality gains associated with gated attention: perplexity improves from 6.03 to 5.70, RULER scores at 128K context nearly double, and attention to the first token, a proxy for attention sinks, drops from 47% to under 4%. Training stability improves markedly, with loss spikes reduced by 98%.", "authors": ["Alfred Shen", "Aaron Shen"], "categories": ["cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-01-12", "url": "https://arxiv.org/abs/2601.15305", "pdf_url": "https://arxiv.org/pdf/2601.15305v1", "arxiv_id": "2601.15305", "doi": "10.48550/arXiv.2601.15305", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4377}
{"id": "6f44d8298ad1ee1486089bec0800b1d87d5c5300504806d3cee5d79f92c33721", "sources": ["arxiv", "semantic_scholar"], "title": "DYCP: Dynamic Context Pruning for Long-Form Dialogue with LLMs", "abstract": "Large Language Models (LLMs) increasingly operate over long-form dialogues with frequent topic shifts. While recent LLMs support extended context windows, efficient management of dialogue history in practice is needed due to inference cost and latency constraints. We present DyCP, a lightweight context management method implemented outside the LLM that dynamically identifies and retrieves relevant dialogue segments conditioned on the current turn, without offline memory construction. DyCP manages dialogue context while preserving the sequential nature of dialogue without predefined topic boundaries, enabling adaptive and efficient context selection. Across three long-form dialogue benchmarks-LoCoMo, MT-Bench+, and SCM4LLMs-and multiple LLM backends, DyCP achieves competitive answer quality in downstream generation, with more selective context usage and improved inference efficiency.", "authors": ["Nayoung Choi", "Jonathan Zhang", "Jinho D. Choi"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-01-12", "url": "https://arxiv.org/abs/2601.07994", "pdf_url": "https://arxiv.org/pdf/2601.07994v5", "arxiv_id": "2601.07994", "doi": "10.48550/arXiv.2601.07994", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4377}
{"id": "0aab31a66da9c8bd3c4b129d9784f7c381c82292a016d37830bdca24e60a0565", "sources": ["arxiv", "semantic_scholar"], "title": "Intelligence Degradation in Long-Context LLMs: Critical Threshold Determination via Natural Length Distribution Analysis", "abstract": "Large Language Models (LLMs) exhibit catastrophic performance degradation when processing contexts approaching certain critical thresholds, even when information remains relevant. This intelligence degradation-defined as over 30% drop in task performance-severely limits long-context applications. This degradation shows a common pattern: models maintain strong performance up to a critical threshold, then collapse catastrophically. We term this shallow long-context adaptation-models adapt for short to medium contexts but fail beyond critical thresholds. This paper presents three contributions: (1) Natural Length Distribution Analysis: We use each sample's natural token length without truncation or padding, providing stronger causal evidence that degradation results from context length itself. (2) Critical Threshold Determination: Through experiments on a mixed dataset (1,000 samples covering 5%-95% of context length), we identify the critical threshold for Qwen2.5-7B at 40-50% of maximum context length, where F1 scores drop from 0.55-0.56 to 0.3 (45.5% degradation), using five-method cross-validation. (3) Unified Framework: We consolidate shallow adaptation, explaining degradation patterns and providing a foundation for mitigation strategies. This work provides the first systematic characterization of intelligence degradation in open-source Qwen models, offering practical guidance for deploying LLMs in long-context scenarios.", "authors": ["Weiwei Wang", "Jiyong Min", "Weijie Zou"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2026-01-07", "url": "https://arxiv.org/abs/2601.15300", "pdf_url": "https://arxiv.org/pdf/2601.15300v1", "arxiv_id": "2601.15300", "doi": "10.48550/arXiv.2601.15300", "citation_count": 5, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.6676}
{"id": "78efcd52b71fc05b3c54ad8d4574e8e24308b057b4e76c1fc2b6f74d67ec3789", "sources": ["arxiv", "semantic_scholar"], "title": "Not All Needles Are Found: How Fact Distribution and Don't Make It Up Prompts Shape Literal Extraction, Logical Inference, and Hallucination Risks in Long-Context LLMs", "abstract": "Large language models (LLMs) increasingly support very long input contexts. Yet it remains unclear how reliably they extract and infer information at scale. Performance varies with context length and strongly interacts with how information is distributed in real-world corpora. Motivated by these observations, we study how fact placement, corpus-level fact distributions, and Don't Make It Up prompts influence model behavior. We introduce an extended needle-in-a-haystack benchmark across four production-scale models: Gemini-2.5-flash, ChatGPT-5-mini, Claude-4.5-haiku, and Deepseek-v3.2-chat. Unlike prior work, we separately evaluate literal extraction, logical inference, and hallucination risk. Our study considers both positional effects and realistic distributions of evidence across long contexts, as well as prompts that explicitly discourage fabrication. We find that longer contexts alone do not guarantee better performance and can be detrimental when relevant evidence is diluted or widely dispersed. Performance varies substantially across models: some show severe degradation under realistic conditions, while others remain more robust at longer context lengths. Anti-hallucination (AH) instructions can make some models overly conservative, sharply reducing accuracy in literal extraction and logical inference. While we do not directly compare retrieval-augmented generation (RAG) and cache-augmented generation (CAG), our results suggest many failures stem from ineffective context utilization. Models often struggle to identify and prioritize relevant information even when it is present. These findings have direct practical implications, as enterprise workflows increasingly involve pasting large volumes of unfiltered documents into LLM prompts. Effective context length and model-specific robustness to long contexts are therefore critical for reliable LLM deployment in research and business.", "authors": ["Amirali Ebrahimzadeh", "Seyyed M. Salili"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2026-01-05", "url": "https://arxiv.org/abs/2601.02023", "pdf_url": "https://arxiv.org/pdf/2601.02023v1", "arxiv_id": "2601.02023", "doi": "10.48550/arXiv.2601.02023", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4297}
{"id": "243f02f2b418695987969db683ec8a2523536d6c403ccf8b41d7038890b528ea", "sources": ["arxiv", "semantic_scholar"], "title": "Context as a Tool: Context Management for Long-Horizon SWE-Agents", "abstract": "Agents based on large language models have recently shown strong potential on real-world software engineering (SWE) tasks that require long-horizon interaction with repository-scale codebases. However, most existing agents rely on append-only context maintenance or passively triggered compression heuristics, which often lead to context explosion, semantic drift, and degraded reasoning in long-running interactions. We propose CAT, a new context management paradigm that elevates context maintenance to a callable tool integrated into the decision-making process of agents. CAT formalizes a structured context workspace consisting of stable task semantics, condensed long-term memory, and high-fidelity short-term interactions, and enables agents to proactively compress historical trajectories into actionable summaries at appropriate milestones. To support context management for SWE-agents, we propose a trajectory-level supervision framework, CAT-GENERATOR, based on an offline data construction pipeline that injects context-management actions into complete interaction trajectories. Using this framework, we train a context-aware model, SWE-Compressor. Experiments on SWE-Bench-Verified demonstrate that SWE-Compressor reaches a 57.6% solved rate and significantly outperforms ReAct-based agents and static compression baselines, while maintaining stable and scalable long-horizon reasoning under a bounded context budget.", "authors": ["Shukai Liu", "Jian Yang", "Bo Jiang", "Yizhi Li", "Jinyang Guo", "Xianglong Liu", "Bryan Dai"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-26", "url": "https://arxiv.org/abs/2512.22087", "pdf_url": "https://arxiv.org/pdf/2512.22087v1", "arxiv_id": "2512.22087", "doi": "10.48550/arXiv.2512.22087", "citation_count": 17, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4182}
{"id": "8394c56ad80db360672a9a08d121eb00f92b5184fdfd810feda4251e8d148cb3", "sources": ["arxiv", "semantic_scholar"], "title": "Context Discipline and Performance Correlation: Analyzing LLM Performance and Quality Degradation Under Varying Context Lengths", "abstract": "The scaling trend in Large Language Models (LLMs) has prioritized increasing the maximum context window to facilitate complex, long-form reasoning and document analysis. However, managing this expanded context introduces severe computational overhead. This paper investigates the critical trade-off between system performance and model quality when dense transformer architectures--specifically Llama-3.1-70B and Qwen1.5-14B--are exposed to large volumes of irrelevant and distracting context. The research identifies a non-linear performance degradation tied to the growth of the Key-Value (KV) cache. Furthermore, an extended analysis of the Mixture-of-Experts (MoE) architecture reveals unique behavioral anomalies at varying context scales, suggesting that architectural benefits may be masked by infrastructure bottlenecks at high token volumes.", "authors": ["Ahilan Ayyachamy Nadar Ponnusamy", "Karthic Chandran", "M Maruf Hossain"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-25", "url": "https://arxiv.org/abs/2601.11564", "pdf_url": "https://arxiv.org/pdf/2601.11564v1", "arxiv_id": "2601.11564", "doi": "10.48550/arXiv.2601.11564", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4171}
{"id": "55203c81c15ca6a804e6ad22c4913450a531af5ac42c965c649bf55bbcd0c5d9", "sources": ["arxiv", "semantic_scholar"], "title": "RePo: Language Models with Context Re-Positioning", "abstract": "In-context learning is fundamental to modern Large Language Models (LLMs); however, prevailing architectures impose a rigid and fixed contextual structure by assigning linear or constant positional indices. The rigid position information poses the full burden of organizing the input structure to attention layers, thus reducing the amount of attention that could be allocated for more critical information. To address this, we propose RePo, a novel mechanism that alleviates the burden for attention layers via context re-positioning. Unlike conventional approaches, RePo utilizes a differentiable module, $f_φ$, to assign token positions that capture contextual dependencies, rather than replying on pre-defined order. By continually pre-training on the OLMo-2 1B \\& 7B models, we demonstrate that RePo consistently enhances performance on tasks involving noisy contexts, structured data, and longer context length, while maintaining competitive performance on general short-context tasks. Analysis reveals that RePo successfully allocates more attention mass to distant but relevant information, assigns positions in a dense and non-linear space, and captures the intrinsic structure of the input context. Our code is at https://github.com/SakanaAI/repo.", "authors": ["Huayang Li", "Tianyu Zhao", "Deng Cai", "Richard Sproat"], "categories": ["cs.LG", "cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-16", "url": "https://arxiv.org/abs/2512.14391", "pdf_url": "https://arxiv.org/pdf/2512.14391v3", "arxiv_id": "2512.14391", "doi": "10.48550/arXiv.2512.14391", "citation_count": 2, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/SakanaAI/repo", "venue": "arXiv.org", "quality_score": 0.6286}
{"id": "18b8148da2d671c9551047bf58aa7c12e0082b6f14ade32ce03bc4f0a67be01e", "sources": ["arxiv", "semantic_scholar"], "title": "Let's (not) just put things in Context: Test-Time Training for Long-Context LLMs", "abstract": "Progress on training and architecture strategies has enabled LLMs with millions of tokens in context length. However, empirical evidence suggests that such long-context LLMs can consume far more text than they can reliably use. On the other hand, it has been shown that inference-time compute can be used to scale performance of LLMs, often by generating thinking tokens, on challenging tasks involving multi-step reasoning. Through controlled experiments on sandbox long-context tasks, we find that such inference-time strategies show rapidly diminishing returns and fail at long context. We attribute these failures to score dilution, a phenomenon inherent to static self-attention. Further, we show that current inference-time strategies cannot retrieve relevant long-context signals under certain conditions. We propose a simple method that, through targeted gradient updates on the given context, provably overcomes limitations of static self-attention. We find that this shift in how inference-time compute is spent leads to consistently large performance improvements across models and long-context benchmarks. Our method leads to large 12.6 and 14.1 percentage point improvements for Qwen3-4B on average across subsets of LongBench-v2 and ZeroScrolls benchmarks. The takeaway is practical: for long context, a small amount of context-specific training is a better use of inference compute than current inference-time scaling strategies like producing more thinking tokens.", "authors": ["Rachit Bansal", "Aston Zhang", "Rishabh Tiwari", "Lovish Madaan", "Sai Surya Duvvuri", "Devvrit Khatri", "David Brandfonbrener", "David Alvarez-Melis", "Prajjwal Bhargava", "Mihir Sanjay Kale", "Samy Jelassi"], "categories": ["cs.LG", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-15", "url": "https://arxiv.org/abs/2512.13898", "pdf_url": "https://arxiv.org/pdf/2512.13898v1", "arxiv_id": "2512.13898", "doi": "10.48550/arXiv.2512.13898", "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4056}
{"id": "0c047ce00d9aad7cff9a60711a9e88631239d6af4e1152ff58ffcf7c0e3714b0", "sources": ["arxiv", "semantic_scholar"], "title": "Extending the Context of Pretrained LLMs by Dropping Their Positional Embeddings", "abstract": "So far, expensive finetuning beyond the pretraining sequence length has been a requirement for effectively extending the context of language models (LM). In this work, we break this key bottleneck by Dropping the Positional Embeddings of LMs after training (DroPE). Our simple method is motivated by three key theoretical and empirical observations. First, positional embeddings (PEs) serve a crucial role during pretraining, providing an important inductive bias that significantly facilitates convergence. Second, over-reliance on this explicit positional information is also precisely what prevents test-time generalization to sequences of unseen length, even when using popular PE-scaling methods. Third, positional embeddings are not an inherent requirement of effective language modeling and can be safely removed after pretraining, following a short recalibration phase. Empirically, DroPE yields seamless zero-shot context extension without any long-context finetuning, quickly adapting pretrained LMs without compromising their capabilities in the original training context. Our findings hold across different models and dataset sizes, far outperforming previous specialized architectures and established rotary positional embedding scaling methods.", "authors": ["Yoav Gelberg", "Koshi Eguchi", "Takuya Akiba", "Edoardo Cetin"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-13", "url": "https://arxiv.org/abs/2512.12167", "pdf_url": "https://arxiv.org/pdf/2512.12167v1", "arxiv_id": "2512.12167", "doi": "10.48550/arXiv.2512.12167", "citation_count": 14, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4033}
{"id": "27d3ee1a30061fd942a4878c391d7378afe0edbfb830f25b07f3fde736306d2d", "sources": ["arxiv", "semantic_scholar"], "title": "In-Context Learning for Seismic Data Processing", "abstract": "Seismic processing transforms raw data into subsurface images essential for geophysical applications. Traditional methods face challenges, such as noisy data, and manual parameter tuning, among others. Recently deep learning approaches have proposed alternative solutions to some of these problems. However, important challenges of existing deep learning approaches are spatially inconsistent results across neighboring seismic gathers and lack of user-control. We address these limitations by introducing ContextSeisNet, an in-context learning model, to seismic demultiple processing. Our approach conditions predictions on a support set of spatially related example pairs: neighboring common-depth point gathers from the same seismic line and their corresponding labels. This allows the model to learn task-specific processing behavior at inference time by observing how similar gathers should be processed, without any retraining. This method provides both flexibility through user-defined examples and improved lateral consistency across seismic lines. On synthetic data, ContextSeisNet outperforms a U-Net baseline quantitatively and demonstrates enhanced spatial coherence between neighboring gathers. On field data, our model achieves superior lateral consistency compared to both traditional Radon demultiple and the U-Net baseline. Relative to the U-Net, ContextSeisNet also delivers improved near-offset performance and more complete multiple removal. Notably, ContextSeisNet achieves comparable field data performance despite being trained on 90% less data, demonstrating substantial data efficiency. These results establish ContextSeisNet as a practical approach for spatially consistent seismic demultiple with potential applicability to other seismic processing tasks.", "authors": ["Fabian Fuchs", "Mario Ruben Fernandez", "Norman Ettrich", "Janis Keuper"], "categories": ["cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-12", "url": "https://arxiv.org/abs/2512.11575", "pdf_url": "https://arxiv.org/pdf/2512.11575v2", "arxiv_id": "2512.11575", "doi": "10.48550/arXiv.2512.11575", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.6216}
{"id": "dd55f26595320c6433dbd209cad47fcda5cd614787774afbd7003cfcada2bc4b", "sources": ["arxiv", "semantic_scholar"], "title": "Learning to Extract Context for Context-Aware LLM Inference", "abstract": "User prompts to large language models (LLMs) are often ambiguous or under-specified, and subtle contextual cues shaped by user intentions, prior knowledge, and risk factors strongly influence what constitutes an appropriate response. Misinterpreting intent or risks may lead to unsafe outputs, while overly cautious interpretations can cause unnecessary refusal of benign requests. In this paper, we question the conventional framework in which LLMs generate immediate responses to requests without considering broader contextual factors. User requests are situated within broader contexts such as intentions, knowledge, and prior experience, which strongly influence what constitutes an appropriate answer. We propose a framework that extracts and leverages such contextual information from the user prompt itself. Specifically, a reinforcement learning based context generator, designed in an autoencoder-like fashion, is trained to infer contextual signals grounded in the prompt and use them to guide response generation. This approach is particularly important for safety tasks, where ambiguous requests may bypass safeguards while benign but confusing requests can trigger unnecessary refusals. Experiments show that our method reduces harmful responses by an average of 5.6% on the SafetyInstruct dataset across multiple foundation models and improves the harmonic mean of attack success rate and compliance on benign prompts by 6.2% on XSTest and WildJailbreak. These results demonstrate the effectiveness of context extraction for safer and more reliable LLM inferences.", "authors": ["Minseon Kim", "Lucas Caccia", "Zhengyan Shi", "Matheus Pereira", "Marc-Alexandre Côté", "Xingdi Yuan", "Alessandro Sordoni"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-12", "url": "https://arxiv.org/abs/2512.11986", "pdf_url": "https://arxiv.org/pdf/2512.11986v1", "arxiv_id": "2512.11986", "doi": "10.48550/arXiv.2512.11986", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4022}
{"id": "79bf5fac5eb1656a2fb12cf366ec134ad0378c0319f064013abdeb26a40fde94", "sources": ["arxiv", "semantic_scholar"], "title": "SWAA: Sliding Window Attention Adaptation for Efficient and Quality Preserving Long Context Processing", "abstract": "The quadratic complexity of self attention in Transformer based LLMs renders long context inference prohibitively expensive. While Sliding Window Attention (SWA), the simplest sparse attention pattern, offers a linear complexity alternative, it suffers from catastrophic long context performance collapse, which stems from two fundamental factors: the training inference mismatch when naively applying SWA to models pretrained with Full Attention (FA), and the inherent structural inability to access distant information when applying SWA to every module at all times. To address these dual challenges, we propose Sliding Window Attention Adaptation (SWAA), a plug and play toolkit of recipes that adapts FA models to SWA without costly pretraining. SWAA systematically combines four core strategies to tackle these distinct issues: (1) Full Attention (FA) Decode and (2) Interleaving FA and SWA layers, which mitigate structural defects by selectively allowing access to distant information; alongside (3) preserving ``sink'' tokens and (4) lightweight fine tuning, which mitigate the training inference mismatch. Our experiments reveal that while isolated strategies are insufficient, specific synergistic combinations effectively recover long context performance. Despite varying computational overheads, our performance efficiency trade off analysis identifies optimal SWAA configurations for diverse scenarios, achieving 30% to 100% speedups for long context inference with acceptable quality retention. Our code, data and model weights are available at https://github.com/yuyijiong/sliding-window-attention-adaptation", "authors": ["Yijiong Yu", "Jiale Liu", "Qingyun Wu", "Huazheng Wang", "Ji Pei"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-11", "url": "https://arxiv.org/abs/2512.10411", "pdf_url": "https://arxiv.org/pdf/2512.10411v5", "arxiv_id": "2512.10411", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/yuyijiong/sliding-window-attention-adaptation", "venue": null, "quality_score": 0.474}
{"id": "2dc5cdca7a316bf7836b965959c874f48f66767090c22e7866fd99a4bfb5bcfc", "sources": ["arxiv", "semantic_scholar"], "title": "Beyond Real: Imaginary Extension of Rotary Position Embeddings for Long-Context LLMs", "abstract": "Rotary Position Embeddings (RoPE) have become a standard for encoding sequence order in Large Language Models (LLMs) by applying rotations to query and key vectors in the complex plane. Standard implementations, however, utilize only the real component of the complex-valued dot product for attention score calculation. This simplification discards the imaginary component, which contains valuable phase information, leading to a potential loss of relational details crucial for modeling long-context dependencies. In this paper, we propose an extension that re-incorporates this discarded imaginary component. Our method leverages the full complex-valued representation to create a dual-component attention score. We theoretically and empirically demonstrate that this approach enhances the modeling of long-context dependencies by preserving more positional information. Furthermore, evaluations on a suite of long-context language modeling benchmarks show that our method consistently improves performance over the standard RoPE, with the benefits becoming more significant as context length increases. The code is available at https://github.com/OpenMOSS/rope_pp.", "authors": ["Xiaoran Liu", "Yuerong Song", "Zhigeng Liu", "Zengfeng Huang", "Qipeng Guo", "Zhaoxiang Liu", "Shiguo Lian", "Ziwei He", "Xipeng Qiu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-08", "url": "https://arxiv.org/abs/2512.07525", "pdf_url": "https://arxiv.org/pdf/2512.07525v1", "arxiv_id": "2512.07525", "doi": "10.48550/arXiv.2512.07525", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/OpenMOSS/rope_pp", "venue": "arXiv.org", "quality_score": 0.6145}
{"id": "52cc2bfb38990295535f3718c728a7b405e6070653647442e1ff779364c70bc5", "sources": ["arxiv", "semantic_scholar"], "title": "Evaluating Long-Context Reasoning in LLM-Based WebAgents", "abstract": "As large language model (LLM)-based agents become increasingly integrated into daily digital interactions, their ability to reason across long interaction histories becomes crucial for providing personalized and contextually aware assistance. However, the performance of these agents in long context scenarios, particularly for action-taking WebAgents operating in realistic web environments, remains largely unexplored. This paper introduces a benchmark for evaluating long context reasoning capabilities of WebAgents through sequentially dependent subtasks that require retrieval and application of information from extended interaction histories. We develop a novel evaluation framework that simulates multi-session user interactions by injecting irrelevant task trajectories between dependent subtasks, creating contexts ranging from 25,000 to 150,000 tokens. Through extensive evaluation of four popular models, Claude-3.7, GPT-4.1, Llama 4, and o4-mini, we observe a dramatic performance degradation as context length increases, with success rates dropping from 40-50\\% in baseline conditions to less than 10\\% in long context scenarios. Our detailed error analysis reveals that agents primarily fail due to getting stuck in loops and losing track of original task objectives. We further propose an implicit RAG approach that provides modest improvements by generating task-relevant summaries, though fundamental limitations in long context reasoning persist. These findings highlight critical challenges for deploying WebAgents in realistic, long-term user interaction scenarios and provide insights for developing more robust agent architectures capable of maintaining coherent task execution across extended contexts.", "authors": ["Andy Chung", "Yichi Zhang", "Kaixiang Lin", "Aditya Rawal", "Qiaozi Gao", "Joyce Chai"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-03", "url": "https://arxiv.org/abs/2512.04307", "pdf_url": "https://arxiv.org/pdf/2512.04307v1", "arxiv_id": "2512.04307", "doi": "10.48550/arXiv.2512.04307", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3919}
{"id": "19ec99a03aec8e1850bdc8c239758780578e584250f1b35e2fb5fb4abdfc809d", "sources": ["arxiv", "semantic_scholar"], "title": "When Refusals Fail: Unstable Safety Mechanisms in Long-Context LLM Agents", "abstract": "Solving complex or long-horizon problems often requires large language models (LLMs) to use external tools and operate over a significantly longer context window. New LLMs enable longer context windows and support tool calling capabilities. Prior works have focused mainly on evaluation of LLMs on long-context prompts, leaving agentic setup relatively unexplored, both from capability and safety perspectives. Our work addresses this gap. We find that LLM agents could be sensitive to length, type, and placement of the context, exhibiting unexpected and inconsistent shifts in task performance and in refusals to execute harmful requests. Models with 1M-2M token context windows show severe degradation already at 100K tokens, with performance drops exceeding 50\\% for both benign and harmful tasks. Refusal rates shift unpredictably: GPT-4.1-nano increases from $\\sim$5\\% to $\\sim$40\\% while Grok 4 Fast decreases from $\\sim$80\\% to $\\sim$10\\% at 200K tokens. Our work shows potential safety issues with agents operating on longer context and opens additional questions on the current metrics and paradigm for evaluating LLM agent safety on long multi-step tasks. In particular, our results on LLM agents reveal a notable divergence in both capability and safety performance compared to prior evaluations of LLMs on similar criteria.", "authors": ["Tsimur Hadeliya", "Mohammad Ali Jauhar", "Nidhi Sakpal", "Diogo Cruz"], "categories": ["cs.LG", "cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-12-02", "url": "https://arxiv.org/abs/2512.02445", "pdf_url": "https://arxiv.org/pdf/2512.02445v1", "arxiv_id": "2512.02445", "doi": "10.48550/arXiv.2512.02445", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3907}
{"id": "a25a9efe0876bfd564e400d2845cfb5464b38f616c6815fe377ac99a2f108676", "sources": ["arxiv", "semantic_scholar"], "title": "SpeContext: Enabling Efficient Long-context Reasoning with Speculative Context Sparsity in LLMs", "abstract": "In this paper, we point out that the objective of the retrieval algorithms is to align with the LLM, which is similar to the objective of knowledge distillation in LLMs. We analyze the similarity in information focus between the distilled language model(DLM) and the original LLM from the perspective of information theory, and thus propose a novel paradigm that leverages a DLM as the retrieval algorithm. Based on the insight, we present SpeContext, an algorithm and system co-design for long-context reasoning. (1) At the algorithm level, SpeContext proposes lightweight retrieval head based on the head-level attention weights of DLM, achieving > 90% parameters reduction by pruning the redundancy. (2) At the system level, SpeContext designs an asynchronous prefetch dataflow via the elastic loading strategy, effectively overlapping KV cache retrieval with the LLM computation. (3) At the compilation level, SpeContext constructs the theoretical memory model and implements an adaptive memory management system to achieve acceleration by maximizing GPU memory utilization. We deploy and evaluate SpeContext in two resourceconstrained environments, cloud and edge. Extensive experiments show that, compared with the Huggingface framework, SpeContext achieves up to 24.89x throughput improvement in cloud and 10.06x speedup in edge with negligible accuracy loss, pushing the Pareto frontier of accuracy and throughput.", "authors": ["Jiaming Xu", "Jiayi Pan", "Hanzhen Wang", "Yongkang Zhou", "Jiancai Ye", "Yu Wang", "Guohao Dai"], "categories": ["cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-11-30", "url": "https://arxiv.org/abs/2512.00722", "pdf_url": "https://arxiv.org/pdf/2512.00722v1", "arxiv_id": "2512.00722", "doi": "10.1145/3779212.3790224", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Architectural Support for Programming Languages and Operating Systems", "quality_score": 0.3884}
{"id": "14c9abb244c05b7002708aae686f7bf2b9fd945e98503498ef4316624d5ff2b0", "sources": ["arxiv", "semantic_scholar"], "title": "LoCoBench-Agent: An Interactive Benchmark for LLM Agents in Long-Context Software Engineering", "abstract": "As large language models (LLMs) evolve into sophisticated autonomous agents capable of complex software development tasks, evaluating their real-world capabilities becomes critical. While existing benchmarks like LoCoBench~\\cite{qiu2025locobench} assess long-context code understanding, they focus on single-turn evaluation and cannot capture the multi-turn interactive nature, tool usage patterns, and adaptive reasoning required by real-world coding agents. We introduce \\textbf{LoCoBench-Agent}, a comprehensive evaluation framework specifically designed to assess LLM agents in realistic, long-context software engineering workflows. Our framework extends LoCoBench's 8,000 scenarios into interactive agent environments, enabling systematic evaluation of multi-turn conversations, tool usage efficiency, error recovery, and architectural consistency across extended development sessions. We also introduce an evaluation methodology with 9 metrics across comprehension and efficiency dimensions. Our framework provides agents with 8 specialized tools (file operations, search, code analysis) and evaluates them across context lengths ranging from 10K to 1M tokens, enabling precise assessment of long-context performance. Through systematic evaluation of state-of-the-art models, we reveal several key findings: (1) agents exhibit remarkable long-context robustness; (2) comprehension-efficiency trade-off exists with negative correlation, where thorough exploration increases comprehension but reduces efficiency; and (3) conversation efficiency varies dramatically across models, with strategic tool usage patterns differentiating high-performing agents. As the first long-context LLM agent benchmark for software engineering, LoCoBench-Agent establishes a rigorous foundation for measuring agent capabilities, identifying performance gaps, and advancing autonomous software development at scale.", "authors": ["Jielin Qiu", "Zuxin Liu", "Zhiwei Liu", "Rithesh Murthy", "Jianguo Zhang", "Haolin Chen", "Shiyu Wang", "Ming Zhu", "Liangwei Yang", "Juntao Tan", "Roshan Ram", "Akshara Prabhakar", "Tulika Awalgaonkar", "Zixiang Chen", "Zhepeng Cen", "Cheng Qian", "Shelby Heinecke", "Weiran Yao", "Silvio Savarese", "Caiming Xiong", "Huan Wang"], "categories": ["cs.SE", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-11-17", "url": "https://arxiv.org/abs/2511.13998", "pdf_url": "https://arxiv.org/pdf/2511.13998v1", "arxiv_id": "2511.13998", "doi": "10.48550/arXiv.2511.13998", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3735}
{"id": "0b672e8db3e5a49e37912c219919e1601fe032965bfd5ad9ec4a3d21a399b343", "sources": ["arxiv", "semantic_scholar"], "title": "PISanitizer: Preventing Prompt Injection to Long-Context LLMs via Prompt Sanitization", "abstract": "Long context LLMs are vulnerable to prompt injection, where an attacker can inject an instruction in a long context to induce an LLM to generate an attacker-desired output. Existing prompt injection defenses are designed for short contexts. When extended to long-context scenarios, they have limited effectiveness. The reason is that an injected instruction constitutes only a very small portion of a long context, making the defense very challenging. In this work, we propose PISanitizer, which first pinpoints and sanitizes potential injected tokens (if any) in a context before letting a backend LLM generate a response, thereby eliminating the influence of the injected instruction. To sanitize injected tokens, PISanitizer builds on two observations: (1) prompt injection attacks essentially craft an instruction that compels an LLM to follow it, and (2) LLMs intrinsically leverage the attention mechanism to focus on crucial input tokens for output generation. Guided by these two observations, we first intentionally let an LLM follow arbitrary instructions in a context and then sanitize tokens receiving high attention that drive the instruction-following behavior of the LLM. By design, PISanitizer presents a dilemma for an attacker: the more effectively an injected instruction compels an LLM to follow it, the more likely it is to be sanitized by PISanitizer. Our extensive evaluation shows that PISanitizer can successfully prevent prompt injection, maintain utility, outperform existing defenses, is efficient, and is robust to optimization-based and strong adaptive attacks. The code is available at https://github.com/sleeepeer/PISanitizer.", "authors": ["Runpeng Geng", "Yanting Wang", "Chenlong Yin", "Minhao Cheng", "Ying Chen", "Jinyuan Jia"], "categories": ["cs.CR", "cs.AI", "cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-11-13", "url": "https://arxiv.org/abs/2511.10720", "pdf_url": "https://arxiv.org/pdf/2511.10720v1", "arxiv_id": "2511.10720", "doi": "10.48550/arXiv.2511.10720", "citation_count": 8, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/sleeepeer/PISanitizer", "venue": "arXiv.org", "quality_score": 0.5702}
{"id": "e0ee946dd35f36c38332cb20a9d48f28537c16ab1099ec263ae6ce3c8449d759", "sources": ["arxiv", "semantic_scholar"], "title": "Trusted Multi-view Learning for Long-tailed Classification", "abstract": "Class imbalance has been extensively studied in single-view scenarios; however, addressing this challenge in multi-view contexts remains an open problem, with even scarcer research focusing on trustworthy solutions. In this paper, we tackle a particularly challenging class imbalance problem in multi-view scenarios: long-tailed classification. We propose TMLC, a Trusted Multi-view Long-tailed Classification framework, which makes contributions on two critical aspects: opinion aggregation and pseudo-data generation. Specifically, inspired by Social Identity Theory, we design a group consensus opinion aggregation mechanism that guides decision making toward the direction favored by the majority of the group. In terms of pseudo-data generation, we introduce a novel distance metric to adapt SMOTE for multi-view scenarios and develop an uncertainty-guided data generation module that produces high-quality pseudo-data, effectively mitigating the adverse effects of class imbalance. Extensive experiments on long-tailed multi-view datasets demonstrate that our model is capable of achieving superior performance. The code is released at https://github.com/cncq-tang/TMLC.", "authors": ["Chuanqing Tang", "Yifei Shi", "Guanghao Lin", "Lei Xing", "Long Shi"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-11-12", "url": "https://arxiv.org/abs/2511.09138", "pdf_url": "https://arxiv.org/pdf/2511.09138v1", "arxiv_id": "2511.09138", "doi": "10.48550/arXiv.2511.09138", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/cncq-tang/TMLC", "venue": "arXiv.org", "quality_score": 0.5684}
{"id": "804a1cf66455df4c460cf0c3681453a5f2cd32cb5b765c0aa549473b8ddacb38", "sources": ["arxiv", "semantic_scholar"], "title": "Sentence-Anchored Gist Compression for Long-Context LLMs", "abstract": "This work investigates context compression for Large Language Models (LLMs) using learned compression tokens to reduce the memory and computational demands of processing long sequences. We demonstrate that pre-trained LLMs can be fine-tuned to compress their context by factors of 2x to 8x without significant performance degradation, as evaluated on both short-context and long-context benchmarks. Furthermore, in experiments on a 3-billion-parameter LLaMA model, our method achieves results on par with alternative compression techniques while attaining higher compression ratios.", "authors": ["Dmitrii Tarasov", "Elizaveta Goncharova", "Kuznetsov Andrey"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-11-11", "url": "https://arxiv.org/abs/2511.08128", "pdf_url": "https://arxiv.org/pdf/2511.08128v1", "arxiv_id": "2511.08128", "doi": "10.48550/arXiv.2511.08128", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3667}
{"id": "72ce38e867c989408f4fa495a4efafe7f8cc28fa59e93642d6745483f4603928", "sources": ["arxiv", "semantic_scholar"], "title": "NoteEx: Interactive Visual Context Manipulation for LLM-Assisted Exploratory Data Analysis in Computational Notebooks", "abstract": "Computational notebooks have become popular for Exploratory Data Analysis (EDA), augmented by LLM-based code generation and result interpretation. Effective LLM assistance hinges on selecting informative context -- the minimal set of cells whose code, data, or outputs suffice to answer a prompt. As notebooks grow long and messy, users can lose track of the mental model of their analysis. They thus fail to curate appropriate contexts for LLM tasks, causing frustration and tedious prompt engineering. We conducted a formative study (n=6) that surfaced challenges in LLM context selection and mental model maintenance. Therefore, we introduce NoteEx, a JupyterLab extension that provides a semantic visualization of the EDA workflow, allowing analysts to externalize their mental model, specify analysis dependencies, and enable interactive selection of task-relevant contexts for LLMs. A user study (n=12) against a baseline shows that NoteEx improved mental model retention and context selection, leading to more accurate and relevant LLM responses.", "authors": ["Mohammad Hasan Payandeh", "Lin-Ping Yuan", "Jian Zhao"], "categories": ["cs.HC", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-11-10", "url": "https://arxiv.org/abs/2511.07223", "pdf_url": "https://arxiv.org/pdf/2511.07223v1", "arxiv_id": "2511.07223", "doi": "10.48550/arXiv.2511.07223", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3655}
{"id": "acf33bc72ac0e511bcf03c00e87758d3c70c52abf89b09a8db313f1225acfc30", "sources": ["arxiv", "semantic_scholar"], "title": "Vocabulary In-Context Learning in Transformers: Benefits of Positional Encoding", "abstract": "Numerous studies have demonstrated that the Transformer architecture possesses the capability for in-context learning (ICL). In scenarios involving function approximation, context can serve as a control parameter for the model, endowing it with the universal approximation property (UAP). In practice, context is represented by tokens from a finite set, referred to as a vocabulary, which is the case considered in this paper, \\emph{i.e.}, vocabulary in-context learning (VICL). We demonstrate that VICL in single-layer Transformers, without positional encoding, does not possess the UAP; however, it is possible to achieve the UAP when positional encoding is included. Several sufficient conditions for the positional encoding are provided. Our findings reveal the benefits of positional encoding from an approximation theory perspective in the context of ICL.", "authors": ["Qian Ma", "Ruoxiang Xu", "Yongqiang Cai"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-11-09", "url": "https://arxiv.org/abs/2511.06376", "pdf_url": "https://arxiv.org/pdf/2511.06376v1", "arxiv_id": "2511.06376", "doi": "10.48550/arXiv.2511.06376", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3644}
{"id": "7c5f09ea6dbe4abad5672aa86b1e1a1e081afbd3569830de117d0f233e81c799", "sources": ["arxiv", "semantic_scholar"], "title": "ContextPilot: Fast Long-Context Inference via Context Reuse", "abstract": "AI applications increasingly depend on long-context inference, where LLMs consume substantial context to support stronger reasoning. Common examples include retrieval-augmented generation, agent memory layers, and multi-agent orchestration. As input contexts get longer, prefill latency becomes the main bottleneck. Yet today's prefill acceleration techniques face a trade-off: they either preserve reasoning quality but deliver little KV-cache reuse, or improve reuse at the cost of degraded reasoning quality. We present ContextPilot, a system that accelerates prefill by introducing context reuse as a new mechanism for faster long-context inference. ContextPilot introduces a context index to identify overlapping context blocks across LLM interactions (e.g., across users and turns). It further proposes context ordering and de-duplication techniques to maximize KV-cache reuse. To preserve reasoning quality under reuse, it introduces succinct context annotations that prevent quality degradation. Finally, ContextPilot is built around a modular architecture with a clean interface that integrates with existing inference engines. Extensive evaluation shows that ContextPilot reduces LLM prefill latency by up to $3\\times{}$ compared to state-of-the-art methods while preserving reasoning quality. At longer context lengths, it can even improve reasoning quality. ContextPilot is open-sourced at: https://github.com/EfficientContext/ContextPilot.", "authors": ["Yinsicheng Jiang", "Yeqi Huang", "Liang Cheng", "Cheng Deng", "Xuan Sun", "Luo Mai"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-11-05", "url": "https://arxiv.org/abs/2511.03475", "pdf_url": "https://arxiv.org/pdf/2511.03475v4", "arxiv_id": "2511.03475", "doi": null, "citation_count": 4, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/EfficientContext/ContextPilot", "venue": null, "quality_score": 0.4252}
{"id": "e0fa8d97b9666cfaaee64b6a639761a889ad0ff2070ea4d9cf6f543c0ae33c99", "sources": ["arxiv", "semantic_scholar"], "title": "On the Role of Context for Discourse Relation Classification in Scientific Writing", "abstract": "With the increasing use of generative Artificial Intelligence (AI) methods to support science workflows, we are interested in the use of discourse-level information to find supporting evidence for AI generated scientific claims. A first step towards this objective is to examine the task of inferring discourse structure in scientific writing. In this work, we present a preliminary investigation of pretrained language model (PLM) and Large Language Model (LLM) approaches for Discourse Relation Classification (DRC), focusing on scientific publications, an under-studied genre for this task. We examine how context can help with the DRC task, with our experiments showing that context, as defined by discourse structure, is generally helpful. We also present an analysis of which scientific discourse relation types might benefit most from context.", "authors": ["Stephen Wan", "Wei Liu", "Michael Strube"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-30", "url": "https://arxiv.org/abs/2510.26354", "pdf_url": "https://arxiv.org/pdf/2510.26354v1", "arxiv_id": "2510.26354", "doi": "10.18653/v1/2025.codi-1.8", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2246}
{"id": "20385278108537ae188721339fbcdfe80917941a0ff55389dbe681308f1aa616", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Context Modeling with Dynamic Hierarchical Sparse Attention for Memory-Constrained LLM Inference", "abstract": "The quadratic cost of attention limits the scalability of long-context LLMs, especially under limited hardware memory budgets. While attention is often sparse, existing static sparse methods cannot adapt to task- or input-dependent variations, and recent dynamic approaches rely on predefined templates or heuristics that may sacrifice generality. We propose Dynamic Hierarchical Sparse Attention (DHSA), a data-driven framework that predicts attention sparsity online while keeping the LLM backbone frozen. DHSA performs hierarchical routing by estimating importance at the chunk level and propagating it to token-level interactions, preserving causally important dependencies while enabling efficient sparsification. Across Needle-in-a-Haystack test, LongBench and RULER, DHSA maintains near-dense accuracy in highly sparse regimes, achieving 12--20% relative accuracy gains over Block Sparse Attention at comparable prefill cost. With a memory-efficient tiled backend, DHSA delivers up to $10\\times$ prefill speedup at 128K context length. On LLaMA-3.1-8B (4-bit), DHSA scales to 100K context on a single 24GB GPU, where dense attention fails. We provide complementary GPU and CPU backends, enabling DHSA to run across diverse hardware environments and multiple open-weight model families. These results demonstrate DHSA as an efficient and adaptable solution for memory-constrained long-context LLM inference.", "authors": ["Siheng Xiong", "Joe Zou", "Faramarz Fekri", "Yae Jee Cho"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-28", "url": "https://arxiv.org/abs/2510.24606", "pdf_url": "https://arxiv.org/pdf/2510.24606v2", "arxiv_id": "2510.24606", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2231}
{"id": "812ac659e383a72b2a654fd7c83845ade882e6e72949b365ff6bb043118f7d88", "sources": ["arxiv", "semantic_scholar"], "title": "Seeing the Unseen: Mask-Driven Positional Encoding and Strip-Convolution Context Modeling for Cross-View Object Geo-Localization", "abstract": "Cross-view object geo-localization enables high-precision object localization through cross-view matching, with critical applications in autonomous driving, urban management, and disaster response. However, existing methods rely on keypoint-based positional encoding, which captures only 2D coordinates while neglecting object shape information, resulting in sensitivity to annotation shifts and limited cross-view matching capability. To address these limitations, we propose a mask-based positional encoding scheme that leverages segmentation masks to capture both spatial coordinates and object silhouettes, thereby upgrading the model from \"location-aware\" to \"object-aware.\" Furthermore, to tackle the challenge of large-span objects (e.g., elongated buildings) in satellite imagery, we design a context enhancement module. This module employs horizontal and vertical strip convolutional kernels to extract long-range contextual features, enhancing feature discrimination among strip-like objects. Integrating MPE and CEM, we present EDGeo, an end-to-end framework for robust cross-view object geo-localization. Extensive experiments on two public datasets (CVOGL and VIGOR-Building) demonstrate that our method achieves state-of-the-art performance, with a 3.39% improvement in localization accuracy under challenging ground-to-satellite scenarios. This work provides a robust positional encoding paradigm and a contextual modeling framework for advancing cross-view geo-localization research.", "authors": ["Shuhan Hu", "Yiru Li", "Yuanyuan Li", "Yingying Zhu"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-23", "url": "https://arxiv.org/abs/2510.20247", "pdf_url": "https://arxiv.org/pdf/2510.20247v1", "arxiv_id": "2510.20247", "doi": "10.48550/arXiv.2510.20247", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3449}
{"id": "d54363971a519ff25e47fb38159eb77b1cd593a797a8e14a4f4e0c51720222cc", "sources": ["arxiv", "semantic_scholar"], "title": "Stream: Scaling up Mechanistic Interpretability to Long Context in LLMs via Sparse Attention", "abstract": "As Large Language Models (LLMs) scale to million-token contexts, traditional Mechanistic Interpretability techniques for analyzing attention scale quadratically with context length, demanding terabytes of memory beyond 100,000 tokens. We introduce Sparse Tracing, a novel technique that leverages dynamic sparse attention to efficiently analyze long context attention patterns. We present Stream, a compilable hierarchical pruning algorithm that estimates per-head sparse attention masks in near-linear time $O(T \\log T)$ and linear space $O(T)$, enabling one-pass interpretability at scale. Stream performs a binary-search-style refinement to retain only the top-$k$ key blocks per query while preserving the model's next-token behavior. We apply Stream to long chain-of-thought reasoning traces and identify thought anchors while pruning 97-99\\% of token interactions. On the RULER benchmark, Stream preserves critical retrieval paths while discarding 90-96\\% of interactions and exposes layer-wise routes from the needle to output. Our method offers a practical drop-in tool for analyzing attention patterns and tracing information flow without terabytes of caches. By making long context interpretability feasible on consumer GPUs, Sparse Tracing helps democratize chain-of-thought monitoring. Code is available at https://anonymous.4open.science/r/stream-03B8/.", "authors": ["J Rosser", "José Luis Redondo García", "Gustavo Penha", "Konstantina Palla", "Hugues Bouchard"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-22", "url": "https://arxiv.org/abs/2510.19875", "pdf_url": "https://arxiv.org/pdf/2510.19875v2", "arxiv_id": "2510.19875", "doi": "10.48550/arXiv.2510.19875", "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5312}
{"id": "afdff49c65b2423e46271bb865cf7515fc982ab9936d60f639418a3b54626e1b", "sources": ["arxiv", "semantic_scholar"], "title": "Investigating LLM Capabilities on Long Context Comprehension for Medical Question Answering", "abstract": "This study is the first to investigate LLM comprehension capabilities over long-context (LC), clinically relevant medical Question Answering (QA) beyond MCQA. Our comprehensive approach considers a range of settings based on content inclusion of varying size and relevance, LLM models of different capabilities and a variety of datasets across task formulations. We reveal insights on model size effects and their limitations, underlying memorization issues and the benefits of reasoning models, while demonstrating the value and challenges of leveraging the full long patient's context. Importantly, we examine the effect of Retrieval Augmented Generation (RAG) on medical LC comprehension, showcasing best settings in single versus multi-document QA datasets. We shed light into some of the evaluation aspects using a multi-faceted approach uncovering common metric challenges. Our quantitative analysis reveals challenging cases where RAG excels while still showing limitations in cases requiring temporal reasoning.", "authors": ["Feras AlMannaa", "Talia Tseriotou", "Jenny Chim", "Maria Liakata"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-21", "url": "https://arxiv.org/abs/2510.18691", "pdf_url": "https://arxiv.org/pdf/2510.18691v2", "arxiv_id": "2510.18691", "doi": "10.48550/arXiv.2510.18691", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3426}
{"id": "d6f3903775d6d4833dcdfdd4ae08111942daabc1eb4f72d6e972af1f1a10f3b5", "sources": ["arxiv", "semantic_scholar"], "title": "MTraining: Distributed Dynamic Sparse Attention for Efficient Ultra-Long Context Training", "abstract": "The adoption of long context windows has become a standard feature in Large Language Models (LLMs), as extended contexts significantly enhance their capacity for complex reasoning and broaden their applicability across diverse scenarios. Dynamic sparse attention is a promising approach for reducing the computational cost of long-context. However, efficiently training LLMs with dynamic sparse attention on ultra-long contexts-especially in distributed settings-remains a significant challenge, due in large part to worker- and step-level imbalance. This paper introduces MTraining, a novel distributed methodology leveraging dynamic sparse attention to enable efficient training for LLMs with ultra-long contexts. Specifically, MTraining integrates three key components: a dynamic sparse training pattern, balanced sparse ring attention, and hierarchical sparse ring attention. These components are designed to synergistically address the computational imbalance and communication overheads inherent in dynamic sparse attention mechanisms during the training of models with extensive context lengths. We demonstrate the efficacy of MTraining by training Qwen2.5-3B, successfully expanding its context window from 32K to 512K tokens on a cluster of 32 A100 GPUs. Our evaluations on a comprehensive suite of downstream tasks, including RULER, PG-19, InfiniteBench, and Needle In A Haystack, reveal that MTraining achieves up to a 6x higher training throughput while preserving model accuracy. Our code is available at https://github.com/microsoft/MInference/tree/main/MTraining.", "authors": ["Wenxuan Li", "Chengruidong Zhang", "Huiqiang Jiang", "Yucheng Li", "Yuqing Yang", "Lili Qiu"], "categories": ["cs.CL", "cs.DC", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-21", "url": "https://arxiv.org/abs/2510.18830", "pdf_url": "https://arxiv.org/pdf/2510.18830v2", "arxiv_id": "2510.18830", "doi": "10.48550/arXiv.2510.18830", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/microsoft/MInference/tree/main/MTraining", "venue": "arXiv.org", "quality_score": 0.5295}
{"id": "a0d5f3b7e831f8d478302d342be7761377b8ef358a99a732d9b036ec3e0e9736", "sources": ["arxiv", "semantic_scholar"], "title": "AcademicEval: Live Long-Context LLM Benchmark", "abstract": "Large Language Models (LLMs) have recently achieved remarkable performance in long-context understanding. However, current long-context LLM benchmarks are limited by rigid context length, labor-intensive annotation, and the pressing challenge of label leakage issues during LLM training. Therefore, we propose \\textsc{AcademicEval}, a live benchmark for evaluating LLMs over long-context generation tasks. \\textsc{AcademicEval} adopts papers on arXiv to introduce several academic writing tasks with long-context inputs, \\textit{i.e.}, \\textsc{Title}, \\textsc{Abstract}, \\textsc{Introduction}, and \\textsc{Related Work}, which cover a wide range of abstraction levels and require no manual labeling. Moreover, \\textsc{AcademicEval} integrates high-quality and expert-curated few-shot demonstrations from a collected co-author graph to enable flexible context length. Especially, \\textsc{AcademicEval} features an efficient live evaluation, ensuring no label leakage. We conduct a holistic evaluation on \\textsc{AcademicEval}, and the results illustrate that LLMs perform poorly on tasks with hierarchical abstraction levels and tend to struggle with long few-shot demonstrations, highlighting the challenge of our benchmark. Through experimental analysis, we also reveal some insights for enhancing LLMs' long-context modeling capabilities. Code is available at https://github.com/ulab-uiuc/AcademicEval", "authors": ["Haozhen Zhang", "Tao Feng", "Pengrui Han", "Jiaxuan You"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-20", "url": "https://arxiv.org/abs/2510.17725", "pdf_url": "https://arxiv.org/pdf/2510.17725v1", "arxiv_id": "2510.17725", "doi": "10.48550/arXiv.2510.17725", "citation_count": 5, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/ulab-uiuc/AcademicEval", "venue": null, "quality_score": 0.4035}
{"id": "48cd18a4f615f6ea024025b3e4f84eefd672293ab62258e432e21a1c6f4a882b", "sources": ["arxiv", "semantic_scholar"], "title": "Glyph: Scaling Context Windows via Visual-Text Compression", "abstract": "Large language models (LLMs) increasingly rely on long-context modeling for tasks such as document understanding, code analysis, and multi-step reasoning. However, scaling context windows to the million-token level brings prohibitive computational and memory costs, limiting the practicality of long-context LLMs. In this work, we take a different perspective-visual context scaling-to tackle this challenge. Instead of extending token-based sequences, we propose Glyph, a framework that renders long texts into images and processes them with vision-language models (VLMs). This approach substantially compresses textual input while preserving semantic information, and we further design an LLM-driven genetic search to identify optimal visual rendering configurations for balancing accuracy and compression. Through extensive experiments, we demonstrate that our method achieves 3-4x token compression while maintaining accuracy comparable to leading LLMs such as Qwen3-8B on various long-context benchmarks. This compression also leads to around 4x faster prefilling and decoding, and approximately 2x faster SFT training. Furthermore, under extreme compression, a 128K-context VLM could scale to handle 1M-token-level text tasks. In addition, the rendered text data benefits real-world multimodal tasks, such as document understanding. Our code and model are released at https://github.com/thu-coai/Glyph.", "authors": ["Jiale Cheng", "Yusen Liu", "Xinyu Zhang", "Yulin Fei", "Wenyi Hong", "Ruiliang Lyu", "Weihan Wang", "Zhe Su", "Xiaotao Gu", "Xiao Liu", "Yushi Bai", "Jie Tang", "Hongning Wang", "Minlie Huang"], "categories": ["cs.CV", "cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-20", "url": "https://arxiv.org/abs/2510.17800", "pdf_url": "https://arxiv.org/pdf/2510.17800v2", "arxiv_id": "2510.17800", "doi": "10.48550/arXiv.2510.17800", "citation_count": 33, "influential_citation_count": 9, "has_code": true, "code_url": "https://github.com/thu-coai/Glyph", "venue": "arXiv.org", "quality_score": 0.5277}
{"id": "97659ae96b6fd7fc4762c741e705aecc1d7568afa7caed45cba3380487837258", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Context Attention Benchmark: From Kernel Efficiency to Distributed Context Parallelism", "abstract": "Transformer-based large language models (LLMs) have achieved remarkable success, yet their standard attention mechanism incurs quadratic computation and memory costs with respect to sequence length, posing a major bottleneck for long-context training. Prior work tackles this challenge along two directions: (1) kernel-level optimizations, which accelerate dense and sparse attention operators; and (2) module-level strategies, often referred to as distributed attention or context parallel training, which scale attention across multiple devices. However, systematic evaluation still remains limited: operator-level comparisons are often incomplete, while context parallel strategies are typically framework-specific, with unclear performance analysis across contexts. To address these gaps, we propose a unified benchmark that integrates representative attention kernels and context parallel mechanisms with a modular and extensible interface for evaluation. The benchmark evaluates methods along two critical dimensions: (1) attention mask patterns, which strongly affect efficiency, scalability, and usability, and (2) sequence length and distributed scale, which determine performance under extreme long-context training. Through comprehensive experiments on the cluster of up to 96 GPUs, our benchmark enables reproducible comparisons, highlights method-specific trade-offs, and provides practical guidance for designing and deploying attention mechanisms in long-context LLM training.", "authors": ["Tao Bu", "Qiangang Wang", "Bowen Zeng", "Hanwen Sun", "Yunpeng Huang", "Chun Cao", "Jingwei Xu"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-19", "url": "https://arxiv.org/abs/2510.17896", "pdf_url": "https://arxiv.org/pdf/2510.17896v1", "arxiv_id": "2510.17896", "doi": "10.48550/arXiv.2510.17896", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3403}
{"id": "675a2b6ad7219c3322a8c4c454577de4d26ace72635fcb9777506c63505967b4", "sources": ["arxiv", "semantic_scholar"], "title": "Extending Audio Context for Long-Form Understanding in Large Audio-Language Models", "abstract": "Large Audio-Language Models (LALMs) are often constrained by short audio context windows, even when their text backbones support long contexts, limiting long-form audio understanding. Prior work has introduced context-extension methods (e.g. YaRN) on unimodal LLMs, yet their application to LALMs remains unexplored. First, building on RoPE-based context extension, we introduce Partial YaRN, a training-free, modality-decoupled extension method that modifies only audio token positions, leaving text positions intact to preserve the base LLM's text capabilities. Second, we propose Virtual Longform Audio Training (VLAT), a training strategy that extends Partial YaRN into a training-time positional augmentation. VLAT simulates diverse audio lengths during training, enabling generalization to inputs far longer than those seen in training. Our experiments on SALMONN and Qwen2-Audio confirm that Partial YaRN outperforms the original models across wide range of settings, and VLAT provides substantial performance improvement on long audio of unseen lengths.", "authors": ["Yuatyong Chaichana", "Pittawat Taveekitworachai", "Warit Sirichotedumrong", "Potsawee Manakul", "Kunat Pipatanakul"], "categories": ["cs.CL", "cs.AI", "cs.SD", "eess.AS"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2025-10-17", "url": "https://arxiv.org/abs/2510.15231", "pdf_url": "https://arxiv.org/pdf/2510.15231v2", "arxiv_id": "2510.15231", "doi": "10.48550/arXiv.2510.15231", "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/yophis/partial-yarn", "venue": "Conference of the European Chapter of the Association for Computational Linguistics", "quality_score": 0.5224}
{"id": "e19eb194abd6f81f0b0e62ac58eb5c69f902e8dbe5662dec715f3c2b8288c8e7", "sources": ["arxiv", "semantic_scholar"], "title": "MMLongCite: A Benchmark for Evaluating Fidelity of Long-Context Vision-Language Models", "abstract": "The rapid advancement of large vision language models (LVLMs) has led to a significant expansion of their context windows. However, an extended context window does not guarantee the effective utilization of the context, posing a critical challenge for real-world applications. Current evaluations of such long-context faithfulness are predominantly focused on the text-only domain, while multimodal assessments remain limited to short contexts. To bridge this gap, we introduce MMLongCite, a comprehensive benchmark designed to evaluate the fidelity of LVLMs in long-context scenarios. MMLongCite comprises 8 distinct tasks spanning 6 context length intervals and incorporates diverse modalities, including text, images, and videos. Our evaluation of state-of-the-art LVLMs reveals their limited faithfulness in handling long multimodal contexts. Furthermore, we provide an in-depth analysis of how context length and the position of crucial content affect the faithfulness of these models.", "authors": ["Keyan Zhou", "Zecheng Tang", "Lingfeng Ming", "Guanghao Zhou", "Qiguang Chen", "Dan Qiao", "Zheming Yang", "Libo Qin", "Minghui Qiu", "Juntao Li", "Min Zhang"], "categories": ["cs.CV", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-15", "url": "https://arxiv.org/abs/2510.13276", "pdf_url": "https://arxiv.org/pdf/2510.13276v1", "arxiv_id": "2510.13276", "doi": "10.48550/arXiv.2510.13276", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3357}
{"id": "74502638f89af2a46116a2ffe8bb811cda38b9e991b9f13155578f48fbb7340b", "sources": ["arxiv", "semantic_scholar"], "title": "SpareCodeSearch: Searching for Code Context When You Have No Spare GPU", "abstract": "Retrieval-Augmented Generation (RAG) frameworks aim to enhance Code Language Models (CLMs) by including another module for retrieving relevant context to construct the input prompt. However, these retrieval modules commonly use semantic search, requiring substantial computational resources for training and hosting these embedded models, making them infeasible to integrate into lightweight applications such as in-IDE AI-based code completion. In this solution paper, we prove that using keyword-search is sufficient to retrieve relevant and useful code context inside large codebases, without the need for extensive GPU resources. The usefulness of code contexts found by our solution is demonstrated through their completion results on the Code Context Competition's benchmark, reaching 0.748 and 0.725 chRF scores on Kotlin and Python tracks, respectively.", "authors": ["Minh Nguyen"], "categories": ["cs.SE", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-14", "url": "https://arxiv.org/abs/2510.12948", "pdf_url": "https://arxiv.org/pdf/2510.12948v1", "arxiv_id": "2510.12948", "doi": "10.1109/ASEW67777.2025.00073", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2129}
{"id": "028437c3d04323ed6bec4d7d9e22f22eba6e8b527b79f248b48ff4abdc25dcb1", "sources": ["arxiv", "semantic_scholar"], "title": "Scaling Long-Horizon LLM Agent via Context-Folding", "abstract": "Large language model (LLM) agents are fundamentally constrained by context length on long-horizon tasks. We introduce Context-Folding, a framework that empowers agents to actively manage their working context. An agent can procedurally branch into a sub-trajectory to handle a subtask and then fold it upon completion, collapsing the intermediate steps while retaining a concise summary of the outcome. To make this behavior learnable, we develop an end-to-end reinforcement learning framework FoldGRPO with specific process rewards to encourage effective task decomposition and context management. On complex long-horizon tasks (Deep Research and SWE), our folding agent matches or outperforms the ReAct baselines while using an active context 10$\\times$ smaller and significantly outperforms models that rely on summarization-based context management.", "authors": ["Weiwei Sun", "Miao Lu", "Zhan Ling", "Kang Liu", "Xuesong Yao", "Yiming Yang", "Jiecao Chen"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-13", "url": "https://arxiv.org/abs/2510.11967", "pdf_url": "https://arxiv.org/pdf/2510.11967v1", "arxiv_id": "2510.11967", "doi": "10.48550/arXiv.2510.11967", "citation_count": 77, "influential_citation_count": 7, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.473}
{"id": "d5117d3c09229ee546df2bd089153ebdb8039bbb425752dd4ea035622ee9cbdc", "sources": ["arxiv", "semantic_scholar"], "title": "UltraLLaDA: Scaling the Context Length to 128K for Diffusion Large Language Models", "abstract": "Diffusion LLMs have attracted growing interest, with plenty of recent work emphasizing their great potential in various downstream tasks; yet the long-context behavior of diffusion LLMs remains largely uncharted. We present a case study of post-training techniques for extending the context window of diffusion LLMs (i.e., LLaDA) without retraining from scratch. We show that a simple modification to the standard Rotary Positional Embeddings (RoPE) extension effectively accommodates the probabilistic modeling inherent in the diffusion process, enabling stable scaling to longer context ranges. We further compare masking strategies used during post-training and analyze their impact on optimization stability and long-range recall. Instantiating these insights, we introduce UltraLLaDA, a diffusion LLM with a 128K-token context window that, in our empirical evaluation on long-context tasks, significantly outperforms training-free baselines. Our experimental results highlight the special positional extension as a key lever for scaling diffusion LLMs to extended contexts and offer practical guidance for practitioners seeking 128K-scale context via efficient post-training.", "authors": ["Guangxin He", "Shen Nie", "Fengqi Zhu", "Yuankang Zhao", "Tianyi Bai", "Ran Yan", "Jie Fu", "Chongxuan Li", "Binhang Yuan"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-12", "url": "https://arxiv.org/abs/2510.10481", "pdf_url": "https://arxiv.org/pdf/2510.10481v1", "arxiv_id": "2510.10481", "doi": "10.48550/arXiv.2510.10481", "citation_count": 11, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3323}
{"id": "c5e08411bb0cbb305e0357714c1f41dbe7f2ac0e7f5ba3636746f422d2164283", "sources": ["arxiv", "semantic_scholar"], "title": "Beyond Turn Limits: Training Deep Search Agents with Dynamic Context Window", "abstract": "While recent advances in reasoning models have demonstrated cognitive behaviors through reinforcement learning, existing approaches struggle to invoke deep reasoning capabilities in multi-turn agents with long-horizon interactions. We propose DeepMiner, a novel framework that elicits such abilities by introducing high-difficulty training tasks and dynamic context window. DeepMiner presents a reverse construction method to generate complex but verifiable question-answer pairs from authentic web sources, which ensures the challenge and reliability of training data while injecting cognitive capabilities into multi-turn reasoning scenarios. We further design an elegant yet effective dynamic context management strategy for both training and inference, utilizing sliding window mechanisms while eliminating the dependency on external summarization models, thereby efficiently empowering the model to handle continuously expanding long-horizon contexts. Through reinforcement learning on Qwen3-32B, we develop DeepMiner-32B, which achieves substantial performance improvements across multiple search agent benchmarks. DeepMiner attains 33.5% accuracy on BrowseComp-en, surpassing the previous best open-source agent by almost 20 percentage points, and demonstrates consistent improvements on BrowseComp-zh, XBench-DeepSearch, and GAIA. Notably, our dynamic context management enables sustained interactions of nearly 100 turns within standard 32k context length, effectively addressing the context limitations that constrain existing multi-turn interaction systems.", "authors": ["Qiaoyu Tang", "Hao Xiang", "Le Yu", "Bowen Yu", "Yaojie Lu", "Xianpei Han", "Le Sun", "WenJuan Zhang", "Pengbo Wang", "Shixuan Liu", "Zhenru Zhang", "Jianhong Tu", "Hongyu Lin", "Junyang Lin"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-09", "url": "https://arxiv.org/abs/2510.08276", "pdf_url": "https://arxiv.org/pdf/2510.08276v1", "arxiv_id": "2510.08276", "doi": "10.48550/arXiv.2510.08276", "citation_count": 17, "influential_citation_count": 3, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5082}
{"id": "e8e89c1fac2e798d5bc5fb97d29865f9a0e3f0359412cae2d7051868bcb59ae2", "sources": ["arxiv", "semantic_scholar"], "title": "Scaling LLM Multi-turn RL with End-to-end Summarization-based Context Management", "abstract": "We study reinforcement learning (RL) fine-tuning of large language model (LLM) agents for long-horizon multi-turn tool use, where context length quickly becomes a fundamental bottleneck. Existing RL pipelines can suffer from degraded instruction following, excessive rollout costs, and most importantly, strict context limits. To address these challenges, we introduce summarization-based context management to training. In specific, it periodically compresses the tool using history by LLM-generated summaries that retain task-relevant information to keep a compact context while enabling the agent to scale beyond the fixed context window. Building on this formulation, we derive a policy gradient representation that seamlessly enables standard LLM RL infrastructures to optimize both tool-use behaviors as well as summarization strategies in an end-to-end fashion. We instantiate this framework with \\underline{SU}mmarization augmented \\underline{P}olicy \\underline{O}ptimization (\\texttt{SUPO}), an LLM RL algorithm that enables long-horizon training beyond a fixed context limit. Experiments on interactive function calling and searching tasks demonstrate that \\texttt{SUPO} significantly improves the success rate while maintaining the same or even lower working context length compared to baselines. We also demonstrate that for complex searching tasks, \\texttt{SUPO} can further improve the evaluation performance when scaling test-time maximum round of summarization beyond that of training time. Our results establish summarization-based context management as a principled and scalable approach for training RL agents beyond a fixed context length limit.", "authors": ["Miao Lu", "Weiwei Sun", "Weihua Du", "Zhan Ling", "Xuesong Yao", "Kang Liu", "Jiecao Chen"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-08", "url": "https://arxiv.org/abs/2510.06727", "pdf_url": "https://arxiv.org/pdf/2510.06727v1", "arxiv_id": "2510.06727", "doi": "10.48550/arXiv.2510.06727", "citation_count": 23, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3451}
{"id": "15c3658b334d5704dc73b338b5a2e9fe83e577f58c904a19e1f9526200964cad", "sources": ["arxiv", "semantic_scholar"], "title": "Haystack Engineering: Context Engineering for Heterogeneous and Agentic Long-Context Evaluation", "abstract": "Modern long-context large language models (LLMs) perform well on synthetic \"needle-in-a-haystack\" (NIAH) benchmarks, but such tests overlook how noisy contexts arise from biased retrieval and agentic workflows. We argue that haystack engineering is necessary to construct noisy long contexts that faithfully capture key real-world factors -- distraction from heterogeneous biased retrievers and cascading errors in agentic workflows -- to test models' long-context robustness. We instantiate it through HaystackCraft, a new NIAH benchmark built on the full English Wikipedia hyperlink network with multi-hop questions. HaystackCraft evaluates how heterogeneous retrieval strategies (e.g., sparse, dense, hybrid, and graph-based) affect distractor composition, haystack ordering, and downstream LLM performance. HaystackCraft further extends NIAH to dynamic, LLM-dependent settings that simulate agentic operations, where models refine queries, reflect on their past reasonings, and decide when to stop. Experiments with 15 long-context models show that (1) while stronger dense retrievers can introduce more challenging distractors, graph-based reranking simultaneously improves retrieval effectiveness and mitigates more harmful distractors; (2) in agentic tests, even advanced models like Gemini 2.5 Pro and GPT-5 suffer cascading failures from self-generated distractors or struggle to perform early stops. These results highlight persistent challenges in agentic long-context reasoning and establish HaystackCraft as a valuable testbed for future progress.", "authors": ["Mufei Li", "Dongqi Fu", "Limei Wang", "Si Zhang", "Hanqing Zeng", "Kaan Sancak", "Ruizhong Qiu", "Haoyu Wang", "Xiaoxin He", "Xavier Bresson", "Yinglong Xia", "Chonglin Sun", "Pan Li"], "categories": ["cs.CL", "cs.AI", "cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-08", "url": "https://arxiv.org/abs/2510.07414", "pdf_url": "https://arxiv.org/pdf/2510.07414v2", "arxiv_id": "2510.07414", "doi": "10.48550/arXiv.2510.07414", "citation_count": 3, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Graph-COM/HaystackCraft", "venue": "arXiv.org", "quality_score": 0.5065}
{"id": "4e9f11ac23099e03ed066a39f4186a6e23ef89d57a7588b85ef4258b3ef3a992", "sources": ["arxiv", "semantic_scholar"], "title": "Relative Positioning Based Code Chunking Method For Rich Context Retrieval In Repository Level Code Completion Task With Code Language Model", "abstract": "Code completion can help developers improve efficiency and ease the development lifecycle. Although code completion is available in modern integrated development environments (IDEs), research lacks in determining what makes a good context for code completion based on the information available to the IDEs for the large language models (LLMs) to perform better. In this paper, we describe an effective context collection strategy to assist the LLMs in performing better at code completion tasks. The key idea of our strategy is to preprocess the repository into smaller code chunks and later use syntactic and semantic similarity-based code chunk retrieval with relative positioning. We found that code chunking and relative positioning of the chunks in the final context improve the performance of code completion tasks.", "authors": ["Imranur Rahman", "Md Rayhanur Rahman"], "categories": ["cs.SE", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-07", "url": "https://arxiv.org/abs/2510.08610", "pdf_url": "https://arxiv.org/pdf/2510.08610v1", "arxiv_id": "2510.08610", "doi": "10.1109/ASEW67777.2025.00077", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2078}
{"id": "4436b9145077f0bfe74f1a664faee6a0b4d820cfa95d00debd06bb2e6356a853", "sources": ["arxiv", "semantic_scholar"], "title": "FinLFQA: Evaluating Attributed Text Generation of LLMs in Financial Long-Form Question Answering", "abstract": "Large Language Models (LLMs) frequently hallucinate to long-form questions, producing plausible yet factually incorrect answers. A common mitigation strategy is to provide attribution to LLM outputs. However, existing benchmarks primarily focus on simple attribution that retrieves supporting textual evidence as references. We argue that in real-world scenarios such as financial applications, attribution goes beyond reference retrieval. We introduce FinLFQA, a benchmark designed to evaluate the ability of LLMs to generate long-form answers to complex financial questions with reliable and nuanced attributions. FinLFQA evaluates three critical aspects of attribution through human annotations: (1) supporting evidence extracted from financial reports, (2) intermediate numerical reasoning steps, and (3) domain-specific financial knowledge that informs the reasoning process. We further provide an automatic evaluation framework covering both answer quality and attribution quality. Through extensive experiments on eight LLMs across multiple attribution-generation paradigms, we find that fine-grained metrics are important to distinguish model capabilities, that end-to-end generation achieves comparable performance to post-hoc approaches, and that iterative refinement only helps when guided by external feedback.", "authors": ["Yitao Long", "Tiansheng Hu", "Yilun Zhao", "Arman Cohan", "Chen Zhao"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-07", "url": "https://arxiv.org/abs/2510.06426", "pdf_url": "https://arxiv.org/pdf/2510.06426v1", "arxiv_id": "2510.06426", "doi": "10.18653/v1/2025.findings-emnlp.908", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.3266}
{"id": "27fe1c2dc8125e5742cc5fe8172b9afa05a1ca98ba4359a0d184a45358dc50b8", "sources": ["arxiv", "semantic_scholar"], "title": "Revisiting Long-context Modeling from Context Denoising Perspective", "abstract": "Long-context models (LCMs) have demonstrated great potential in processing long sequences, facilitating many real-world applications. The success of LCMs can be attributed to their ability to locate implicit critical information within the context for further prediction. However, recent research reveals that LCMs are often susceptible to contextual noise, i.e., irrelevant tokens, that can mislead model attention. In this paper, we conduct a fine-grained analysis of the context noise and propose an effective metric, the Integrated Gradient (IG) score, to detect and quantify the noise information within the context. Our findings reveal that even simple mitigation of detected context noise can substantially boost the model's attention on critical tokens and benefit subsequent predictions. Building on this insight, we propose Context Denoising Training (CDT), a straightforward yet effective training strategy that improves attention on critical tokens while reinforcing their influence on model predictions. Extensive experiments across four tasks, under both context window scaling and long-context alignment settings, demonstrate the superiority of CDT. Notably, when trained with CDT, an open-source 8B model can achieve performance (50.92) comparable to GPT-4o (51.00).", "authors": ["Zecheng Tang", "Baibei Ji", "Juntao Li", "Lijun Wu", "Haijia Gui", "Min Zhang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-07", "url": "https://arxiv.org/abs/2510.05862", "pdf_url": "https://arxiv.org/pdf/2510.05862v2", "arxiv_id": "2510.05862", "doi": "10.48550/arXiv.2510.05862", "citation_count": 3, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5047}
{"id": "0515e6bfad18cdf5cde533d669f1596d21f24037f07513baeb824f99b2860fc5", "sources": ["arxiv", "semantic_scholar"], "title": "Challenge on Optimization of Context Collection for Code Completion", "abstract": "The rapid advancement of workflows and methods for software engineering using AI emphasizes the need for a systematic evaluation and analysis of their ability to leverage information from entire projects, particularly in large code bases. In this challenge on optimization of context collection for code completion, organized by JetBrains in collaboration with Mistral AI as part of the ASE 2025 conference, participants developed efficient mechanisms for collecting context from source code repositories to improve fill-in-the-middle code completions for Python and Kotlin. We constructed a large dataset of real-world code in these two programming languages using permissively licensed open-source projects. The submissions were evaluated based on their ability to maximize completion quality for multiple state-of-the-art neural models using the chrF metric. During the public phase of the competition, nineteen teams submitted solutions to the Python track and eight teams submitted solutions to the Kotlin track. In the private phase, six teams competed, of which five submitted papers to the workshop.", "authors": ["Dmitry Ustalov", "Egor Bogomolov", "Alexander Bezzubov", "Yaroslav Golubev", "Evgeniy Glukhov", "Georgii Levtsov", "Vladimir Kovalenko"], "categories": ["cs.SE", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-05", "url": "https://arxiv.org/abs/2510.04349", "pdf_url": "https://arxiv.org/pdf/2510.04349v1", "arxiv_id": "2510.04349", "doi": "10.1109/ASEW67777.2025.00072", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": null, "quality_score": 0.3832}
{"id": "faaf68314e5f79cf9d59bcb5cbc44bf18c0b1135e6b195a7baee0985782243bf", "sources": ["arxiv", "semantic_scholar"], "title": "No Tokens Wasted: Leveraging Long Context in Biomedical Vision-Language Models", "abstract": "Embedding vision-language models (VLMs) are typically pretrained with short text windows (<77 tokens), which forces the truncation of long-format captions. Yet, the distribution of biomedical captions from large-scale open source literature reveals that a huge portion of captions far exceed 77 tokens. To this end, we investigate the impact of pretraining on long-format biomedical captions by extending the context length of text encoders in VLMs. We find that longer context (thus, enabling additional supervision provided in long-format captions) correlates with better retrieval and classification performance. Given this finding, we introduce BIOMEDICA-LongCAP, a dataset of 1M image-caption pairs enriched with context-aware descriptions from full-text articles, providing longer and additional textual supervision. Using BIOMEDICA-LongCAP, we train BMC-LongCLIP, a long-context biomedical VLM with a text encoder supporting windows of up to 512 tokens. Our model extends context capacity by 6.6x, reducing token waste from 55% to just 2.2%. On long-caption retrieval benchmarks, BMC-LongCLIP achieves up to +30% absolute gains in Recall@1 and +2% average improvements in classification, while also converging faster than short-context. Our results demonstrate that long-context modeling is a promising direction for advancing biomedical VLMs.", "authors": ["Min Woo Sun", "Alejandro Lozano", "Javier Gamazo Tejero", "Vishwesh Nath", "Xiao Xiao Sun", "James Burgess", "Yuhui Zhang", "Kun Yuan", "Robert Tibshirani", "Sean Huver", "Serena Yeung-Levy"], "categories": ["cs.CV", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-04", "url": "https://arxiv.org/abs/2510.03978", "pdf_url": "https://arxiv.org/pdf/2510.03978v1", "arxiv_id": "2510.03978", "doi": "10.48550/arXiv.2510.03978", "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4994}
{"id": "f621bb0a541c8eb6b0a45b7f20787a6f4ad6a906e5e6edaaf4411ac7b0689bd2", "sources": ["arxiv", "semantic_scholar"], "title": "ACON: Optimizing Context Compression for Long-horizon LLM Agents", "abstract": "Large language models (LLMs) are increasingly deployed as agents in dynamic real-world environments, where success depends on maintaining precise records of actions and observations. However, the resulting unbounded context growth in long-horizon agentic tasks makes two critical bottlenecks: prohibitive inference memory costs and reasoning degradation due to irrelevant information. Existing compression methods fail to fully address this, often relying on brittle heuristics or requiring parameter updates impractical for proprietary or large-scale LLMs. We introduce Agent Context Optimization (ACON), a unified framework that optimally compresses both observations and history into concise, informative representations. Distinct from prior works, ACON employs an optimization in natural language space: it iteratively refines compression guidelines based on failure analysis of the agent, ensuring critical state information is preserved without model fine-tuning. To further minimize computational overhead, we distill the optimized compressor into smaller models. Experiments on AppWorld, OfficeBench, and Multi-objective QA demonstrate that ACON reduces peak token usage by 26-54% while improving task success over existing compression baselines. Notably, it enables smaller LMs to function effectively as long-horizon agents, achieving up to 46% performance improvement by mitigating context distraction. Our code is available at https://github.com/microsoft/acon.", "authors": ["Minki Kang", "Wei-Ning Chen", "Dongge Han", "Huseyin A. Inan", "Lukas Wutschitz", "Yanzhi Chen", "Robert Sim", "Saravan Rajmohan"], "categories": ["cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-10-01", "url": "https://arxiv.org/abs/2510.00615", "pdf_url": "https://arxiv.org/pdf/2510.00615v3", "arxiv_id": "2510.00615", "doi": "10.48550/arXiv.2510.00615", "citation_count": 51, "influential_citation_count": 4, "has_code": true, "code_url": "https://github.com/microsoft/acon", "venue": "arXiv.org", "quality_score": 0.4941}
{"id": "3a0601943f9e74aa2fba1673bd923bdea741727902c88d3766055c7a8918e01f", "sources": ["arxiv", "semantic_scholar"], "title": "SparseServe: Unlocking Parallelism for Dynamic Sparse Attention in Long-Context LLM Serving", "abstract": "Serving long-context LLMs is costly because attention computation grows linearly with context length. Dynamic sparse attention algorithms (DSAs) mitigate this by attending only to the key-value (KV) cache of critical tokens. However, with DSAs, the main performance bottleneck shifts from HBM bandwidth to HBM capacity: KV caches for unselected tokens must remain in HBM for low-latency decoding, constraining parallel batch size and stalling further throughput gains. Offloading these underutilized KV caches to DRAM could free HBM capacity, allowing larger parallel batch sizes. Yet, achieving such hierarchical HBM-DRAM storage raises new challenges, including fragmented KV cache access, HBM cache contention, and high HBM demands of hybrid batching, that remain unresolved in prior work. This paper proposes SparseServe, an LLM serving system that unlocks the parallel potential of DSAs through efficient hierarchical HBM-DRAM management. SparseServe introduces three key innovations to address the challenges mentioned above: (1) fragmentation-aware KV cache transfer, which accelerates HBM-DRAM data movement through GPU-direct loading (FlashH2D) and CPU-assisted saving (FlashD2H); (2) working-set-aware batch size control that adjusts batch sizes based on real-time working set estimation to minimize HBM cache thrashing; (3) layer-segmented prefill that bounds HBM use during prefill to a single layer, enabling efficient execution even for long prompts. Extensive experimental results demonstrate that SparseServe achieves up to 9.26x lower mean time-to-first-token (TTFT) latency and up to 3.14x higher token generation throughput compared to state-of-the-art LLM serving systems.", "authors": ["Qihui Zhou", "Peiqi Yin", "Pengfei Zuo", "James Cheng"], "categories": ["cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-29", "url": "https://arxiv.org/abs/2509.24626", "pdf_url": "https://arxiv.org/pdf/2509.24626v1", "arxiv_id": "2509.24626", "doi": "10.48550/arXiv.2509.24626", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3174}
{"id": "ee01236eb5ac3fd5e5c35d38f01d7cbaaadbbe3dbfce95fa800ef8268e30fea1", "sources": ["arxiv", "semantic_scholar"], "title": "The Impact of Role Design in In-Context Learning for Large Language Models", "abstract": "In-context learning (ICL) enables Large Language Models (LLMs) to generate predictions based on prompts without additional fine-tuning. While prompt engineering has been widely studied, the impact of role design within prompts remains underexplored. This study examines the influence of role configurations in zero-shot and few-shot learning scenarios using GPT-3.5 and GPT-4o from OpenAI and Llama2-7b and Llama2-13b from Meta. We evaluate the models' performance across datasets, focusing on tasks like sentiment analysis, text classification, question answering, and math reasoning. Our findings suggest the potential of role-based prompt structuring to enhance LLM performance.", "authors": ["Hamidreza Rouzegar", "Masoud Makrehchi"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-27", "url": "https://arxiv.org/abs/2509.23501", "pdf_url": "https://arxiv.org/pdf/2509.23501v1", "arxiv_id": "2509.23501", "doi": "10.48550/arXiv.2509.23501", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/hrouzegar/Role_Based-In-Context-Learning", "venue": "arXiv.org", "quality_score": 0.487}
{"id": "74f835756e00ec94bcae543415838584a995232546f918c5270087df83ddb783", "sources": ["arxiv", "semantic_scholar"], "title": "Graph of Agents: Principled Long Context Modeling by Emergent Multi-Agent Collaboration", "abstract": "As a model-agnostic approach to long context modeling, multi-agent systems can process inputs longer than a large language model's context window without retraining or architectural modifications. However, their performance often heavily relies on hand-crafted multi-agent collaboration strategies and prompt engineering, which limit generalizability. In this work, we introduce a principled framework that formalizes the model-agnostic long context modeling problem as a compression problem, yielding an information-theoretic compression objective. Building on this framework, we propose Graph of Agents (GoA), which dynamically constructs an input-dependent collaboration structure that maximizes this objective. For Llama 3.1 8B and Qwen3 8B across six document question answering benchmarks, GoA improves the average $F_1$ score of retrieval-augmented generation by 5.7\\% and a strong multi-agent baseline using a fixed collaboration structure by 16.35\\%, respectively. Even with only a 2K context window, GoA surpasses the 128K context window Llama 3.1 8B on LongBench, showing a dramatic increase in effective context length. Our source code is available at https://github.com/tjoo512/graph-of-agents.", "authors": ["Taejong Joo", "Shu Ishida", "Ivan Sosnovik", "Bryan Lim", "Sahand Rezaei-Shoshtari", "Adam Gaier", "Robert Giaquinto"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-26", "url": "https://arxiv.org/abs/2509.21848", "pdf_url": "https://arxiv.org/pdf/2509.21848v1", "arxiv_id": "2509.21848", "doi": "10.48550/arXiv.2509.21848", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/tjoo512/graph-of-agents", "venue": "arXiv.org", "quality_score": 0.4852}
{"id": "8effe7c8fc809137de508b60be08f6df305beedaec7ecdb54162d72487ea25fc", "sources": ["arxiv", "semantic_scholar"], "title": "InfiniPipe: Elastic Pipeline Parallelism for Efficient Variable-Length Long-Context LLM Training", "abstract": "Long context training is crucial for LLM's context extension. Existing schemes, such as sequence parallelism, incur substantial communication overhead. Pipeline parallelism (PP) reduces this cost, but its effectiveness hinges on partitioning granularity. Batch-level PP employing sequence packing exhibits high memory consumption in long-context scenarios, whereas token-level PP splitting sequences into slices alleviates memory overhead but may incur hardware under-utilization. Moreover, the skewed distribution of sequence length in real-world datasets renders monolithic and static granularity PP's sub-optimal performance. In this paper, we propose 1) \\textit{Elastic Pipeline Parallelism} (EPP) that orchestrates token-level PP and batch-level PP to adapt to resource and workload heterogeneity, and 2) \\textit{Stage-Aware Chunk-Level Adaptive Checkpointing} that efficiently integrates gradient checkpointing with EPP. Comprehensive experiments demonstrate that InfiniPipe achieves a 1.69x speedup over state-of-the-art systems. Our code is open-sourced at https://github.com/wsjdsg/InfiniPipe-code.git.", "authors": ["Shiju Wang", "Yujie Wang", "Ao Sun", "Fangcheng Fu", "Zijian Zhu", "Bin Cui", "Xu Han", "Kaisheng Ma"], "categories": ["cs.DC", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-25", "url": "https://arxiv.org/abs/2509.21275", "pdf_url": "https://arxiv.org/pdf/2509.21275v4", "arxiv_id": "2509.21275", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/wsjdsg/InfiniPipe-code.git", "venue": null, "quality_score": 0.3697}
{"id": "793293c4df76a6ccaa97b98fe6b9e18f2de67267134acfe7dccfe7d73c7bada5", "sources": ["arxiv", "semantic_scholar"], "title": "Steering Multimodal Large Language Models Decoding for Context-Aware Safety", "abstract": "Multimodal Large Language Models (MLLMs) are increasingly deployed in real-world applications, yet their ability to make context-aware safety decisions remains limited. Existing methods often fail to balance oversensitivity (unjustified refusals of benign queries) and undersensitivity (missed detection of visually grounded risks), leaving a persistent gap in safety alignment. To address this issue, we introduce Safety-aware Contrastive Decoding (SafeCoDe), a lightweight and model-agnostic decoding framework that dynamically adjusts token generation based on multimodal context. SafeCoDe operates in two stages: (1) a contrastive decoding mechanism that highlights tokens sensitive to visual context by contrasting real and Gaussian-noised images, and (2) a global-aware token modulation strategy that integrates scene-level reasoning with token-level adjustment to adapt refusals according to the predicted safety verdict. Extensive experiments across diverse MLLM architectures and safety benchmarks, covering undersensitivity, oversensitivity, and general safety evaluations, show that SafeCoDe consistently improves context-sensitive refusal behaviors while preserving model helpfulness.", "authors": ["Zheyuan Liu", "Zhangchen Xu", "Guangyao Dou", "Xiangchi Yuan", "Zhaoxuan Tan", "Radha Poovendran", "Meng Jiang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-23", "url": "https://arxiv.org/abs/2509.19212", "pdf_url": "https://arxiv.org/pdf/2509.19212v1", "arxiv_id": "2509.19212", "doi": "10.48550/arXiv.2509.19212", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3105}
{"id": "889f411227f99fd37cb2edef9de51005a97324baae67d31898f3e7df9e1b6a0d", "sources": ["arxiv", "semantic_scholar"], "title": "CompLLM: Compression for Long Context Q&A", "abstract": "Large Language Models (LLMs) face significant computational challenges when processing long contexts due to the quadratic complexity of self-attention. While soft context compression methods, which map input text to smaller latent representations, have shown promise, their real-world adoption is limited. Existing techniques typically compress the context as a single unit, which leads to quadratic compression complexity and an inability to reuse computations across queries with overlapping contexts. In this work, we introduce CompLLM, a soft compression technique designed for practical deployment. Instead of processing the context holistically, CompLLM divides it into segments and compresses each one independently. This simple design choice yields three critical properties: efficiency, as the compression step scales linearly with the context length; scalability, enabling models trained on short sequences (e.g., 1k tokens) to generalize to contexts of 100k tokens; and reusability, allowing compressed segments to be cached and reused across different queries. Our experiments show that with a 2x compression rate, at high context lengths CompLLM speeds up Time To First Token (TTFT) by up to 4x and reduces the KV cache size by 50%. Furthermore, CompLLM achieves performance comparable to that obtained with the uncompressed context, and even surpasses it on very long sequences, demonstrating its effectiveness and practical utility.", "authors": ["Gabriele Berton", "Jayakrishnan Unnikrishnan", "Son Tran", "Mubarak Shah"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-23", "url": "https://arxiv.org/abs/2509.19228", "pdf_url": "https://arxiv.org/pdf/2509.19228v1", "arxiv_id": "2509.19228", "doi": "10.48550/arXiv.2509.19228", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3105}
{"id": "e66fadaa916c8d8d13ab9a0f4b5d1935ec3f6b2b383e96683734b6e5b93931b8", "sources": ["arxiv", "semantic_scholar"], "title": "Towards Adaptive Context Management for Intelligent Conversational Question Answering", "abstract": "This particular paper introduces an Adaptive Context Management (ACM) framework for the Conversational Question Answering (ConvQA) systems. The key objective of the ACM framework is to optimize the use of the conversation history by dynamically managing context for maximizing the relevant information provided to a ConvQA model within its token limit. Our approach incorporates a Context Manager (CM) Module, a Summarization (SM) Module, and an Entity Extraction (EE) Module in a bid to handle the conversation history efficaciously. The CM Module dynamically adjusts the context size, thereby preserving the most relevant and recent information within a model's token limit. The SM Module summarizes the older parts of the conversation history via a sliding window. When the summarization window exceeds its limit, the EE Module identifies and retains key entities from the oldest conversation turns. Experimental results demonstrate the effectiveness of our envisaged framework in generating accurate and contextually appropriate responses, thereby highlighting the potential of the ACM framework to enhance the robustness and scalability of the ConvQA systems.", "authors": ["Manoj Madushanka Perera", "Adnan Mahmood", "Kasun Eranda Wijethilake", "Quan Z. Sheng"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-22", "url": "https://arxiv.org/abs/2509.17829", "pdf_url": "https://arxiv.org/pdf/2509.17829v1", "arxiv_id": "2509.17829", "doi": "10.1007/978-981-96-0847-8_25", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Advanced Data Mining and Applications", "quality_score": 0.3094}
{"id": "a7f1def78ecc405403d40dcf8233542405a2506f50c7d0d2359a4e012a40e97f", "sources": ["arxiv", "semantic_scholar"], "title": "Context Is What You Need: The Maximum Effective Context Window for Real World Limits of LLMs", "abstract": "Large language model (LLM) providers boast big numbers for maximum context window sizes. To test the real world use of context windows, we 1) define a concept of maximum effective context window, 2) formulate a testing method of a context window's effectiveness over various sizes and problem types, and 3) create a standardized way to compare model efficacy for increasingly larger context window sizes to find the point of failure. We collected hundreds of thousands of data points across several models and found significant differences between reported Maximum Context Window (MCW) size and Maximum Effective Context Window (MECW) size. Our findings show that the MECW is, not only, drastically different from the MCW but also shifts based on the problem type. A few top of the line models in our test group failed with as little as 100 tokens in context; most had severe degradation in accuracy by 1000 tokens in context. All models fell far short of their Maximum Context Window by as much as 99 percent. Our data reveals the Maximum Effective Context Window shifts based on the type of problem provided, offering clear and actionable insights into how to improve model accuracy and decrease model hallucination rates.", "authors": ["Norman Paulsen"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-21", "url": "https://arxiv.org/abs/2509.21361", "pdf_url": "https://arxiv.org/pdf/2509.21361v2", "arxiv_id": "2509.21361", "doi": "10.54364/AAIML.2026.61268", "citation_count": 11, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Advances in Artificial Intelligence and Machine Learning", "quality_score": 0.3082}
{"id": "b3e10f2d96322ccf22fa5ea1d6e53db94e062a12d4c8d10812c0ea4e0bbba692", "sources": ["arxiv", "semantic_scholar"], "title": "LiteLong: Resource-Efficient Long-Context Data Synthesis for LLMs", "abstract": "High-quality long-context data is essential for training large language models (LLMs) capable of processing extensive documents, yet existing synthesis approaches using relevance-based aggregation face challenges of computational efficiency. We present LiteLong, a resource-efficient method for synthesizing long-context data through structured topic organization and multi-agent debate. Our approach leverages the BISAC book classification system to provide a comprehensive hierarchical topic organization, and then employs a debate mechanism with multiple LLMs to generate diverse, high-quality topics within this structure. For each topic, we use lightweight BM25 retrieval to obtain relevant documents and concatenate them into 128K-token training samples. Experiments on HELMET and Ruler benchmarks demonstrate that LiteLong achieves competitive long-context performance and can seamlessly integrate with other long-dependency enhancement methods. LiteLong makes high-quality long-context data synthesis more accessible by reducing both computational and data engineering costs, facilitating further research in long-context language training.", "authors": ["Junlong Jia", "Xing Wu", "Chaochen Gao", "Ziyang Chen", "Zijia Lin", "Zhongzhi Li", "Weinong Wang", "Haotian Xu", "Donghui Jin", "Debing Zhang", "Binghui Guo"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-19", "url": "https://arxiv.org/abs/2509.15568", "pdf_url": "https://arxiv.org/pdf/2509.15568v1", "arxiv_id": "2509.15568", "doi": "10.48550/arXiv.2509.15568", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "AAAI Conference on Artificial Intelligence", "quality_score": 0.3059}
{"id": "ccdd2e2551e1b3db70d49fcd1d7b8604d99f3a6c54d1232d643fc861b22d6d33", "sources": ["arxiv", "semantic_scholar"], "title": "Q-ROAR: Outlier-Aware Rescaling for RoPE Position Interpolation in Quantized Long-Context LLMs", "abstract": "Extending LLM context windows is crucial for long range tasks. RoPE-based position interpolation (PI) methods like linear and frequency-aware scaling extend input lengths without retraining, while post-training quantization (PTQ) enables practical deployment. We show that combining PI with PTQ degrades accuracy due to coupled effects long context aliasing, dynamic range dilation, axis grid anisotropy, and outlier shifting that induce position-dependent logit noise. We provide the first systematic analysis of PI plus PTQ and introduce two diagnostics: Interpolation Pressure (per-band phase scaling sensitivity) and Tail Inflation Ratios (outlier shift from short to long contexts). To address this, we propose Q-ROAR, a RoPE-aware, weight-only stabilization that groups RoPE dimensions into a few frequency bands and performs a small search over per-band scales for W_Q,W_K, with an optional symmetric variant to preserve logit scale. The diagnostics guided search uses a tiny long-context dev set and requires no fine-tuning, kernel, or architecture changes. Empirically, Q-ROAR recovers up to 0.7% accuracy on standard tasks and reduces GovReport perplexity by more than 10%, while preserving short-context performance and compatibility with existing inference stacks.", "authors": ["Ye Qiao", "Sitao Huang"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-17", "url": "https://arxiv.org/abs/2509.14391", "pdf_url": "https://arxiv.org/pdf/2509.14391v1", "arxiv_id": "2509.14391", "doi": "10.48550/arXiv.2509.14391", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3036}
{"id": "71fcadc494f606ba5546b64ef626c0f61385e3e9a2b2f8b91696d2a2992ec4e9", "sources": ["arxiv", "semantic_scholar"], "title": "HoPE: Hyperbolic Rotary Positional Encoding for Stable Long-Range Dependency Modeling in Large Language Models", "abstract": "Positional encoding mechanisms enable Transformers to model sequential structure and long-range dependencies in text. While absolute positional encodings struggle with extrapolation to longer sequences due to fixed positional representations, and relative approaches like Alibi exhibit performance degradation on extremely long contexts, the widely-used Rotary Positional Encoding (RoPE) introduces oscillatory attention patterns that hinder stable long-distance dependency modelling. We address these limitations through a geometric reformulation of positional encoding. Drawing inspiration from Lorentz transformations in hyperbolic geometry, we propose Hyperbolic Rotary Positional Encoding (HoPE), which leverages hyperbolic functions to implement Lorentz rotations on token representations. Theoretical analysis demonstrates that RoPE is a special case of our generalized formulation. HoPE fundamentally resolves RoPE's slation issues by enforcing monotonic decay of attention weights with increasing token distances. Extensive experimental results, including perplexity evaluations under several extended sequence benchmarks, show that HoPE consistently exceeds existing positional encoding methods. These findings underscore HoPE's enhanced capacity for representing and generalizing long-range dependencies. Data and code will be available.", "authors": ["Chang Dai", "Hongyu Shan", "Mingyang Song", "Di Liang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-09-05", "url": "https://arxiv.org/abs/2509.05218", "pdf_url": "https://arxiv.org/pdf/2509.05218v2", "arxiv_id": "2509.05218", "doi": "10.48550/arXiv.2509.05218", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2899}
{"id": "8cb4bd6b8dfe654fd49cd84edefdc0d23475b7bb3089f14e4d43fc55605a9e9b", "sources": ["arxiv", "semantic_scholar"], "title": "The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Management", "abstract": "Large Language Model (LLM)-based agents solve complex tasks through iterative reasoning, exploration, and tool-use, a process that can result in long, expensive context histories. While state-of-the-art Software Engineering (SE) agents like OpenHands or Cursor use LLM-based summarization to tackle this issue, it is unclear whether the increased complexity offers tangible performance benefits compared to simply omitting older observations. We present a systematic comparison of these approaches within SWE-agent on SWE-bench Verified across five diverse model configurations. Moreover, we show initial evidence of our findings generalizing to the OpenHands agent scaffold. We find that a simple environment observation masking strategy halves cost relative to the raw agent while matching, and sometimes slightly exceeding, the solve rate of LLM summarization. Additionally, we introduce a novel hybrid approach that further reduces costs by 7% and 11% compared to just observation masking or LLM summarization, respectively. Our findings raise concerns regarding the trend towards pure LLM summarization and provide initial evidence of untapped cost reductions by pushing the efficiency-effectiveness frontier. We release code and data for reproducibility.", "authors": ["Tobias Lindenbauer", "Igor Slinko", "Ludwig Felder", "Egor Bogomolov", "Yaroslav Zharov"], "categories": ["cs.SE", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-29", "url": "https://arxiv.org/abs/2508.21433", "pdf_url": "https://arxiv.org/pdf/2508.21433v3", "arxiv_id": "2508.21433", "doi": "10.48550/arXiv.2508.21433", "citation_count": 16, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3076}
{"id": "183102849cdc0c5046218694e65522308a6db59bd024b3da96c6dcf63341b280", "sources": ["arxiv", "semantic_scholar"], "title": "Joint Enhancement of Relational Reasoning for Long-Context LLMs", "abstract": "Despite significant progress, large language models (LLMs) still struggle with long contexts due to memory limitations and their inability to tackle complex and long-context tasks. Additionally, LLMs often suffer from a lack of transparency and are prone to producing hallucinations. To address these challenges, we propose \\textbf{JERR}, a novel framework designed to enhance long-context comprehension via graph-based reasoning in LLMs. JERR integrates three key components: synopsis extraction, graph construction, and relational reasoning. First, synopsis is extracted by chunking text strategically, allowing the model to summarize and understand information more efficiently. Second, we build a directed acyclic graph (DAG) to resolve redundancy, ensuring logical consistency and clarity. Finally, we incorporate Monte Carlo Tree Search (MCTS) to help the model navigate complex reasoning paths, ensuring more accurate and interpretable outputs. This framework provides a novel solution that enables LLMs to handle extended contexts and complex reasoning tasks with improved reliability and transparency. Experimental results show that JERR consistently outperforms all baselines on the ROUGE and F1 metrics, achieving the highest scores on the LLM-Rater evaluation.", "authors": ["Zhirui Chen", "Wei Shen", "Jiashui Huang", "Ling Shao"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-28", "url": "https://arxiv.org/abs/2508.20351", "pdf_url": "https://arxiv.org/pdf/2508.20351v1", "arxiv_id": "2508.20351", "doi": "10.48550/arXiv.2508.20351", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.2807}
{"id": "94e109f569823be75fe4531c6ae74da30aa26de5b59e246207d4555b0720c153", "sources": ["arxiv", "semantic_scholar"], "title": "Strata: Hierarchical Context Caching for Long Context Language Model Serving", "abstract": "Large Language Models (LLMs) with expanding context windows face significant performance hurdles. While caching key-value (KV) states is critical for avoiding redundant computation, the storage footprint of long-context caches quickly exceeds GPU memory capacity, forcing production systems to adopt hierarchical caching across memory hierarchies. However, transferring large cached contexts back to the GPU introduces severe performance bottlenecks: fragmented I/O from paged layouts prevents full bandwidth utilization, and existing schedulers fail to account for cache-loading delays, leaving systems loading-bound rather than compute-bound. We present Strata, a hierarchical context caching framework designed for efficient long context LLM serving. Strata introduces GPU-assisted I/O to combat KV cache fragmentation, decoupling GPU and CPU memory layouts and employs cache-aware request scheduling to balance compute with I/O latency and overlapping unavoidable stalls with complementary tasks. Built on SGLang and deployed in production, Strata achieves up to 5x lower Time-To-First-Token (TTFT) compared to vLLM + LMCache and 3.75x speedup over NVIDIA TensorRT-LLM on long-context benchmarks, without degrading short-context performance.", "authors": ["Zhiqiang Xie", "Ziyi Xu", "Mark Zhao", "Yuwei An", "Vikram Sharma Mailthody", "Scott Mahlke", "Michael Garland", "Christos Kozyrakis"], "categories": ["cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-26", "url": "https://arxiv.org/abs/2508.18572", "pdf_url": "https://arxiv.org/pdf/2508.18572v1", "arxiv_id": "2508.18572", "doi": "10.48550/arXiv.2508.18572", "citation_count": 20, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3495}
{"id": "b3e8262df3986432c64958ac20a65a5ba3744be4225a941e3e7e0e4a97403a9c", "sources": ["arxiv", "semantic_scholar"], "title": "Agri-Query: A Case Study on RAG vs. Long-Context LLMs for Cross-Lingual Technical Question Answering", "abstract": "We present a case study evaluating large language models (LLMs) with 128K-token context windows on a technical question answering (QA) task. Our benchmark is built on a user manual for an agricultural machine, available in English, French, and German. It simulates a cross-lingual information retrieval scenario where questions are posed in English against all three language versions of the manual. The evaluation focuses on realistic \"needle-in-a-haystack\" challenges and includes unanswerable questions to test for hallucinations. We compare nine long-context LLMs using direct prompting against three Retrieval-Augmented Generation (RAG) strategies (keyword, semantic, hybrid), with an LLM-as-a-judge for evaluation. Our findings for this specific manual show that Hybrid RAG consistently outperforms direct long-context prompting. Models like Gemini 2.5 Flash and the smaller Qwen 2.5 7B achieve high accuracy (over 85%) across all languages with RAG. This paper contributes a detailed analysis of LLM performance in a specialized industrial domain and an open framework for similar evaluations, highlighting practical trade-offs and challenges.", "authors": ["Julius Gun", "Timo Oksanen"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-25", "url": "https://arxiv.org/abs/2508.18093", "pdf_url": "https://arxiv.org/pdf/2508.18093v2", "arxiv_id": "2508.18093", "doi": "10.48550/arXiv.2508.18093", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2773}
{"id": "1a4af397a4e464a6599e048770f9df1b4d15d44ea1ccf1016982d1b8a2d5246f", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Context Speech Synthesis with Context-Aware Memory", "abstract": "In long-text speech synthesis, current approaches typically convert text to speech at the sentence-level and concatenate the results to form pseudo-paragraph-level speech. These methods overlook the contextual coherence of paragraphs, leading to reduced naturalness and inconsistencies in style and timbre across the long-form speech. To address these issues, we propose a Context-Aware Memory (CAM)-based long-context Text-to-Speech (TTS) model. The CAM block integrates and retrieves both long-term memory and local context details, enabling dynamic memory updates and transfers within long paragraphs to guide sentence-level speech synthesis. Furthermore, the prefix mask enhances the in-context learning ability by enabling bidirectional attention on prefix tokens while maintaining unidirectional generation. Experimental results demonstrate that the proposed method outperforms baseline and state-of-the-art long-context methods in terms of prosody expressiveness, coherence and context inference cost across paragraph-level speech.", "authors": ["Zhipeng Li", "Xiaofen Xing", "Jingyuan Xing", "Hangrui Hu", "Heng Lu", "Xiangmin Xu"], "categories": ["eess.AS", "cs.SD"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2025-08-20", "url": "https://arxiv.org/abs/2508.14713", "pdf_url": "https://arxiv.org/pdf/2508.14713v1", "arxiv_id": "2508.14713", "doi": "10.48550/arXiv.2508.14713", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Interspeech", "quality_score": 0.2716}
{"id": "d6b1a326f50e7b2ff2763b493ead5f36a519565c1d673ba528020642848689b4", "sources": ["arxiv", "semantic_scholar"], "title": "Chunks as Arms: Multi-Armed Bandit-Guided Sampling for Long-Context LLM Preference Optimization", "abstract": "Long-context modeling is critical for a wide range of real-world tasks, including long-context question answering, summarization, and complex reasoning tasks. Recent studies have explored fine-tuning Large Language Models (LLMs) with synthetic data to enhance their long-context capabilities. However, the effectiveness of such approaches is often limited by the low diversity and factual inconsistencies in the generated data. To address these challenges, we propose LongMab, a novel framework that leverages a Multi-Armed Bandit (MAB) rollout strategy to identify the most informative chunks from the given long context for sampling high-quality and diverse responses and constructing preference data pairs for Direct Preference Optimization (DPO) training. Specifically, we treat context chunks as arms of MAB, select chunks based on their expected reward scores to input into LLMs to generate responses, and iteratively update these scores based on reward feedback. Both exploration and exploitation during the rollout process enable the LLM to focus on the most relevant context segments, thereby generating and collecting high-quality and diverse responses. Experimental results on both Llama and Qwen show the effectiveness of LongMab by achieving more than a 4% improvement on long-context reasoning benchmarks. All data and code will be released on https://github.com/NEUIR/LongMab-PO.", "authors": ["Shaohua Duan", "Pengcheng Huang", "Xinze Li", "Zhenghao Liu", "Xiaoyuan Yi", "Yukun Yan", "Shuo Wang", "Yu Gu", "Ge Yu", "Maosong Sun"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-19", "url": "https://arxiv.org/abs/2508.13993", "pdf_url": "https://arxiv.org/pdf/2508.13993v2", "arxiv_id": "2508.13993", "doi": "10.48550/arXiv.2508.13993", "citation_count": 7, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/NEUIR/LongMab-PO", "venue": "arXiv.org", "quality_score": 0.4179}
{"id": "a5ae7594d1f362ae809bbe82d9c5f14b490438e1faaa398374917dd37d4fd041", "sources": ["arxiv", "semantic_scholar"], "title": "ENCODE: Breaking the Trade-Off Between Performance and Efficiency in Long-Term User Behavior Modeling", "abstract": "Long-term user behavior sequences are a goldmine for businesses to explore users' interests to improve Click-Through Rate. However, it is very challenging to accurately capture users' long-term interests from their long-term behavior sequences and give quick responses from the online serving systems. To meet such requirements, existing methods \"inadvertently\" destroy two basic requirements in long-term sequence modeling: R1) make full use of the entire sequence to keep the information as much as possible; R2) extract information from the most relevant behaviors to keep high relevance between learned interests and current target items. The performance of online serving systems is significantly affected by incomplete and inaccurate user interest information obtained by existing methods. To this end, we propose an efficient two-stage long-term sequence modeling approach, named as EfficieNt Clustering based twO-stage interest moDEling (ENCODE), consisting of offline extraction stage and online inference stage. It not only meets the aforementioned two basic requirements but also achieves a desirable balance between online service efficiency and precision. Specifically, in the offline extraction stage, ENCODE clusters the entire behavior sequence and extracts accurate interests. To reduce the overhead of the clustering process, we design a metric learning-based dimension reduction algorithm that preserves the relative pairwise distances of behaviors in the new feature space. While in the online inference stage, ENCODE takes the off-the-shelf user interests to predict the associations with target items. Besides, to further ensure the relevance between user interests and target items, we adopt the same relevance metric throughout the whole pipeline of ENCODE. The extensive experiment and comparison with SOTA have demonstrated the effectiveness and efficiency of our proposed ENCODE.", "authors": ["Wenji Zhou", "Yuhang Zheng", "Yinfu Feng", "Yunan Ye", "Rong Xiao", "Long Chen", "Xiaosong Yang", "Jun Xiao"], "categories": ["cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-19", "url": "https://arxiv.org/abs/2508.13567", "pdf_url": "https://arxiv.org/pdf/2508.13567v1", "arxiv_id": "2508.13567", "doi": "10.1109/TKDE.2024.3486445", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Transactions on Knowledge and Data Engineering", "quality_score": 0.2704}
{"id": "10b493cf841ec9a59d736d2196eed3357044138333e943f9e3ecc46cd9eea687", "sources": ["arxiv", "semantic_scholar"], "title": "Video-EM: Event-Centric Episodic Memory for Long-Form Video Understanding", "abstract": "Video Large Language Models (Video-LLMs) have shown strong video understanding, yet their application to long-form videos remains constrained by limited context windows. A common workaround is to compress long videos into a handful of representative frames via retrieval or summarization. However, most existing pipelines score frames in isolation, implicitly assuming that frame-level saliency is sufficient for downstream reasoning. This often yields redundant selections, fragmented temporal evidence, and weakened narrative grounding for long-form video question answering. We present \\textbf{Video-EM}, a training-free, event-centric episodic memory framework that reframes long-form VideoQA as \\emph{episodic event construction} followed by \\emph{memory refinement}. Instead of treating retrieved keyframes as independent visuals, Video-EM employs an LLM as an active memory agent to orchestrate off-the-shelf tools: it first localizes query-relevant moments via multi-grained semantic matching, then groups and segments them into temporally coherent events, and finally encodes each event as a grounded episodic memory with explicit temporal indices and spatio-temporal cues (capturing \\emph{when}, \\emph{where}, \\emph{what}, and involved entities). To further suppress verbosity and noise from imperfect upstream signals, Video-EM integrates a reasoning-driven self-reflection loop that iteratively verifies evidence sufficiency and cross-event consistency, removes redundancy, and adaptively adjusts event granularity. The outcome is a compact yet reliable \\emph{event timeline} -- a minimal but sufficient episodic memory set that can be directly consumed by existing Video-LLMs without additional training or architectural changes.", "authors": ["Yun Wang", "Long Zhang", "Jingren Liu", "Jiaqi Yan", "Zhanjie Zhang", "Jiahao Zheng", "Ao Ma", "Run Ling", "Xun Yang", "Dapeng Wu", "Xiangyu Chen", "Xuelong Li"], "categories": ["cs.CV", "cs.AI", "cs.MM"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-13", "url": "https://arxiv.org/abs/2508.09486", "pdf_url": "https://arxiv.org/pdf/2508.09486v2", "arxiv_id": "2508.09486", "doi": null, "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2113}
{"id": "36001c9997f1eb9afcd2abaaa9bc044544476fda717f7f88f64202c42b4011f3", "sources": ["arxiv", "semantic_scholar"], "title": "Beyond Naïve Prompting: Strategies for Improved Context-aided Forecasting with LLMs", "abstract": "Real-world forecasting requires models to integrate not only historical data but also relevant contextual information provided in textual form. While large language models (LLMs) show promise for context-aided forecasting, critical challenges remain: we lack diagnostic tools to understand failure modes, performance remains far below their potential, and high computational costs limit practical deployment. We introduce a unified framework of four strategies that address these limitations along three orthogonal dimensions: model diagnostics, accuracy, and efficiency. Through extensive evaluation across model families from small open-source models to frontier models including Gemini, GPT, and Claude, we uncover both fundamental insights and practical solutions. Our findings span three key dimensions: diagnostic strategies reveal the \"Execution Gap\" where models correctly explain how context affects forecasts but fail to apply this reasoning; accuracy-focused strategies achieve substantial performance improvements of 25-50%; and efficiency-oriented approaches show that adaptive routing between small and large models can approach large model accuracy on average while significantly reducing inference costs. These orthogonal strategies can be flexibly integrated based on deployment constraints, providing practitioners with a comprehensive toolkit for practical LLM-based context-aided forecasting.", "authors": ["Arjun Ashok", "Andrew Robert Williams", "Vincent Zhihao Zheng", "Irina Rish", "Nicolas Chapados", "Étienne Marcotte", "Valentina Zantedeschi", "Alexandre Drouin"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-13", "url": "https://arxiv.org/abs/2508.09904", "pdf_url": "https://arxiv.org/pdf/2508.09904v2", "arxiv_id": "2508.09904", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": null, "quality_score": 0.3115}
{"id": "1f91c9945f7edabc4068d1183c4a87eb660bd5920bf70da18a300a546e01e7d2", "sources": ["arxiv", "semantic_scholar"], "title": "Architecting Long-Context LLM Acceleration with Packing-Prefetch Scheduler and Ultra-Large Capacity On-Chip Memories", "abstract": "Long-context Large Language Model (LLM) inference faces increasing compute bottlenecks as attention calculations scale with context length, primarily due to the growing KV-cache transfer overhead that saturates High Bandwidth Memory (HBM). While prefetching techniques mitigate cache misses by fetching KV data in advance, their spatial and temporal benefits present new opportunities to exploit. This work proposes a packing-prefetch scheduling architecture with monolithic 3D (M3D) back-end-of-line (BEOL) compatible embedded memories with ultra-large on-chip capacity to accelerate long-context LLM inference. Our optimizations demonstrate 8.06x decode speedup and 1.83x overall latency reduction on Llama3.1-8B using TPUv6e-like hardware with additional 512MB BEOL memories over the serial execution. Evaluations of multi-request workloads on TPU-like architectures show 1.7x-2.4x throughput improvement and 1.5x-2.4x HBM bandwidth reduction compared to packing-only methods on Llama3.1-8B and Llama3.1-70B models. With the co-design of packing, prefetching, and BEOL memories, our approach alleviates HBM constraints and enables efficient long-context LLM inference.", "authors": ["Ming-Yen Lee", "Faaiq Waqar", "Hanchen Yang", "Muhammed Ahosan Ul Karim", "Harsono Simka", "Shimeng Yu"], "categories": ["cs.AR", "cs.ET"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-11", "url": "https://arxiv.org/abs/2508.08457", "pdf_url": "https://arxiv.org/pdf/2508.08457v1", "arxiv_id": "2508.08457", "doi": "10.48550/arXiv.2508.08457", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Micro", "quality_score": 0.2612}
{"id": "78e66fa9e9eaf019774cfa200727812ce90505b2a905458e571ea7bb72ba4600", "sources": ["arxiv", "semantic_scholar"], "title": "Positional Biases Shift as Inputs Approach Context Window Limits", "abstract": "Large Language Models (LLMs) often struggle to use information across long inputs effectively. Prior work has identified positional biases, such as the Lost in the Middle (LiM) effect, where models perform better when information appears at the beginning (primacy bias) or end (recency bias) of the input, rather than in the middle. However, long-context studies have not consistently replicated these effects, raising questions about their intensity and the conditions under which they manifest. To address this, we conducted a comprehensive analysis using relative rather than absolute input lengths, defined with respect to each model's context window. Our findings reveal that the LiM effect is strongest when inputs occupy up to 50% of a model's context window. Beyond that, the primacy bias weakens, while recency bias remains relatively stable. This effectively eliminates the LiM effect; instead, we observe a distance-based bias, where model performance is better when relevant information is closer to the end of the input. Furthermore, our results suggest that successful retrieval is a prerequisite for reasoning in LLMs, and that the observed positional biases in reasoning are largely inherited from retrieval. These insights have implications for long-context tasks, the design of future LLM benchmarks, and evaluation methodologies for LLMs handling extended inputs.", "authors": ["Blerta Veseli", "Julian Chibane", "Mariya Toneva", "Alexander Koller"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-10", "url": "https://arxiv.org/abs/2508.07479", "pdf_url": "https://arxiv.org/pdf/2508.07479v1", "arxiv_id": "2508.07479", "doi": "10.48550/arXiv.2508.07479", "citation_count": 11, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2698}
{"id": "97ce2f6032aaa14f174432e261c3379cd6311a9b7c5990112fd5d6770634b879", "sources": ["arxiv", "semantic_scholar"], "title": "Context Engineering for Multi-Agent LLM Code Assistants Using Elicit, NotebookLM, ChatGPT, and Claude Code", "abstract": "Large Language Models (LLMs) have shown promise in automating code generation and software engineering tasks, yet they often struggle with complex, multi-file projects due to context limitations and knowledge gaps. We propose a novel context engineering workflow that combines multiple AI components: an Intent Translator (GPT-5) for clarifying user requirements, an Elicit-powered semantic literature retrieval for injecting domain knowledge, NotebookLM-based document synthesis for contextual understanding, and a Claude Code multi-agent system for code generation and validation. Our integrated approach leverages intent clarification, retrieval-augmented generation, and specialized sub-agents orchestrated via Claude's agent framework. We demonstrate that this method significantly improves the accuracy and reliability of code assistants in real-world repositories, yielding higher single-shot success rates and better adherence to project context than baseline single-agent approaches. Qualitative results on a large Next.js codebase show the multi-agent system effectively plans, edits, and tests complex features with minimal human intervention. We compare our system with recent frameworks like CodePlan, MASAI, and HyperAgent, highlighting how targeted context injection and agent role decomposition lead to state-of-the-art performance. Finally, we discuss the implications for deploying LLM-based coding assistants in production, along with lessons learned on context management and future research directions.", "authors": ["Muhammad Haseeb"], "categories": ["cs.SE", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-09", "url": "https://arxiv.org/abs/2508.08322", "pdf_url": "https://arxiv.org/pdf/2508.08322v1", "arxiv_id": "2508.08322", "doi": "10.48550/arXiv.2508.08322", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.259}
{"id": "f3b46a1d04f06b6a5ba6c395fb31ec9b07f94953279e8d9237bd0dc6bf724e66", "sources": ["arxiv", "semantic_scholar"], "title": "SlimInfer: Accelerating Long-Context LLM Inference via Dynamic Token Pruning", "abstract": "Long-context inference for Large Language Models (LLMs) is heavily limited by high computational demands. While several existing methods optimize attention computation, they still process the full set of hidden states at each layer, limiting overall efficiency. In this work, we propose SlimInfer, an innovative framework that aims to accelerate inference by directly pruning less critical prompt tokens during the forward pass. Our key insight is an information diffusion phenomenon: As information from critical tokens propagates through layers, it becomes distributed across the entire sequence. This diffusion process suggests that LLMs can maintain their semantic integrity when excessive tokens, even including these critical ones, are pruned in hidden states. Motivated by this, SlimInfer introduces a dynamic fine-grained pruning mechanism that accurately removes redundant tokens of hidden state at intermediate layers. This layer-wise pruning naturally enables an asynchronous KV cache manager that prefetches required token blocks without complex predictors, reducing both memory usage and I/O costs. Extensive experiments show that SlimInfer can achieve up to $\\mathbf{2.53\\times}$ time-to-first-token (TTFT) speedup and $\\mathbf{1.88\\times}$ end-to-end latency reduction for LLaMA3.1-8B-Instruct on a single RTX 4090, without sacrificing performance on LongBench. Our code is available at https://github.com/Longxmas/SlimInfer.", "authors": ["Lingkun Long", "Rubing Yang", "Yushi Huang", "Desheng Hui", "Ao Zhou", "Jianlei Yang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-08", "url": "https://arxiv.org/abs/2508.06447", "pdf_url": "https://arxiv.org/pdf/2508.06447v2", "arxiv_id": "2508.06447", "doi": "10.48550/arXiv.2508.06447", "citation_count": 8, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/Longxmas/SlimInfer", "venue": "AAAI Conference on Artificial Intelligence", "quality_score": 0.3984}
{"id": "cb9103378f1d2950c4d5e4e0cf92e48d23e1382dfe743ce496a66efce4ce24ec", "sources": ["arxiv", "semantic_scholar"], "title": "Sculptor: Empowering LLMs with Cognitive Agency via Active Context Management", "abstract": "Large Language Models (LLMs) suffer from significant performance degradation when processing long contexts due to proactive interference, where irrelevant information in earlier parts of the context disrupts reasoning and memory recall. While most research focuses on external memory systems to augment LLMs' capabilities, we propose a complementary approach: empowering LLMs with Active Context Management (ACM) tools to actively sculpt their internal working memory. We introduce Sculptor, a framework that equips LLMs with three categories of tools: (1) context fragmentation, (2) summary, hide, and restore, and (3) precise search. Our approach enables LLMs to proactively manage their attention and working memory, analogous to how humans selectively focus on relevant information while filtering out distractions. Experimental evaluation on diverse long-context benchmarks demonstrates that Sculptor significantly improves performance even without specific training, leveraging LLMs' inherent tool-calling and instruction-following capabilities. To further optimize these strategies, we introduce a novel dynamic context-aware reinforcement learning (RL) approach, advancing the training of an agent that actively modifies its own conversational history. By enabling Active Context Management, Sculptor not only mitigates proactive interference but also provides a cognitive foundation for more reliable reasoning across diverse long-context tasks-highlighting that explicit context-control strategies, rather than merely larger token windows, are key to robustness at scale.", "authors": ["Mo Li", "L. H. Xu", "Qitai Tan", "Long Ma", "Ting Cao", "Yunxin Liu"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-06", "url": "https://arxiv.org/abs/2508.04664", "pdf_url": "https://arxiv.org/pdf/2508.04664v2", "arxiv_id": "2508.04664", "doi": "10.48550/arXiv.2508.04664", "citation_count": 6, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2555}
{"id": "5601a73b5f6820311b8d56d09f804a5c10194dd606beec72ed0ee4616d17296f", "sources": ["arxiv", "semantic_scholar"], "title": "LaMPE: Length-aware Multi-grained Positional Encoding for Adaptive Long-context Scaling Without Training", "abstract": "Large language models (LLMs) experience significant performance degradation when the input exceeds the pretraining context window, primarily due to the out-of-distribution (OOD) behavior of Rotary Position Embedding (RoPE). Recent studies mitigate this problem by remapping OOD positions into the in-distribution range with fixed mapping strategies, ignoring the dynamic relationship between input length and the model's effective context window. To this end, we propose Length-aware Multi-grained Positional Encoding (LaMPE), a training-free method that fully utilizes the model's effective context window for adaptive long-context scaling in LLMs. Motivated by the left-skewed frequency distribution of relative positions, LaMPE establishes a dynamic relationship between mapping length and input length through a parametric scaled sigmoid function to adaptively allocate positional capacity across varying input lengths. Meanwhile, LaMPE devises a novel multi-grained attention mechanism that strategically allocates positional resolution across different sequence regions to capture both fine-grained locality and long-range dependencies. Our method can be seamlessly applied to a wide range of RoPE-based LLMs without training. Extensive experiments on three representative LLMs across five mainstream long-context benchmarks demonstrate that LaMPE achieves significant performance improvements compared to existing length extrapolation methods. The code will be released at https://github.com/scar-on/LaMPE.", "authors": ["Sikui Zhang", "Guangze Gao", "Ziyun Gan", "Chunfeng Yuan", "Zefeng Lin", "Houwen Peng", "Bing Li", "Weiming Hu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-04", "url": "https://arxiv.org/abs/2508.02308", "pdf_url": "https://arxiv.org/pdf/2508.02308v2", "arxiv_id": "2508.02308", "doi": "10.48550/arXiv.2508.02308", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/scar-on/LaMPE", "venue": "arXiv.org", "quality_score": 0.3914}
{"id": "f7cf427c1ca6d555955f1cf27acaf53f64df9b57b48fe6e48307a7c95e66d1de", "sources": ["arxiv", "semantic_scholar"], "title": "KCR: Resolving Long-Context Knowledge Conflicts via Reasoning in LLMs", "abstract": "Knowledge conflicts commonly arise across diverse sources, and their prevalence has increased with the advent of LLMs. When dealing with conflicts between multiple contexts, also known as \\emph{inter-context knowledge conflicts}, LLMs are often confused by lengthy and conflicting contexts. To address this challenge, we propose the Knowledge Conflict Reasoning (KCR) framework, which enhances the ability of LLMs to resolve conflicting knowledge. The key idea of KCR is to train backbone LLMs to establish a correct reasoning process by rewarding them for selecting and adhering to the context with stronger logical consistency when presented with conflicting contexts. Specifically, we first extract reasoning paths, represented by either text or local knowledge graphs, from the conflicting long contexts. Subsequently, we employ Reinforcement Learning to encourage the model to learn the paradigm of reasoning process that follows correct reasoning paths rather than the incorrect counterparts. This enables the backbone models to genuinely acquire the capability to resolve inter-context knowledge conflicts within long contexts. Experimental results demonstrate that our framework significantly improves the ability of various backbone models to resolve knowledge conflicts in long-context scenarios, yielding substantial performance gains.", "authors": ["Xianda Zheng", "Zijian Huang", "Meng-Fen Chiang", "Michael J. Witbrock", "Kaiqi Zhao"], "categories": ["cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-08-02", "url": "https://arxiv.org/abs/2508.01273", "pdf_url": "https://arxiv.org/pdf/2508.01273v2", "arxiv_id": "2508.01273", "doi": "10.48550/arXiv.2508.01273", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2509}
{"id": "f7fec5fdbb6b3261037a7939ab2eb3a22258074ca8f0110e926b2464d072135e", "sources": ["arxiv", "semantic_scholar"], "title": "TextQuests: How Good are LLMs at Text-Based Video Games?", "abstract": "Evaluating AI agents within complex, interactive environments that mirror real-world challenges is critical for understanding their practical capabilities. While existing agent benchmarks effectively assess skills like tool use or performance on structured tasks, they often do not fully capture an agent's ability to operate autonomously in exploratory environments that demand sustained, self-directed reasoning over a long and growing context. To enable a more accurate assessment of AI agents in challenging exploratory environments, we introduce TextQuests, a benchmark based on the Infocom suite of interactive fiction games. These text-based adventures, which can take human players over 30 hours and require hundreds of precise actions to solve, serve as an effective proxy for evaluating AI agents on focused, stateful tasks. The benchmark is specifically designed to assess an LLM agent's capacity for self-contained problem-solving by precluding the use of external tools, thereby focusing on intrinsic long-context reasoning capabilities in an exploratory environment characterized by the need for trial-and-error learning and sustained problem-solving within a single interactive session. We release TextQuests at https://textquests.ai.", "authors": ["Long Phan", "Mantas Mazeika", "Andy Zou", "Dan Hendrycks"], "categories": ["cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-07-31", "url": "https://arxiv.org/abs/2507.23701", "pdf_url": "https://arxiv.org/pdf/2507.23701v3", "arxiv_id": "2507.23701", "doi": "10.48550/arXiv.2507.23701", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2486}
{"id": "ce50f30f946e3e4eb28e58d9ffd32d52a05948f0eceb8e801603994b576e17db", "sources": ["arxiv", "semantic_scholar"], "title": "Context-aware Rotary Position Embedding", "abstract": "Positional encoding is a vital component of Transformer architectures, enabling models to incorporate sequence order into self-attention mechanisms. Rotary Positional Embeddings (RoPE) have become a widely adopted solution due to their compatibility with relative position encoding and computational efficiency. However, RoPE relies on static, input-independent sinusoidal frequency patterns, limiting its ability to model context-sensitive relationships. In this work, we propose CARoPE (Context-Aware Rotary Positional Embedding), a novel generalization of RoPE that dynamically generates head-specific frequency patterns conditioned on token embeddings. This design introduces token- and context-sensitive positional representations while preserving RoPE efficiency and architectural simplicity. CARoPE computes input-dependent phase shifts using a bounded transformation of token embeddings and integrates them into the rotary mechanism across attention heads. We evaluate CARoPE on the FineWeb-Edu-10B dataset using GPT-2 variants trained on next-token prediction tasks. Experimental results show that CARoPE consistently outperforms RoPE and other common positional encoding baselines, achieving significantly lower perplexity, even at longer context lengths. Additionally, CARoPE enables faster training throughput without sacrificing model stability. These findings demonstrate that CARoPE offers a scalable, expressive, and efficient upgrade to existing positional encoding strategies in Transformer models.", "authors": ["Ali Veisi", "Delaram Fartoot", "Hamidreza Amirzadeh"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-07-30", "url": "https://arxiv.org/abs/2507.23083", "pdf_url": "https://arxiv.org/pdf/2507.23083v1", "arxiv_id": "2507.23083", "doi": "10.48550/arXiv.2507.23083", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2475}
{"id": "c2ef6d830d7fa6c88863d676d1aed062c698c1d186f7cc11fce356d9d75c277d", "sources": ["arxiv", "semantic_scholar"], "title": "MCP4EDA: LLM-Powered Model Context Protocol RTL-to-GDSII Automation with Backend Aware Synthesis Optimization", "abstract": "This paper presents MCP4EDA, the first Model Context Protocol server that enables Large Language Models (LLMs) to control and optimize the complete open-source RTL-to-GDSII design flow through natural language interaction. The system integrates Yosys synthesis, Icarus Verilog simulation, OpenLane place-and-route, GTKWave analysis, and KLayout visualization into a unified LLM-accessible interface, enabling designers to execute complex multi-tool EDA workflows conversationally via AI assistants such as Claude Desktop and Cursor IDE. The principal contribution is a backend-aware synthesis optimization methodology wherein LLMs analyze actual post-layout timing, power, and area metrics from OpenLane results to iteratively refine synthesis TCL scripts, establishing a closed-loop optimization system that bridges the traditional gap between synthesis estimates and physical implementation reality. In contrast to conventional flows that rely on wire-load models, this methodology leverages real backend performance data to guide synthesis parameter tuning, optimization sequence selection, and constraint refinement, with the LLM functioning as an intelligent design space exploration agent. Experimental evaluation on representative digital designs demonstrates 15-30% improvements in timing closure and 10-20% area reduction compared to default synthesis flows, establishing MCP4EDA as the first practical LLM-controlled end-to-end open-source EDA automation system. The code and demo are avaiable at: http://www.agent4eda.com/", "authors": ["Yiting Wang", "Wanghao Ye", "Yexiao He", "Yiran Chen", "Gang Qu", "Ang Li"], "categories": ["cs.AR", "cs.MA"], "fields_of_study": ["Computer Science"], "published_date": "2025-07-25", "url": "https://arxiv.org/abs/2507.19570", "pdf_url": "https://arxiv.org/pdf/2507.19570v1", "arxiv_id": "2507.19570", "doi": "10.48550/arXiv.2507.19570", "citation_count": 8, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3736}
{"id": "2aa123efcd4e069b1b7d241f372053eb9fd6479ab1f10883c32df57c55f276fe", "sources": ["arxiv", "semantic_scholar"], "title": "Smooth Reading: Bridging the Gap of Recurrent LLM to Self-Attention LLM on Long-Context Tasks", "abstract": "Recently, recurrent large language models (Recurrent LLMs) with linear computational complexity have re-emerged as efficient alternatives to self-attention-based LLMs (Self-Attention LLMs), which have quadratic complexity. However, Recurrent LLMs often underperform on long-context tasks due to their limited fixed-size memory. Previous research has primarily focused on enhancing the memory capacity of Recurrent LLMs through architectural innovations, but these approaches have not yet enabled Recurrent LLMs to match the performance of Self-Attention LLMs on long-context tasks. We argue that this limitation arises because processing the entire context at once is not well-suited for Recurrent LLMs. In this paper, we propose Smooth Reading, a chunk-wise inference method inspired by human reading strategies. Smooth Reading processes context in chunks and iteratively summarizes the contextual information, thereby reducing memory demands and making the approach more compatible with Recurrent LLMs. Our experimental results show that this method substantially narrows the performance gap between Recurrent and Self-Attention LLMs on long-context tasks, while preserving the efficiency advantages of Recurrent LLMs. Our Smooth Reading boosts SWA-3B-4k (a Recurrent LLM) from 5.68% lower to 3.61% higher performance than Self-Attention LLMs on LongBench. Besides, our method maintains the high efficiency, training 3x faster and inferring 2x faster at 64k context compared to Self-Attention LLMs. To our knowledge, this is the first work to achieve comparable performance using Recurrent LLMs compared with Self-Attention LLMs on long-context tasks. We hope our method will inspire future research in this area. To facilitate further progress, we will release code and dataset.", "authors": ["Kai Liu", "Zhan Su", "Peijie Dong", "Fengran Mo", "Jianfei Gao", "ShaoTing Zhang", "Kai Chen"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-07-25", "url": "https://arxiv.org/abs/2507.19353", "pdf_url": "https://arxiv.org/pdf/2507.19353v1", "arxiv_id": "2507.19353", "doi": "10.48550/arXiv.2507.19353", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2418}
{"id": "3d443fa3d27479a27ff9d90ebe3b0c17152bf18c6770bf08280ef1f43b0e10d5", "sources": ["arxiv", "semantic_scholar"], "title": "Ref-Long: Benchmarking the Long-context Referencing Capability of Long-context Language Models", "abstract": "Long-context language models (LCLMs) have exhibited impressive capabilities in long-context understanding tasks. Among these, long-context referencing -- a crucial task that requires LCLMs to attribute items of interest to specific parts of long-context data -- remains underexplored. To bridge this gap, this paper proposes Referencing Evaluation for Long-context Language Models (Ref-Long), a novel benchmark designed to assess the long-context referencing capability of LCLMs. Specifically, Ref-Long requires LCLMs to identify the indexes of documents that reference a specific key, emphasizing contextual relationships between the key and the documents over simple retrieval. Based on the task design, we construct three subsets ranging from synthetic to realistic scenarios to form the Ref-Long benchmark. Experimental results of 13 LCLMs reveal significant shortcomings in long-context referencing, even among advanced models like GPT-4o. To further investigate these challenges, we conduct comprehensive analyses, including human evaluations, task format adjustments, fine-tuning experiments, and error analyses, leading to several key insights. Our data and code can be found in https://github. com/wujunjie1998/Ref-Long.", "authors": ["Junjie Wu", "Gefei Gu", "Yanan Zheng", "Dit-Yan Yeung", "Arman Cohan"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-07-13", "url": "https://arxiv.org/abs/2507.09506", "pdf_url": "https://arxiv.org/pdf/2507.09506v2", "arxiv_id": "2507.09506", "doi": "10.18653/v1/2025.acl-long.1162", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.228}
{"id": "e5cbd5333f70e0dfaa0b81f3f5e5550e4838a7dd40deb51ebcabe0ee9d4f4b38", "sources": ["arxiv", "semantic_scholar"], "title": "ETT: Expanding the Long Context Understanding Capability of LLMs at Test-Time", "abstract": "Transformer-based Language Models' computation and memory overhead increase quadratically as a function of sequence length. The quadratic cost poses challenges when employing LLMs for processing long sequences. In this work, we introduce \\ourmodelacronym~(Extend at Test-Time), method for extending the context length of short context Transformer-based LLMs, with constant memory requirement and linear computation overhead. ETT enable the extension of the context length at test-time by efficient fine-tuning the model's parameters on the input context, chunked into overlapping small subsequences. We evaluate ETT on LongBench by extending the context length of GPT-Large and Phi-2 up to 32 times, increasing from 1k to 32k tokens. This results in up to a 30 percent improvement in the model's accuracy. We also study how context can be stored in LLM's weights effectively and efficiently. Through a detailed ablation study, we examine which Transformer modules are most beneficial to fine-tune at test-time. Interestingly, we find that fine-tuning the second layer of the FFNs is more effective than full fine-tuning, leading to a further improvement in the models' accuracy.", "authors": ["Kiarash Zahirnia", "Zahra Golpayegani", "Walid Ahmed", "Yang Liu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-07-08", "url": "https://arxiv.org/abs/2507.06313", "pdf_url": "https://arxiv.org/pdf/2507.06313v3", "arxiv_id": "2507.06313", "doi": "10.48550/arXiv.2507.06313", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2223}
{"id": "c39b6b6cdedb7b5defe6fbbca555e574abb4e33da13794c776931d295c20196e", "sources": ["arxiv", "semantic_scholar"], "title": "PERK: Long-Context Reasoning as Parameter-Efficient Test-Time Learning", "abstract": "Long-context reasoning requires accurately identifying relevant information in extensive, noisy input contexts. Previous research shows that using test-time learning to encode context directly into model parameters can effectively enable reasoning over noisy information. However, meta-learning methods for enabling test-time learning are prohibitively memory-intensive, preventing their application to long context settings. In this work, we propose PERK (Parameter Efficient Reasoning over Knowledge), a scalable approach for learning to encode long input contexts using gradient updates to a lightweight model adapter at test time. Specifically, PERK employs two nested optimization loops in a meta-training phase. The inner loop rapidly encodes contexts into a low-rank adapter (LoRA) that serves as a parameter-efficient memory module for the base model. Concurrently, the outer loop learns to use the updated adapter to accurately recall and reason over relevant information from the encoded long context. Our evaluations on several long-context reasoning tasks show that PERK significantly outperforms the standard prompt-based long-context baseline, achieving average absolute performance gains of up to 90% for smaller models (GPT-2) and up to 27% for our largest evaluated model, Qwen-2.5-0.5B. In general, PERK is more robust to reasoning complexity, length extrapolation, and the locations of relevant information in contexts. Finally, we show that while PERK is memory-intensive during training, it scales more efficiently at inference time than prompt-based long-context inference.", "authors": ["Zeming Chen", "Angelika Romanou", "Gail Weiss", "Antoine Bosselut"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-07-08", "url": "https://arxiv.org/abs/2507.06415", "pdf_url": "https://arxiv.org/pdf/2507.06415v2", "arxiv_id": "2507.06415", "doi": "10.48550/arXiv.2507.06415", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2223}
{"id": "d1884cc1ad02cb070003cfe41df0f90090fd0fda9f9cbffe695609e28ebdf415", "sources": ["arxiv", "semantic_scholar"], "title": "Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities", "abstract": "In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving.", "authors": ["Gheorghe Comanici", "Eric Bieber", "Mike Schaekermann", "Ice Pasupat", "Noveen Sachdeva", "Inderjit Dhillon", "Marcel Blistein", "Ori Ram", "Dan Zhang", "Evan Rosen", "Luke Marris", "Sam Petulla", "Colin Gaffney", "Asaf Aharoni", "Nathan Lintz", "Tiago Cardal Pais", "Henrik Jacobsson", "Idan Szpektor", "Nan-Jiang Jiang", "Krishna Haridasan", "Ahmed Omran", "Nikunj Saunshi", "Dara Bahri", "Gaurav Mishra", "Eric Chu", "Toby Boyd", "Brad Hekman", "Aaron Parisi", "Chaoyi Zhang", "Kornraphop Kawintiranon", "Tania Bedrax-Weiss", "Oliver Wang", "Ya Xu", "Ollie Purkiss", "Uri Mendlovic", "Ilaï Deutel", "Nam Nguyen", "Adam Langley", "Flip Korn", "Lucia Rossazza", "Alexandre Ramé", "Sagar Waghmare", "Helen Miller", "Nathan Byrd", "Ashrith Sheshan", "Raia Hadsell", "Sangnie Bhardwaj", "Pawel Janus", "Tero Rissa", "Dan Horgan", "Alvin Abdagic", "Lior Belenki", "James Allingham", "Anima Singh", "Theo Guidroz", "Srivatsan Srinivasan", "Herman Schmit", "Kristen Chiafullo", "Andre Elisseeff", "Nilpa Jha", "Prateek Kolhar", "Leonard Berrada", "Frank Ding", "Xiance Si", "Shrestha Basu Mallick", "Franz Och", "Sofia Erell", "Eric Ni", "Tejasi Latkar", "Sherry Yang", "Petar Sirkovic", "Ziqiang Feng", "Robert Leland", "Rachel Hornung", "Gang Wu", "Charles Blundell", "Hamidreza Alvari", "Po-Sen Huang", "Cathy Yip", "Sanja Deur", "Li Liu", "Gabriela Surita", "Pablo Duque", "Dima Damen", "Johnson Jia", "Arthur Guez", "Markus Mircea", "Animesh Sinha", "Alberto Magni", "Paweł Stradomski", "Tal Marian", "Vlado Galić", "Wenhu Chen", "Hisham Husain", "Achintya Singhal", "Dominik Grewe", "François-Xavier Aubet", "Shuang Song", "Lorenzo Blanco", "Leland Rechis", "Lewis Ho", "Rich Munoz", "Kelvin Zheng", "Jessica Hamrick", "Kevin Mather", "Hagai Taitelbaum", "Eliza Rutherford", "Yun Lei", "Kuangyuan Chen", "Anand Shukla", "Erica Moreira", "Eric Doi", "Berivan Isik", "Nir Shabat", "Dominika Rogozińska", "Kashyap Kolipaka", "Jason Chang", "Eugen Vušak", "Srinivasan Venkatachary", "Shadi Noghabi", "Tarun Bharti", "Younghoon Jun", "Aleksandr Zaks", "Simon Green", "Jeshwanth Challagundla", "William Wong", "Muqthar Mohammad", "Dean Hirsch", "Yong Cheng", "Iftekhar Naim", "Lev Proleev", "Damien Vincent", "Aayush Singh", "Maxim Krikun", "Dilip Krishnan", "Zoubin Ghahramani", "Aviel Atias", "Rajeev Aggarwal", "Christo Kirov", "Dimitrios Vytiniotis", "Christy Koh", "Alexandra Chronopoulou", "Pawan Dogra", "Vlad-Doru Ion", "Gladys Tyen", "Jason Lee", "Felix Weissenberger", "Trevor Strohman", "Ashwin Balakrishna", "Jack Rae", "Marko Velic", "Raoul de Liedekerke", "Oded Elyada", "Wentao Yuan", "Canoee Liu", "Lior Shani", "Sergey Kishchenko", "Bea Alessio", "Yandong Li", "Richard Song", "Sam Kwei", "Orion Jankowski", "Aneesh Pappu", "Youhei Namiki", "Yenai Ma", "Nilesh Tripuraneni", "Colin Cherry", "Marissa Ikonomidis", "Yu-Cheng Ling", "Colin Ji", "Beka Westberg", "Auriel Wright", "Da Yu", "David Parkinson", "Swaroop Ramaswamy", "Jerome Connor", "Soheil Hassas Yeganeh", "Snchit Grover", "George Kenwright", "Lubo Litchev", "Chris Apps", "Alex Tomala", "Felix Halim", "Alex Castro-Ros", "Zefei Li", "Anudhyan Boral", "Pauline Sho", "Michal Yarom", "Eric Malmi", "David Klinghoffer", "Rebecca Lin", "Alan Ansell", "Pradeep Kumar S", "Shubin Zhao", "Siqi Zuo", "Adam Santoro", "Heng-Tze Cheng", "Solomon Demmessie", "Yuchi Liu", "Nicole Brichtova", "Allie Culp", "Nathaniel Braun", "Dan Graur", "Will Ng", "Nikhil Mehta", "Aaron Phillips", "Patrik Sundberg", "Varun Godbole", "Fangyu Liu", "Yash Katariya", "David Rim", "Mojtaba Seyedhosseini", "Sean Ammirati", "Jonas Valfridsson", "Mahan Malihi", "Timothy Knight", "Andeep Toor", "Thomas Lampe", "Abe Ittycheriah", "Lewis Chiang", "Chak Yeung", "Alexandre Fréchette", "Jinmeng Rao", "Huisheng Wang", "Himanshu Srivastava", "Richard Zhang", "Rocky Rhodes", "Ariel Brand", "Dean Weesner", "Ilya Figotin", "Felix Gimeno", "Rachana Fellinger", "Pierre Marcenac", "José Leal", "Eyal Marcus", "Victor Cotruta", "Rodrigo Cabrera", "Sheryl Luo", "Dan Garrette", "Vera Axelrod", "Sorin Baltateanu", "David Barker", "Dongkai Chen", "Horia Toma", "Ben Ingram", "Jason Riesa", "Chinmay Kulkarni", "Yujing Zhang", "Hongbin Liu", "Chao Wang", "Martin Polacek", "Will Wu", "Kai Hui", "Adrian N Reyes", "Yi Su", "Megan Barnes", "Ishaan Malhi", "Anfal Siddiqui", "Qixuan Feng", "Mihai Damaschin", "Daniele Pighin", "Andreas Steiner", "Samuel Yang", "Ramya Sree Boppana", "Simeon Ivanov", "Arun Kandoor", "Aditya Shah", "Asier Mujika", "Da Huang", "Christopher A. Choquette-Choo", "Mohak Patel", "Tianhe Yu", "Toni Creswell", " Jerry", " Liu", "Catarina Barros", "Yasaman Razeghi", "Aurko Roy", "Phil Culliton", "Binbin Xiong", "Jiaqi Pan", "Thomas Strohmann", "Tolly Powell", "Babi Seal", "Doug DeCarlo", "Pranav Shyam", "Kaan Katircioglu", "Xuezhi Wang", "Cassidy Hardin", "Immanuel Odisho", "Josef Broder", "Oscar Chang", "Arun Nair", "Artem Shtefan", "Maura O'Brien", "Manu Agarwal", "Sahitya Potluri", "Siddharth Goyal", "Amit Jhindal", "Saksham Thakur", "Yury Stuken", "James Lyon", "Kristina Toutanova", "Fangxiaoyu Feng", "Austin Wu", "Ben Horn", "Alek Wang", "Alex Cullum", "Gabe Taubman", "Disha Shrivastava", "Chongyang Shi", "Hamish Tomlinson", "Roma Patel", "Tao Tu", "Ada Maksutaj Oflazer", "Francesco Pongetti", "Mingyao Yang", "Adrien Ali Taïga", "Vincent Perot", "Nuo Wang Pierse", "Feng Han", "Yoel Drori", "Iñaki Iturrate", "Ayan Chakrabarti", "Legg Yeung", "Dave Dopson", "Yi-ting Chen", "Apoorv Kulshreshtha", "Tongfei Guo", "Philip Pham", "Tal Schuster", "Junquan Chen", "Alex Polozov", "Jinwei Xing", "Huanjie Zhou", "Praneeth Kacham", "Doron Kukliansky", "Antoine Miech", "Sergey Yaroshenko", "Ed Chi", "Sholto Douglas", "Hongliang Fei", "Mathieu Blondel", "Preethi Myla", "Lior Madmoni", "Xing Wu", "Daniel Keysers", "Kristian Kjems", "Isabela Albuquerque", "Lijun Yu", "Joel D'sa", "Michelle Plantan", "Vlad Ionescu", "Jaume Sanchez Elias", "Abhirut Gupta", "Manish Reddy Vuyyuru", "Fred Alcober", "Tong Zhou", "Kaiyang Ji", "Florian Hartmann", "Subha Puttagunta", "Hugo Song", "Ehsan Amid", "Anca Stefanoiu", "Andrew Lee", "Paul Pucciarelli", "Emma Wang", "Amit Raul", "Slav Petrov", "Isaac Tian", "Valentin Anklin", "Nana Nti", "Victor Gomes", "Max Schumacher", "Grace Vesom", "Alex Panagopoulos", "Konstantinos Bousmalis", "Daniel Andor", "Josh Jacob", "Yuan Zhang", "Bill Rosgen", "Matija Kecman", "Matthew Tung", "Alexandra Belias", "Noah Goodman", "Paul Covington", "Brian Wieder", "Nikita Saxena", "Elnaz Davoodi", "Muhuan Huang", "Sharath Maddineni", "Vincent Roulet", "Folawiyo Campbell-Ajala", "Pier Giuseppe Sessa", " Xintian", " Wu", "Guangda Lai", "Paul Collins", "Alex Haig", "Vytenis Sakenas", "Xiaowei Xu", "Marissa Giustina", "Laurent El Shafey", "Pichi Charoenpanit", "Shefali Garg", "Joshua Ainslie", "Boone Severson", "Montse Gonzalez Arenas", "Shreya Pathak", "Sujee Rajayogam", "Jie Feng", "Michiel Bakker", "Sheng Li", "Nevan Wichers", "Jamie Rogers", "Xinyang Geng", "Yeqing Li", "Rolf Jagerman", "Chao Jia", "Nadav Olmert", "David Sharon", "Matthew Mauger", "Sandeep Mariserla", "Hongxu Ma", "Megha Mohabey", "Kyuyeun Kim", "Alek Andreev", "Scott Pollom", "Juliette Love", "Vihan Jain", "Priyanka Agrawal", "Yannick Schroecker", "Alisa Fortin", "Manfred Warmuth", "Ji Liu", "Andrew Leach", "Irina Blok", "Ganesh Poomal Girirajan", "Roee Aharoni", "Benigno Uria", "Andrei Sozanschi", "Dan Goldberg", "Lucian Ionita", "Marco Tulio Ribeiro", "Martin Zlocha", "Vighnesh Birodkar", "Sami Lachgar", "Liangzhe Yuan", "Himadri Choudhury", "Matt Ginsberg", "Fei Zheng", "Gregory Dibb", "Emily Graves", "Swachhand Lokhande", "Gabriel Rasskin", "George-Cristian Muraru", "Corbin Quick", "Sandeep Tata", "Pierre Sermanet", "Aditya Chawla", "Itay Karo", "Yan Wang", "Susan Zhang", "Orgad Keller", "Anca Dragan", "Guolong Su", "Ian Chou", "Xi Liu", "Yiqing Tao", "Shruthi Prabhakara", "Marc Wilson", "Ruibo Liu", "Shibo Wang", "Georgie Evans", "David Du", "Alfonso Castaño", "Gautam Prasad", "Mona El Mahdy", "Sebastian Gerlach", "Machel Reid", "Jarrod Kahn", "Amir Zait", "Thanumalayan Sankaranarayana Pillai", "Thatcher Ulrich", "Guanyu Wang", "Jan Wassenberg", "Efrat Farkash", "Kiran Yalasangi", "Congchao Wang", "Maria Bauza", "Simon Bucher", "Ting Liu", "Jun Yan", "Gary Leung", "Vikas Sindhwani", "Parker Barnes", "Avi Singh", "Ivan Jurin", "Jichuan Chang", "Niket Kumar Bhumihar", "Sivan Eiger", "Gui Citovsky", "Ben Withbroe", "Zhang Li", "Siyang Xue", "Niccolò Dal Santo", "Georgi Stoyanov", "Yves Raimond", "Steven Zheng", "Yilin Gao", "Vít Listík", "Sławek Kwasiborski", "Rachel Saputro", "Adnan Ozturel", "Ganesh Mallya", "Kushal Majmundar", "Ross West", "Paul Caron", "Jinliang Wei", "Lluis Castrejon", "Sharad Vikram", "Deepak Ramachandran", "Nikhil Dhawan", "Jiho Park", "Sara Smoot", "George van den Driessche", "Yochai Blau", "Chase Malik", "Wei Liang", "Roy Hirsch", "Cicero Nogueira dos Santos", "Eugene Weinstein", "Aäron van den Oord", "Sid Lall", "Nicholas FitzGerald", "Zixuan Jiang", "Xuan Yang", "Dale Webster", "Ali Elqursh", "Aedan Pope", "Georges Rotival", "David Raposo", "Wanzheng Zhu", "Jeff Dean", "Sami Alabed", "Dustin Tran", "Arushi Gupta", "Zach Gleicher", "Jessica Austin", "Edouard Rosseel", "Megh Umekar", "Dipanjan Das", "Yinghao Sun", "Kai Chen", "Karolis Misiunas", "Xiang Zhou", "Yixian Di", "Alyssa Loo", "Josh Newlan", "Bo Li", "Vinay Ramasesh", "Ying Xu", "Alex Chen", "Sudeep Gandhe", "Radu Soricut", "Nikita Gupta", "Shuguang Hu", "Seliem El-Sayed", "Xavier Garcia", "Idan Brusilovsky", "Pu-Chin Chen", "Andrew Bolt", "Lu Huang", "Alex Gurney", "Zhiying Zhang", "Alexander Pritzel", "Jarek Wilkiewicz", "Bryan Seybold", "Bhargav Kanagal Shamanna", "Felix Fischer", "Josef Dean", "Karan Gill", "Ross Mcilroy", "Abhishek Bhowmick", "Jeremy Selier", "Antoine Yang", "Derek Cheng", "Vladimir Magay", "Jie Tan", "Dhriti Varma", "Christian Walder", "Tomas Kocisky", "Ryo Nakashima", "Paul Natsev", "Mike Kwong", "Ionel Gog", "Chiyuan Zhang", "Sander Dieleman", "Thomas Jimma", "Andrey Ryabtsev", "Siddhartha Brahma", "David Steiner", "Dayou Du", "Ante Žužul", "Mislav Žanić", "Mukund Raghavachari", "Willi Gierke", "Zeyu Zheng", "Dessie Petrova", "Yann Dauphin", "Yuchuan Liu", "Ido Kessler", "Steven Hand", "Chris Duvarney", "Seokhwan Kim", "Hyo Lee", "Léonard Hussenot", "Jeffrey Hui", "Josh Smith", "Deepali Jain", "Jiawei Xia", "Gaurav Singh Tomar", "Keyvan Amiri", "Du Phan", "Fabian Fuchs", "Tobias Weyand", "Nenad Tomasev", "Alexandra Cordell", "Xin Liu", "Jonathan Mallinson", "Pankaj Joshi", "Andy Crawford", "Arun Suggala", "Steve Chien", "Nick Fernando", "Mariella Sanchez-Vargas", "Duncan Williams", "Phil Crone", "Xiyang Luo", "Igor Karpov", "Jyn Shan", "Terry Thurk", "Robin Strudel", "Paul Voigtlaender", "Piyush Patil", "Tim Dozat", "Ali Khodaei", "Sahil Singla", "Piotr Ambroszczyk", "Qiyin Wu", "Yifan Chang", "Brian Roark", "Chaitra Hegde", "Tianli Ding", "Angelos Filos", "Zhongru Wu", "André Susano Pinto", "Shuang Liu", "Saarthak Khanna", "Aditya Pandey", "Siobhan Mcloughlin", "Qiujia Li", "Sam Haves", "Allan Zhou", "Elena Buchatskaya", "Isabel Leal", "Peter de Boursac", "Nami Akazawa", "Nina Anderson", "Terry Chen", "Krishna Somandepalli", "Chen Liang", "Sheela Goenka", "Stephanie Winkler", "Alexander Grushetsky", "Yifan Ding", "Jamie Smith", "Fan Ye", "Jordi Pont-Tuset", "Eric Li", "Ruichao Li", "Tomer Golany", "Dawid Wegner", "Tao Jiang", "Omer Barak", "Yuan Shangguan", "Eszter Vértes", "Renee Wong", "Jörg Bornschein", "Alex Tudor", "Michele Bevilacqua", "Tom Schaul", "Ankit Singh Rawat", "Yang Zhao", "Kyriakos Axiotis", "Lei Meng", "Cory McLean", "Jonathan Lai", "Jennifer Beattie", "Nate Kushman", "Yaxin Liu", "Blair Kutzman", "Fiona Lang", "Jingchen Ye", "Praneeth Netrapalli", "Pushkar Mishra", "Myriam Khan", "Megha Goel", "Rob Willoughby", "David Tian", "Honglei Zhuang", "JD Chen", "Zak Tsai", "Tasos Kementsietsidis", "Arjun Khare", "James Keeling", "Keyang Xu", "Nathan Waters", "Florent Altché", "Ashok Popat", "Bhavishya Mittal", "David Saxton", "Dalia El Badawy", "Michael Mathieu", "Zheng Zheng", "Hao Zhou", "Nishant Ranka", "Richard Shin", "Qingnan Duan", "Tim Salimans", "Ioana Mihailescu", "Uri Shaham", "Ming-Wei Chang", "Yannis Assael", "Nishanth Dikkala", "Martin Izzard", "Vincent Cohen-Addad", "Cat Graves", "Vlad Feinberg", "Grace Chung", "DJ Strouse", "Danny Karmon", "Sahand Sharifzadeh", "Zoe Ashwood", "Khiem Pham", "Jon Blanton", "Alex Vasiloff", "Jarred Barber", "Mark Geller", "Aurick Zhou", "Fedir Zubach", "Tzu-Kuo Huang", "Lei Zhang", "Himanshu Gupta", "Matt Young", "Julia Proskurnia", "Ronny Votel", "Valentin Gabeur", "Gabriel Barcik", "Aditya Tripathi", "Hongkun Yu", "Geng Yan", "Beer Changpinyo", "Filip Pavetić", "Amy Coyle", "Yasuhisa Fujii", "Jorge Gonzalez Mendez", "Tianhao Zhou", "Harish Rajamani", "Blake Hechtman", "Eddie Cao", "Da-Cheng Juan", "Yi-Xuan Tan", "Valentin Dalibard", "Yilun Du", "Natalie Clay", "Kaisheng Yao", "Wenhao Jia", "Dimple Vijaykumar", "Yuxiang Zhou", "Xinyi Bai", "Wei-Chih Hung", "Steven Pecht", "Georgi Todorov", "Nikhil Khadke", "Pramod Gupta", "Preethi Lahoti", "Arnaud Autef", "Karthik Duddu", "James Lee-Thorp", "Alexander Bykovsky", "Tautvydas Misiunas", "Sebastian Flennerhag", "Santhosh Thangaraj", "Jed McGiffin", "Zack Nado", "Markus Kunesch", "Andreas Noever", "Amir Hertz", "Marco Liang", "Victor Stone", "Evan Palmer", "Samira Daruki", "Arijit Pramanik", "Siim Põder", "Austin Kyker", "Mina Khan", "Evgeny Sluzhaev", "Marvin Ritter", "Avraham Ruderman", "Wenlei Zhou", "Chirag Nagpal", "Kiran Vodrahalli", "George Necula", "Paul Barham", "Ellie Pavlick", "Jay Hartford", "Izhak Shafran", "Long Zhao", "Maciej Mikuła", "Tom Eccles", "Hidetoshi Shimokawa", "Kanav Garg", "Luke Vilnis", "Hanwen Chen", "Ilia Shumailov", "Kuang-Huei Lee", "Abdelrahman Abdelhamed", "Meiyan Xie", "Vered Cohen", "Ester Hlavnova", "Dan Malkin", "Chawin Sitawarin", "James Lottes", "Pauline Coquinot", "Tianli Yu", "Sandeep Kumar", "Jingwei Zhang", "Aroma Mahendru", "Zafarali Ahmed", "James Martens", "Tao Chen", "Aviel Boag", "Daiyi Peng", "Coline Devin", "Arseniy Klimovskiy", "Mary Phuong", "Danny Vainstein", "Jin Xie", "Bhuvana Ramabhadran", "Nathan Howard", "Xinxin Yu", "Gitartha Goswami", "Jingyu Cui", "Sam Shleifer", "Mario Pinto", "Chih-Kuan Yeh", "Ming-Hsuan Yang", "Sara Javanmardi", "Dan Ethier", "Chace Lee", "Jordi Orbay", "Suyog Kotecha", "Carla Bromberg", "Pete Shaw", "James Thornton", "Adi Gerzi Rosenthal", "Shane Gu", "Matt Thomas", "Ian Gemp", "Aditya Ayyar", "Asahi Ushio", "Aarush Selvan", "Joel Wee", "Chenxi Liu", "Maryam Majzoubi", "Weiren Yu", "Jake Abernethy", "Tyler Liechty", "Renke Pan", "Hoang Nguyen", " Qiong", " Hu", "Sarah Perrin", "Abhinav Arora", "Emily Pitler", "Weiyi Wang", "Kaushik Shivakumar", "Flavien Prost", "Ben Limonchik", "Jing Wang", "Yi Gao", "Timothee Cour", "Shyamal Buch", "Huan Gui", "Maria Ivanova", "Philipp Neubeck", "Kelvin Chan", "Lucy Kim", "Huizhong Chen", "Naman Goyal", "Da-Woon Chung", "Lu Liu", "Yao Su", "Anastasia Petrushkina", "Jiajun Shen", "Armand Joulin", "Yuanzhong Xu", "Stein Xudong Lin", "Yana Kulizhskaya", "Ciprian Chelba", "Shobha Vasudevan", "Eli Collins", "Vasilisa Bashlovkina", "Tony Lu", "Doug Fritz", "Jongbin Park", "Yanqi Zhou", "Chen Su", "Richard Tanburn", "Mikhail Sushkov", "Mitchelle Rasquinha", "Jinning Li", "Jennifer Prendki", "Yiming Li", "Pallavi LV", "Shriya Sharma", "Hen Fitoussi", "Hui Huang", "Andrew Dai", "Phuong Dao", "Mike Burrows", "Henry Prior", "Danfeng Qin", "Golan Pundak", "Lars Lowe Sjoesund", "Art Khurshudov", "Zhenkai Zhu", "Albert Webson", "Elizabeth Kemp", "Tat Tan", "Saurabh Agrawal", "Susie Sargsyan", "Liqun Cheng", "Jim Stephan", "Tom Kwiatkowski", "David Reid", "Arunkumar Byravan", "Assaf Hurwitz Michaely", "Nicolas Heess", "Luowei Zhou", "Sonam Goenka", "Viral Carpenter", "Anselm Levskaya", "Bo Wang", "Reed Roberts", "Rémi Leblond", "Sharat Chikkerur", "Stav Ginzburg", "Max Chang", "Robert Riachi", " Chuqiao", " Xu", "Zalán Borsos", "Michael Pliskin", "Julia Pawar", "Morgane Lustman", "Hannah Kirkwood", "Ankit Anand", "Aditi Chaudhary", "Norbert Kalb", "Kieran Milan", "Sean Augenstein", "Anna Goldie", "Laurel Prince", "Karthik Raman", "Yanhua Sun", "Vivian Xia", "Aaron Cohen", "Zhouyuan Huo", "Josh Camp", "Seher Ellis", "Lukas Zilka", "David Vilar Torres", "Lisa Patel", "Sho Arora", "Betty Chan", "Jonas Adler", "Kareem Ayoub", "Jacky Liang", "Fayaz Jamil", "Jiepu Jiang", "Simon Baumgartner", "Haitian Sun", "Yael Karov", "Yaroslav Akulov", "Hui Zheng", "Irene Cai", "Claudio Fantacci", "James Rubin", "Alex Rav Acha", "Mengchao Wang", "Nina D'Souza", "Rohit Sathyanarayana", "Shengyang Dai", "Simon Rowe", "Andrey Simanovsky", "Omer Goldman", "Yuheng Kuang", "Xiaoyue Pan", "Andrew Rosenberg", "Tania Rojas-Esponda", "Praneet Dutta", "Amy Zeng", "Irina Jurenka", "Greg Farquhar", "Yamini Bansal", "Shariq Iqbal", "Becca Roelofs", "Ga-Young Joung", "Parker Beak", "Changwan Ryu", "Ryan Poplin", "Yan Wu", "Jean-Baptiste Alayrac", "Senaka Buthpitiya", "Olaf Ronneberger", "Caleb Habtegebriel", "Wei Li", "Paul Cavallaro", "Aurora Wei", "Guy Bensky", "Timo Denk", "Harish Ganapathy", "Jeff Stanway", "Pratik Joshi", "Francesco Bertolini", "Jessica Lo", "Olivia Ma", "Zachary Charles", "Geta Sampemane", "Himanshu Sahni", "Xu Chen", "Harry Askham", "David Gaddy", "Peter Young", "Jiewen Tan", "Matan Eyal", "Arthur Bražinskas", "Li Zhong", "Zhichun Wu", "Mark Epstein", "Kai Bailey", "Andrew Hard", "Kamyu Lee", "Sasha Goldshtein", "Alex Ruiz", "Mohammed Badawi", "Matthias Lochbrunner", "JK Kearns", "Ashley Brown", "Fabio Pardo", "Theophane Weber", "Haichuan Yang", "Pan-Pan Jiang", "Berkin Akin", "Zhao Fu", "Marcus Wainwright", "Chi Zou", "Meenu Gaba", "Pierre-Antoine Manzagol", "Wendy Kan", "Yang Song", "Karina Zainullina", "Rui Lin", "Jeongwoo Ko", "Salil Deshmukh", "Apoorv Jindal", "James Svensson", "Divya Tyam", "Heri Zhao", "Christine Kaeser-Chen", "Scott Baird", "Pooya Moradi", "Jamie Hall", "Qiuchen Guo", "Vincent Tsang", "Bowen Liang", "Fernando Pereira", "Suhas Ganesh", "Ivan Korotkov", "Jakub Adamek", "Sridhar Thiagarajan", "Vinh Tran", "Charles Chen", "Chris Tar", "Sanil Jain", "Ishita Dasgupta", "Taylan Bilal", "David Reitter", "Kai Zhao", "Giulia Vezzani", "Yasmin Gehman", "Pulkit Mehta", "Lauren Beltrone", "Xerxes Dotiwalla", "Sergio Guadarrama", "Zaheer Abbas", "Stefani Karp", "Petko Georgiev", "Chun-Sung Ferng", "Marc Brockschmidt", "Liqian Peng", "Christoph Hirnschall", "Vikas Verma", "Yingying Bi", "Ying Xiao", "Avigail Dabush", "Kelvin Xu", "Phil Wallis", "Randall Parker", "Qifei Wang", "Yang Xu", "Ilkin Safarli", "Dinesh Tewari", "Yin Zhang", "Seungyeon Kim", "Andrea Gesmundo", "Mackenzie Thomas", "Sergey Levi", "Ahmed Chowdhury", "Kanishka Rao", "Peter Garst", "Sam Conway-Rahman", "Helen Ran", "Kay McKinney", "Zhisheng Xiao", "Wenhao Yu", "Rohan Agrawal", "Axel Stjerngren", "Catalin Ionescu", "Jingjing Chen", "Vivek Sharma", "Justin Chiu", "Fei Liu", "Ken Franko", "Clayton Sanford", "Xingyu Cai", "Paul Michel", "Sanjay Ganapathy", "Jane Labanowski", "Zachary Garrett", "Ben Vargas", "Sean Sun", "Bryan Gale", "Thomas Buschmann", "Guillaume Desjardins", "Nimesh Ghelani", "Palak Jain", "Mudit Verma", "Chulayuth Asawaroengchai", "Julian Eisenschlos", "Jitendra Harlalka", "Hideto Kazawa", "Don Metzler", "Joshua Howland", "Ying Jian", "Jake Ades", "Viral Shah", "Tynan Gangwani", "Seungji Lee", "Roman Ring", "Steven M. Hernandez", "Dean Reich", "Amer Sinha", "Ashutosh Sathe", "Joe Kovac", "Ashleah Gill", "Ajay Kannan", "Andrea D'olimpio", "Martin Sevenich", "Jay Whang", "Been Kim", "Khe Chai Sim", "Jilin Chen", "Jiageng Zhang", "Shuba Lall", "Yossi Matias", "Bill Jia", "Abe Friesen", "Sara Nasso", "Ashish Thapliyal", "Bryan Perozzi", "Ting Yu", "Anna Shekhawat", "Safeen Huda", "Peter Grabowski", "Eric Wang", "Ashwin Sreevatsa", "Hilal Dib", "Mehadi Hassen", "Parker Schuh", "Vedrana Milutinovic", "Chris Welty", "Michael Quinn", "Ali Shah", "Bangju Wang", "Gabe Barth-Maron", "Justin Frye", "Natalie Axelsson", "Tao Zhu", "Yukun Ma", "Irene Giannoumis", "Hanie Sedghi", "Chang Ye", "Yi Luan", "Kevin Aydin", "Bilva Chandra", "Vivek Sampathkumar", "Ronny Huang", "Victor Lavrenko", "Ahmed Eleryan", "Zhi Hong", "Steven Hansen", "Sara Mc Carthy", "Bidisha Samanta", "Domagoj Ćevid", "Xin Wang", "Fangtao Li", "Michael Voznesensky", "Matt Hoffman", "Andreas Terzis", "Vikash Sehwag", "Gil Fidel", "Luheng He", "Mu Cai", "Yanzhang He", "Alex Feng", "Martin Nikoltchev", "Samrat Phatale", "Jason Chase", "Rory Lawton", "Ming Zhang", "Tom Ouyang", "Manuel Tragut", "Mehdi Hafezi Manshadi", "Arjun Narayanan", "Jiaming Shen", "Xu Gao", "Tolga Bolukbasi", "Nick Roy", "Xin Li", "Daniel Golovin", "Liviu Panait", "Zhen Qin", "Guangxing Han", "Thomas Anthony", "Sneha Kudugunta", "Viorica Patraucean", "Aniket Ray", "Xinyun Chen", "Xiaochen Yang", "Tanuj Bhatia", "Pranav Talluri", "Alex Morris", "Andrija Ražnatović", "Bethanie Brownfield", "James An", "Sheng Peng", "Patrick Kane", "Ce Zheng", "Nico Duduta", "Joshua Kessinger", "James Noraky", "Siqi Liu", "Keran Rong", "Petar Veličković", "Keith Rush", "Alex Goldin", "Fanny Wei", "Shiva Mohan Reddy Garlapati", "Caroline Pantofaru", "Okwan Kwon", "Jianmo Ni", "Eric Noland", "Julia Di Trapani", "Françoise Beaufays", "Abhijit Guha Roy", "Yinlam Chow", "Aybuke Turker", "Geoffrey Cideron", "Lantao Mei", "Jon Clark", "Qingyun Dou", "Matko Bošnjak", "Ralph Leith", "Yuqing Du", "Amir Yazdanbakhsh", "Milad Nasr", "Chester Kwak", "Suraj Satishkumar Sheth", "Alex Kaskasoli", "Ankesh Anand", "Balaji Lakshminarayanan", "Sammy Jerome", "David Bieber", "Chun-Te Chu", "Alexandre Senges", "Tianxiao Shen", "Mukund Sridhar", "Ndaba Ndebele", "Benjamin Beyret", "Shakir Mohamed", "Mia Chen", "Markus Freitag", "Jiaxian Guo", "Luyang Liu", "Paul Roit", "Heng Chen", "Shen Yan", "Tom Stone", "JD Co-Reyes", "Jeremy Cole", "Salvatore Scellato", "Shekoofeh Azizi", "Hadi Hashemi", "Alicia Jin", "Anand Iyer", "Marcella Valentine", "András György", "Arun Ahuja", "Daniel Hernandez Diaz", "Chen-Yu Lee", "Nathan Clement", "Weize Kong", "Drew Garmon", "Ishaan Watts", "Kush Bhatia", "Khyatti Gupta", "Matt Miecnikowski", "Hugo Vallet", "Ankur Taly", "Edward Loper", "Saket Joshi", "James Atwood", "Jo Chick", "Mark Collier", "Fotis Iliopoulos", "Ryan Trostle", "Beliz Gunel", "Ramiro Leal-Cavazos", "Arnar Mar Hrafnkelsson", "Michael Guzman", "Xiaoen Ju", "Andy Forbes", "Jesse Emond", "Kushal Chauhan", "Ben Caine", "Li Xiao", "Wenjun Zeng", "Alexandre Moufarek", "Daniel Murphy", "Maya Meng", "Nitish Gupta", "Felix Riedel", "Anil Das", "Elijah Lawal", "Shashi Narayan", "Tiberiu Sosea", "James Swirhun", "Linda Friso", "Behnam Neyshabur", "Jing Lu", "Sertan Girgin", "Michael Wunder", "Edouard Yvinec", "Aroonalok Pyne", "Victor Carbune", "Shruti Rijhwani", "Yang Guo", "Tulsee Doshi", "Anton Briukhov", "Max Bain", "Ayal Hitron", "Xuanhui Wang", "Ashish Gupta", "Ke Chen", "Cosmo Du", "Weiyang Zhang", "Dhruv Shah", "Arjun Akula", "Max Dylla", "Ashyana Kachra", "Weicheng Kuo", "Tingting Zou", "Lily Wang", "Luyao Xu", "Jifan Zhu", "Justin Snyder", "Sachit Menon", "Orhan Firat", "Igor Mordatch", "Yuan Yuan", "Natalia Ponomareva", "Rory Blevins", "Lawrence Moore", "Weijun Wang", "Phil Chen", "Martin Scholz", "Artur Dwornik", "Jason Lin", "Sicheng Li", "Diego Antognini", "Te I", "Xiaodan Song", "Matt Miller", "Uday Kalra", "Adam Raveret", "Oscar Akerlund", "Felix Wu", "Andrew Nystrom", "Namrata Godbole", "Tianqi Liu", "Hannah DeBalsi", "Jewel Zhao", "Buhuang Liu", "Avi Caciularu", "Lauren Lax", "Urvashi Khandelwal", "Victoria Langston", "Eric Bailey", "Silvio Lattanzi", "Yufei Wang", "Neel Kovelamudi", "Sneha Mondal", "Guru Guruganesh", "Nan Hua", "Ofir Roval", "Paweł Wesołowski", "Rishikesh Ingale", "Jonathan Halcrow", "Tim Sohn", "Christof Angermueller", "Bahram Raad", "Eli Stickgold", "Eva Lu", "Alec Kosik", "Jing Xie", "Timothy Lillicrap", "Austin Huang", "Lydia Lihui Zhang", "Dominik Paulus", "Clement Farabet", "Alex Wertheim", "Bing Wang", "Rishabh Joshi", "Chu-ling Ko", "Yonghui Wu", "Shubham Agrawal", "Lily Lin", "XiangHai Sheng", "Peter Sung", "Tyler Breland-King", "Christina Butterfield", "Swapnil Gawde", "Sumeet Singh", "Qiao Zhang", "Raj Apte", "Shilpa Shetty", "Adrian Hutter", "Tao Li", "Elizabeth Salesky", "Federico Lebron", "Jonni Kanerva", "Michela Paganini", "Arthur Nguyen", "Rohith Vallu", "Jan-Thorsten Peter", "Sarmishta Velury", "David Kao", "Jay Hoover", "Anna Bortsova", "Colton Bishop", "Shoshana Jakobovits", "Alessandro Agostini", "Alekh Agarwal", "Chang Liu", "Charles Kwong", "Sasan Tavakkol", "Ioana Bica", "Alex Greve", "Anirudh GP", "Jake Marcus", "Le Hou", "Tom Duerig", "Rivka Moroshko", "Dave Lacey", "Andy Davis", "Julien Amelot", "Guohui Wang", "Frank Kim", "Theofilos Strinopoulos", "Hui Wan", "Charline Le Lan", "Shankar Krishnan", "Haotian Tang", "Peter Humphreys", "Junwen Bai", "Idan Heimlich Shtacher", "Diego Machado", "Chenxi Pang", "Ken Burke", "Dangyi Liu", "Renga Aravamudhan", "Yue Song", "Ed Hirst", "Abhimanyu Singh", "Brendan Jou", "Liang Bai", "Francesco Piccinno", "Chuyuan Kelly Fu", "Robin Alazard", "Barak Meiri", "Daniel Winter", "Charlie Chen", "Mingda Zhang", "Jens Heitkaemper", "John Lambert", "Jinhyuk Lee", "Alexander Frömmgen", "Sergey Rogulenko", "Pranav Nair", "Paul Niemczyk", "Anton Bulyenov", "Bibo Xu", "Hadar Shemtov", "Morteza Zadimoghaddam", "Serge Toropov", "Mateo Wirth", "Hanjun Dai", "Sreenivas Gollapudi", "Daniel Zheng", "Alex Kurakin", "Chansoo Lee", "Kalesha Bullard", "Nicolas Serrano", "Ivana Balazevic", "Yang Li", "Johan Schalkwyk", "Mark Murphy", "Mingyang Zhang", "Kevin Sequeira", "Romina Datta", "Nishant Agrawal", "Charles Sutton", "Nithya Attaluri", "Mencher Chiang", "Wael Farhan", "Gregory Thornton", "Kate Lin", "Travis Choma", "Hung Nguyen", "Kingshuk Dasgupta", "Dirk Robinson", "Iulia Comşa", "Michael Riley", "Arjun Pillai", "Basil Mustafa", "Ben Golan", "Amir Zandieh", "Jean-Baptiste Lespiau", "Billy Porter", "David Ross", "Sujeevan Rajayogam", "Mohit Agarwal", "Subhashini Venugopalan", "Bobak Shahriari", "Qiqi Yan", "Hao Xu", "Taylor Tobin", "Pavel Dubov", "Hongzhi Shi", "Adrià Recasens", "Anton Kovsharov", "Sebastian Borgeaud", "Lucio Dery", "Shanthal Vasanth", "Elena Gribovskaya", "Linhai Qiu", "Mahdis Mahdieh", "Wojtek Skut", "Elizabeth Nielsen", "CJ Zheng", "Adams Yu", "Carrie Grimes Bostock", "Shaleen Gupta", "Aaron Archer", "Chris Rawles", "Elinor Davies", "Alexey Svyatkovskiy", "Tomy Tsai", "Yoni Halpern", "Christian Reisswig", "Bartek Wydrowski", "Bo Chang", "Joan Puigcerver", "Mor Hazan Taege", "Jian Li", "Eva Schnider", "Xinjian Li", "Dragos Dena", "Yunhan Xu", "Umesh Telang", "Tianze Shi", "Heiga Zen", "Kyle Kastner", "Yeongil Ko", "Neesha Subramaniam", "Aviral Kumar", "Pete Blois", "Zhuyun Dai", "John Wieting", "Yifeng Lu", "Yoel Zeldes", "Tian Xie", "Anja Hauth", "Alexandru Ţifrea", "Yuqi Li", "Sam El-Husseini", "Dan Abolafia", "Howard Zhou", "Wen Ding", "Sahra Ghalebikesabi", "Carlos Guía", "Andrii Maksai", "Ágoston Weisz", "Sercan Arik", "Nick Sukhanov", "Aga Świetlik", "Xuhui Jia", "Luo Yu", "Weiyue Wang", "Mark Brand", "Dawn Bloxwich", "Sean Kirmani", "Zhe Chen", "Alec Go", "Pablo Sprechmann", "Nithish Kannen", "Alen Carin", "Paramjit Sandhu", "Isabel Edkins", "Leslie Nooteboom", "Jai Gupta", "Loren Maggiore", "Javad Azizi", "Yael Pritch", "Pengcheng Yin", "Mansi Gupta", "Danny Tarlow", "Duncan Smith", "Desi Ivanov", "Mohammad Babaeizadeh", "Ankita Goel", "Satish Kambala", "Grace Chu", "Matej Kastelic", "Michelle Liu", "Hagen Soltau", "Austin Stone", "Shivani Agrawal", "Min Kim", "Kedar Soparkar", "Srinivas Tadepalli", "Oskar Bunyan", "Rachel Soh", "Arvind Kannan", "DY Kim", "Blake JianHang Chen", "Afief Halumi", "Sudeshna Roy", "Yulong Wang", "Olcan Sercinoglu", "Gena Gibson", "Sijal Bhatnagar", "Motoki Sano", "Daniel von Dincklage", "Qingchun Ren", "Blagoj Mitrevski", "Mirek Olšák", "Jennifer She", "Carl Doersch", " Jilei", " Wang", "Bingyuan Liu", "Qijun Tan", "Tamar Yakar", "Tris Warkentin", "Alex Ramirez", "Carl Lebsack", "Josh Dillon", "Rajiv Mathews", "Tom Cobley", "Zelin Wu", "Zhuoyuan Chen", "Jon Simon", "Swaroop Nath", "Tara Sainath", "Alexei Bendebury", "Ryan Julian", "Bharath Mankalale", "Daria Ćurko", "Paulo Zacchello", "Adam R. Brown", "Kiranbir Sodhia", "Heidi Howard", "Sergi Caelles", "Abhinav Gupta", "Gareth Evans", "Anna Bulanova", "Lesley Katzen", "Roman Goldenberg", "Anton Tsitsulin", "Joe Stanton", "Benoit Schillings", "Vitaly Kovalev", "Corey Fry", "Rushin Shah", "Kuo Lin", "Shyam Upadhyay", "Cheng Li", "Soroush Radpour", "Marcello Maggioni", "Jing Xiong", "Lukas Haas", "Jenny Brennan", "Aishwarya Kamath", "Nikolay Savinov", "Arsha Nagrani", "Trevor Yacovone", "Ryan Kappedal", "Kostas Andriopoulos", "Li Lao", "YaGuang Li", "Grigory Rozhdestvenskiy", "Kazuma Hashimoto", "Andrew Audibert", "Sophia Austin", "Daniel Rodriguez", "Anian Ruoss", "Garrett Honke", "Deep Karkhanis", "Xi Xiong", "Qing Wei", "James Huang", "Zhaoqi Leng", "Vittal Premachandran", "Stan Bileschi", "Georgios Evangelopoulos", "Thomas Mensink", "Jay Pavagadhi", "Denis Teplyashin", "Paul Chang", "Linting Xue", "Garrett Tanzer", "Sally Goldman", "Kaushal Patel", "Shixin Li", "Jeremy Wiesner", "Ivy Zheng", "Ian Stewart-Binks", "Jie Han", "Zhi Li", "Liangchen Luo", "Karel Lenc", "Mario Lučić", "Fuzhao Xue", "Ryan Mullins", "Alexey Guseynov", "Chung-Ching Chang", "Isaac Galatzer-Levy", "Adam Zhang", "Garrett Bingham", "Grace Hu", "Ale Hartman", "Yue Ma", "Jordan Griffith", "Alex Irpan", "Carey Radebaugh", "Summer Yue", "Lijie Fan", "Victor Ungureanu", "Christina Sorokin", "Hannah Teufel", "Peiran Li", "Rohan Anil", "Dimitris Paparas", "Todd Wang", "Chu-Cheng Lin", "Hui Peng", "Megan Shum", "Goran Petrovic", "Demetra Brady", "Richard Nguyen", "Klaus Macherey", "Zhihao Li", "Harman Singh", "Madhavi Yenugula", "Mariko Iinuma", "Xinyi Chen", "Kavya Kopparapu", "Alexey Stern", "Shachi Dave", "Chandu Thekkath", "Florence Perot", "Anurag Kumar", "Fangda Li", "Yang Xiao", "Matthew Bilotti", "Mohammad Hossein Bateni", "Isaac Noble", "Lisa Lee", "Amelio Vázquez-Reina", "Julian Salazar", "Xiaomeng Yang", "Boyu Wang", "Ela Gruzewska", "Anand Rao", "Sindhu Raghuram", "Zheng Xu", "Eyal Ben-David", "Jieru Mei", "Sid Dalmia", "Zhaoyi Zhang", "Yuchen Liu", "Gagan Bansal", "Helena Pankov", "Steven Schwarcz", "Andrea Burns", "Christine Chan", "Sumit Sanghai", "Ricky Liang", "Ethan Liang", "Antoine He", "Amy Stuart", "Arun Narayanan", "Yukun Zhu", "Christian Frank", "Bahar Fatemi", "Amit Sabne", "Oran Lang", "Indro Bhattacharya", "Shane Settle", "Maria Wang", "Brendan McMahan", "Andrea Tacchetti", "Livio Baldini Soares", "Majid Hadian", "Serkan Cabi", "Timothy Chung", "Nikita Putikhin", "Gang Li", "Jeremy Chen", "Austin Tarango", "Henryk Michalewski", "Mehran Kazemi", "Hussain Masoom", "Hila Sheftel", "Rakesh Shivanna", "Archita Vadali", "Ramona Comanescu", "Doug Reid", "Joss Moore", "Arvind Neelakantan", "Michaël Sander", "Jonathan Herzig", "Aviv Rosenberg", "Mostafa Dehghani", "JD Choi", "Michael Fink", "Reid Hayes", "Eric Ge", "Shitao Weng", "Chia-Hua Ho", "John Karro", "Kalpesh Krishna", "Lam Nguyen Thiet", "Amy Skerry-Ryan", "Daniel Eppens", "Marco Andreetto", "Navin Sarma", "Silvano Bonacina", "Burcu Karagol Ayan", "Megha Nawhal", "Zhihao Shan", "Mike Dusenberry", "Shantanu Thakoor", "Sagar Gubbi", "Duc Dung Nguyen", "Reut Tsarfaty", "Samuel Albanie", "Jovana Mitrović", "Meet Gandhi", "Bo-Juen Chen", "Alessandro Epasto", "Georgi Stephanov", "Ye Jin", "Samuel Gehman", "Aida Amini", "Jack Weber", "Feryal Behbahani", "Shawn Xu", "Miltos Allamanis", "Xi Chen", "Myle Ott", "Claire Sha", "Michal Jastrzebski", "Hang Qi", "David Greene", "Xinyi Wu", "Abodunrinwa Toki", "Daniel Vlasic", "Jane Shapiro", "Ragha Kotikalapudi", "Zhe Shen", "Takaaki Saeki", "Sirui Xie", "Albin Cassirer", "Shikhar Bharadwaj", "Tatsuya Kiyono", "Srinadh Bhojanapalli", "Elan Rosenfeld", "Sam Ritter", "Jieming Mao", "João Gabriel Oliveira", "Zoltan Egyed", "Bernd Bandemer", "Emilio Parisotto", "Keisuke Kinoshita", "Juliette Pluto", "Petros Maniatis", "Steve Li", "Yaohui Guo", "Golnaz Ghiasi", "Jean Tarbouriech", "Srimon Chatterjee", "Julie Jin", " Katrina", " Xu", "Jennimaria Palomaki", "Séb Arnold", "Madhavi Sewak", "Federico Piccinini", "Mohit Sharma", "Ben Albrecht", "Sean Purser-haskell", "Ashwin Vaswani", "Chongyan Chen", "Matheus Wisniewski", "Qin Cao", "John Aslanides", "Nguyet Minh Phu", "Maximilian Sieb", "Lauren Agubuzu", "Anne Zheng", "Daniel Sohn", "Marco Selvi", "Anders Andreassen", "Krishan Subudhi", "Prem Eruvbetine", "Oliver Woodman", "Tomas Mery", "Sebastian Krause", "Xiaoqi Ren", "Xiao Ma", "Jincheng Luo", "Dawn Chen", "Wei Fan", "Henry Griffiths", "Christian Schuler", "Alice Li", "Shujian Zhang", "Jean-Michel Sarr", "Shixin Luo", "Riccardo Patana", "Matthew Watson", "Dani Naboulsi", "Michael Collins", "Sailesh Sidhwani", "Emiel Hoogeboom", "Sharon Silver", "Emily Caveness", "Xiaokai Zhao", "Mikel Rodriguez", "Maxine Deines", "Libin Bai", "Patrick Griffin", "Marco Tagliasacchi", "Emily Xue", "Spandana Raj Babbula", "Bo Pang", "Nan Ding", "Gloria Shen", "Elijah Peake", "Remi Crocker", "Shubha Srinivas Raghvendra", "Danny Swisher", "Woohyun Han", "Richa Singh", "Ling Wu", "Vladimir Pchelin", "Tsendsuren Munkhdalai", "Dana Alon", "Geoff Bacon", "Efren Robles", "Jannis Bulian", "Melvin Johnson", "George Powell", "Felipe Tiengo Ferreira", "Yaoyiran Li", "Frederik Benzing", "Mihajlo Velimirović", "Hubert Soyer", "William Kong", " Tony", " Nguyên", "Zhen Yang", "Jeremiah Liu", "Joost van Amersfoort", "Daniel Gillick", "Baochen Sun", "Nathalie Rauschmayr", "Katie Zhang", "Serena Zhan", "Tao Zhou", "Alexey Frolov", "Chengrun Yang", "Denis Vnukov", "Louis Rouillard", "Hongji Li", "Amol Mandhane", "Nova Fallen", "Rajesh Venkataraman", "Clara Huiyi Hu", "Jennifer Brennan", "Jenny Lee", "Jerry Chang", "Martin Sundermeyer", "Zhufeng Pan", "Rosemary Ke", "Simon Tong", "Alex Fabrikant", "William Bono", "Jindong Gu", "Ryan Foley", "Yiran Mao", "Manolis Delakis", "Dhruva Bhaswar", "Roy Frostig", "Nick Li", "Avital Zipori", "Cath Hope", "Olga Kozlova", "Swaroop Mishra", "Josip Djolonga", "Craig Schiff", "Majd Al Merey", "Eleftheria Briakou", "Peter Morgan", "Andy Wan", "Avinatan Hassidim", "RJ Skerry-Ryan", "Kuntal Sengupta", "Mary Jasarevic", "Praveen Kallakuri", "Paige Kunkle", "Hannah Brennan", "Tom Lieber", "Hassan Mansoor", "Julian Walker", "Bing Zhang", "Annie Xie", "Goran Žužić", "Adaeze Chukwuka", "Alex Druinsky", "Donghyun Cho", "Rui Yao", "Ferjad Naeem", "Shiraz Butt", "Eunyoung Kim", "Zhipeng Jia", "Mandy Jordan", "Adam Lelkes", "Mark Kurzeja", "Sophie Wang", "James Zhao", "Andrew Over", "Abhishek Chakladar", "Marcel Prasetya", "Neha Jha", "Sriram Ganapathy", "Yale Cong", "Prakash Shroff", "Carl Saroufim", "Sobhan Miryoosefi", "Mohamed Hammad", "Tajwar Nasir", "Weijuan Xi", "Yang Gao", "Young Maeng", "Ben Hora", "Chin-Yi Cheng", "Parisa Haghani", "Yoad Lewenberg", "Caden Lu", "Martin Matysiak", "Naina Raisinghani", "Huiyu Wang", "Lexi Baugher", "Rahul Sukthankar", "Minh Giang", "John Schultz", "Noah Fiedel", "Minmin Chen", "Cheng-Chun Lee", "Tapomay Dey", "Hao Zheng", "Shachi Paul", "Celine Smith", "Andy Ly", "Yicheng Wang", "Rishabh Bansal", "Bartek Perz", "Susanna Ricco", "Stasha Blank", "Vaishakh Keshava", "Deepak Sharma", "Marvin Chow", "Kunal Lad", "Komal Jalan", "Simon Osindero", "Craig Swanson", "Jacob Scott", "Anastasija Ilić", "Xiaowei Li", "Siddhartha Reddy Jonnalagadda", "Afzal Shama Soudagar", "Yan Xiong", "Bat-Orgil Batsaikhan", "Daniel Jarrett", "Naveen Kumar", "Maulik Shah", "Matt Lawlor", "Austin Waters", "Mark Graham", "Rhys May", "Sabela Ramos", "Sandra Lefdal", "Zeynep Cankara", "Nacho Cano", "Brendan O'Donoghue", "Jed Borovik", "Frederick Liu", "Jordan Grimstad", "Mahmoud Alnahlawi", "Katerina Tsihlas", "Tom Hudson", "Nikolai Grigorev", "Yiling Jia", "Terry Huang", "Tobenna Peter Igwe", "Sergei Lebedev", "Xiaodan Tang", "Igor Krivokon", "Frankie Garcia", "Melissa Tan", "Eric Jia", "Peter Stys", "Shikhar Vashishth", "Yu Liang", "Balaji Venkatraman", "Chenjie Gu", "Anastasios Kementsietsidis", "Chen Zhu", "Junehyuk Jung", "Yunfei Bai", "Mohammad Javad Hosseini", "Faruk Ahmed", "Aditya Gupta", "Xin Yuan", "Shereen Ashraf", "Shitij Nigam", "Gautam Vasudevan", "Pranjal Awasthi", "Adi Mayrav Gilady", "Zelda Mariet", "Ramy Eskander", "Haiguang Li", "Hexiang Hu", "Guillermo Garrido", "Philippe Schlattner", "George Zhang", "Rohun Saxena", "Petar Dević", "Kritika Muralidharan", "Ashwin Murthy", "Yiqian Zhou", "Min Choi", "Arissa Wongpanich", "Zhengdong Wang", "Premal Shah", "Yuntao Xu", "Yiling Huang", "Stephen Spencer", "Alice Chen", "James Cohan", "Junjie Wang", "Jonathan Tompson", "Junru Wu", "Ruba Haroun", "Haiqiong Li", "Blanca Huergo", "Fan Yang", "Tongxin Yin", "James Wendt", "Michael Bendersky", "Rahma Chaabouni", "Javier Snaider", "Johan Ferret", "Abhishek Jindal", "Tara Thompson", "Andrew Xue", "Will Bishop", "Shubham Milind Phal", "Archit Sharma", "Yunhsuan Sung", "Prabakar Radhakrishnan", "Mo Shomrat", "Reeve Ingle", "Roopali Vij", "Justin Gilmer", "Mihai Dorin Istin", "Sam Sobell", "Yang Lu", "Emily Nottage", "Dorsa Sadigh", "Jeremiah Willcock", "Tingnan Zhang", "Steve Xu", "Sasha Brown", "Katherine Lee", "Gary Wang", "Yun Zhu", "Yi Tay", "Cheolmin Kim", "Audrey Gutierrez", "Abhanshu Sharma", "Yongqin Xian", "Sungyong Seo", "Claire Cui", "Elena Pochernina", "Cip Baetu", "Krzysztof Jastrzębski", "Mimi Ly", "Mohamed Elhawaty", "Dan Suh", "Eren Sezener", "Pidong Wang", "Nancy Yuen", "George Tucker", "Jiahao Cai", "Zuguang Yang", "Cindy Wang", "Alex Muzio", "Hai Qian", "Jae Yoo", "Derek Lockhart", "Kevin R. McKee", "Mandy Guo", "Malika Mehrotra", "Artur Mendonça", "Sanket Vaibhav Mehta", "Sherry Ben", "Chetan Tekur", "Jiaqi Mu", "Muye Zhu", "Victoria Krakovna", "Hongrae Lee", "AJ Maschinot", "Sébastien Cevey", "HyunJeong Choe", "Aijun Bai", "Hansa Srinivasan", "Derek Gasaway", "Nick Young", "Patrick Siegler", "Dan Holtmann-Rice", "Vihari Piratla", "Kate Baumli", "Roey Yogev", "Alex Hofer", "Hado van Hasselt", "Svetlana Grant", "Yuri Chervonyi", "David Silver", "Andrew Hogue", "Ayushi Agarwal", "Kathie Wang", "Preeti Singh", "Four Flynn", "Josh Lipschultz", "Robert David", "Lizzetth Bellot", "Yao-Yuan Yang", "Long Le", "Filippo Graziano", "Kate Olszewska", "Kevin Hui", "Akanksha Maurya", "Nikos Parotsidis", "Weijie Chen", "Tayo Oguntebi", "Joe Kelley", "Anirudh Baddepudi", "Johannes Mauerer", "Gregory Shaw", "Alex Siegman", "Lin Yang", "Shravya Shetty", "Subhrajit Roy", "Yunting Song", "Wojciech Stokowiec", "Ryan Burnell", "Omkar Savant", "Robert Busa-Fekete", "Jin Miao", "Samrat Ghosh", "Liam MacDermed", "Phillip Lippe", "Mikhail Dektiarev", "Zach Behrman", "Fabian Mentzer", "Kelvin Nguyen", "Meng Wei", "Siddharth Verma", "Chris Knutsen", "Sudeep Dasari", "Zhipeng Yan", "Petr Mitrichev", "Xingyu Wang", "Virat Shejwalkar", "Jacob Austin", "Srinivas Sunkara", "Navneet Potti", "Yan Virin", "Christian Wright", "Gaël Liu", "Oriana Riva", "Etienne Pot", "Greg Kochanski", "Quoc Le", "Gargi Balasubramaniam", "Arka Dhar", "Yuguo Liao", "Adam Bloniarz", "Divyansh Shukla", "Elizabeth Cole", "Jong Lee", "Sheng Zhang", "Sushant Kafle", "Siddharth Vashishtha", "Parsa Mahmoudieh", "Grace Chen", "Raphael Hoffmann", "Pranesh Srinivasan", "Agustin Dal Lago", "Yoav Ben Shalom", "Zi Wang", "Michael Elabd", "Anuj Sharma", "Junhyuk Oh", "Suraj Kothawade", "Maigo Le", "Marianne Monteiro", "Shentao Yang", "Kaiz Alarakyia", "Robert Geirhos", "Diana Mincu", "Håvard Garnes", "Hayato Kobayashi", "Soroosh Mariooryad", "Kacper Krasowiak", " Zhixin", " Lai", "Shibl Mourad", "Mingqiu Wang", "Fan Bu", "Ophir Aharoni", "Guanjie Chen", "Abhimanyu Goyal", "Vadim Zubov", "Ankur Bapna", "Elahe Dabir", "Nisarg Kothari", "Kay Lamerigts", "Nicola De Cao", "Jeremy Shar", "Christopher Yew", "Nitish Kulkarni", "Dre Mahaarachchi", "Mandar Joshi", "Zhenhai Zhu", "Jared Lichtarge", "Yichao Zhou", "Hannah Muckenhirn", "Vittorio Selo", "Oriol Vinyals", "Peter Chen", "Anthony Brohan", "Vaibhav Mehta", "Sarah Cogan", "Ruth Wang", "Ty Geri", "Wei-Jen Ko", "Wei Chen", "Fabio Viola", "Keshav Shivam", "Lisa Wang", "Madeleine Clare Elish", "Raluca Ada Popa", "Sébastien Pereira", "Jianqiao Liu", "Raphael Koster", "Donnie Kim", "Gufeng Zhang", "Sayna Ebrahimi", "Partha Talukdar", "Yanyan Zheng", "Petra Poklukar", "Ales Mikhalap", "Dale Johnson", "Anitha Vijayakumar", "Mark Omernick", "Matt Dibb", "Ayush Dubey", "Qiong Hu", "Apurv Suman", "Vaibhav Aggarwal", "Ilya Kornakov", "Fei Xia", "Wing Lowe", "Alexey Kolganov", "Ted Xiao", "Vitaly Nikolaev", "Steven Hemingray", "Bonnie Li", "Joana Iljazi", "Mikołaj Rybiński", "Ballie Sandhu", "Peggy Lu", "Thang Luong", "Rodolphe Jenatton", "Vineetha Govindaraj", " Hui", " Li", "Gabriel Dulac-Arnold", "Wonpyo Park", "Henry Wang", "Abhinit Modi", "Jean Pouget-Abadie", "Kristina Greller", "Rahul Gupta", "Robert Berry", "Prajit Ramachandran", "Jinyu Xie", "Liam McCafferty", "Jianling Wang", "Kilol Gupta", "Hyeontaek Lim", "Blaž Bratanič", "Andy Brock", "Ilia Akolzin", "Jim Sproch", "Dan Karliner", "Duhyeon Kim", "Adrian Goedeckemeyer", "Noam Shazeer", "Cordelia Schmid", "Daniele Calandriello", "Parul Bhatia", "Krzysztof Choromanski", "Ceslee Montgomery", "Dheeru Dua", "Ana Ramalho", "Helen King", "Yue Gao", "Lynn Nguyen", "David Lindner", "Divya Pitta", "Oleaser Johnson", "Khalid Salama", "Diego Ardila", "Michael Han", "Erin Farnese", "Seth Odoom", "Ziyue Wang", "Xiangzhuo Ding", "Norman Rink", "Ray Smith", "Harshal Tushar Lehri", "Eden Cohen", "Neera Vats", "Tong He", "Parthasarathy Gopavarapu", "Adam Paszke", "Miteyan Patel", "Wouter Van Gansbeke", "Lucia Loher", "Luis Castro", "Maria Voitovich", "Tamara von Glehn", "Nelson George", "Simon Niklaus", "Zach Eaton-Rosen", "Nemanja Rakićević", "Erik Jue", "Sagi Perel", "Carrie Zhang", "Yuval Bahat", "Angéline Pouget", "Zhi Xing", "Fantine Huot", "Ashish Shenoy", "Taylor Bos", "Vincent Coriou", "Bryan Richter", "Natasha Noy", "Yaqing Wang", "Santiago Ontanon", "Siyang Qin", "Gleb Makarchuk", "Demis Hassabis", "Zhuowan Li", "Mandar Sharma", "Kumaran Venkatesan", "Iurii Kemaev", "Roxanne Daniel", "Shiyu Huang", "Saloni Shah", "Octavio Ponce", " Warren", " Chen", "Manaal Faruqui", "Jialin Wu", "Slavica Andačić", "Szabolcs Payrits", "Daniel McDuff", "Tom Hume", "Yuan Cao", "MH Tessler", "Qingze Wang", "Yinan Wang", "Ivor Rendulic", "Eirikur Agustsson", "Matthew Johnson", "Tanya Lando", "Andrew Howard", "Sri Gayatri Sundara Padmanabhan", "Mayank Daswani", "Andrea Banino", "Michael Kilgore", "Jonathan Heek", "Ziwei Ji", "Alvaro Caceres", "Conglong Li", "Nora Kassner", "Alexey Vlaskin", "Zeyu Liu", "Alex Grills", "Yanhan Hou", "Roykrong Sukkerd", "Gowoon Cheon", "Nishita Shetty", "Larisa Markeeva", "Piotr Stanczyk", "Tejas Iyer", "Yuan Gong", "Shawn Gao", "Keerthana Gopalakrishnan", "Tim Blyth", "Malcolm Reynolds", "Avishkar Bhoopchand", "Misha Bilenko", "Dero Gharibian", "Vicky Zayats", "Aleksandra Faust", "Abhinav Singh", "Min Ma", "Hongyang Jiao", "Sudheendra Vijayanarasimhan", "Lora Aroyo", "Vikas Yadav", "Sarah Chakera", "Ashwin Kakarla", "Vilobh Meshram", "Karol Gregor", "Gabriela Botea", "Evan Senter", "Dawei Jia", "Geza Kovacs", "Neha Sharma", "Sebastien Baur", "Kai Kang", "Yifan He", "Lin Zhuo", "Marija Kostelac", "Itay Laish", "Songyou Peng", "Louis O'Bryan", "Daniel Kasenberg", "Girish Ramchandra Rao", "Edouard Leurent", "Biao Zhang", "Sage Stevens", "Ana Salazar", "Ye Zhang", "Ivan Lobov", "Jake Walker", "Allen Porter", "Morgan Redshaw", "Han Ke", "Abhishek Rao", "Alex Lee", "Hoi Lam", "Michael Moffitt", "Jaeyoun Kim", "Siyuan Qiao", "Terry Koo", "Robert Dadashi", "Xinying Song", "Mukund Sundararajan", "Peng Xu", "Chizu Kawamoto", "Yan Zhong", "Clara Barbu", "Apoorv Reddy", "Mauro Verzetti", "Leon Li", "George Papamakarios", "Hanna Klimczak-Plucińska", "Mary Cassin", "Koray Kavukcuoglu", "Rigel Swavely", "Alain Vaucher", "Jeffrey Zhao", "Ross Hemsley", "Michael Tschannen", "Heming Ge", "Gaurav Menghani", "Yang Yu", "Natalie Ha", "Wei He", "Xiao Wu", "Maggie Song", "Rachel Sterneck", "Stefan Zinke", "Dan A. Calian", "Annie Marsden", "Alejandro Cruzado Ruiz", "Matteo Hessel", "Almog Gueta", "Benjamin Lee", "Brian Farris", "Manish Gupta", "Yunjie Li", "Mohammad Saleh", "Vedant Misra", "Kefan Xiao", "Piermaria Mendolicchio", "Gavin Buttimore", "Varvara Krayvanova", "Nigamaa Nayakanti", "Matthew Wiethoff", "Yash Pande", "Azalia Mirhoseini", "Ni Lao", "Jasmine Liu", "Yiqing Hua", "Angie Chen", "Yury Malkov", "Dmitry Kalashnikov", "Shubham Gupta", "Kartik Audhkhasi", "Yuexiang Zhai", "Sudhindra Kopalle", "Prateek Jain", "Eran Ofek", "Clemens Meyer", "Khuslen Baatarsukh", "Hana Strejček", "Jun Qian", "James Freedman", "Ricardo Figueira", "Michal Sokolik", "Olivier Bachem", "Raymond Lin", "Dia Kharrat", "Chris Hidey", "Pingmei Xu", "Dennis Duan", "Yin Li", "Muge Ersoy", "Richard Everett", "Kevin Cen", "Rebeca Santamaria-Fernandez", "Amir Taubenfeld", "Ian Mackinnon", "Linda Deng", "Polina Zablotskaia", "Shashank Viswanadha", "Shivanker Goel", "Damion Yates", "Yunxiao Deng", "Peter Choy", "Mingqing Chen", "Abhishek Sinha", "Alex Mossin", "Yiming Wang", "Arthur Szlam", "Susan Hao", "Paul Kishan Rubenstein", "Metin Toksoz-Exley", "Miranda Aperghis", "Yin Zhong", "Junwhan Ahn", "Michael Isard", "Olivier Lacombe", "Florian Luisier", "Chrysovalantis Anastasiou", "Yogesh Kalley", "Utsav Prabhu", "Emma Dunleavy", "Shaan Bijwadia", "Justin Mao-Jones", "Kelly Chen", "Rama Pasumarthi", "Emily Wood", "Adil Dostmohamed", "Nate Hurley", "Jiri Simsa", "Alicia Parrish", "Mantas Pajarskas", "Matt Harvey", "Ondrej Skopek", "Yony Kochinski", "Javier Rey", "Verena Rieser", "Denny Zhou", "Sun Jae Lee", "Trilok Acharya", "Guowang Li", "Joe Jiang", "Xiaofan Zhang", "Bryant Gipson", "Ethan Mahintorabi", "Marco Gelmi", "Nima Khajehnouri", "Angel Yeh", "Kayi Lee", "Loic Matthey", "Leslie Baker", "Trang Pham", "Han Fu", "Alex Pak", "Prakhar Gupta", "Cristina Vasconcelos", "Adam Sadovsky", "Brian Walker", "Sissie Hsiao", "Patrik Zochbauer", "Andreea Marzoca", "Noam Velan", "Junhao Zeng", "Gilles Baechler", "Danny Driess", "Divya Jain", "Yanping Huang", "Lizzie Tao", "John Maggs", "Nir Levine", "Jon Schneider", "Erika Gemzer", "Samuel Petit", "Shan Han", "Zach Fisher", "Dustin Zelle", "Courtney Biles", "Eugene Ie", "Asya Fadeeva", "Casper Liu", "Juliana Vicente Franco", "Adrian Collister", "Hao Zhang", "Renshen Wang", "Ruizhe Zhao", "Leandro Kieliger", "Kurt Shuster", "Rui Zhu", "Boqing Gong", "Lawrence Chan", "Ruoxi Sun", "Sujoy Basu", "Roland Zimmermann", "Jamie Hayes", "Abhishek Bapna", "Jasper Snoek", "Weel Yang", "Puranjay Datta", "Jad Al Abdallah", "Kevin Kilgour", "Lu Li", "SQ Mah", "Yennie Jun", "Morgane Rivière", "Abhijit Karmarkar", "Tammo Spalink", "Tao Huang", "Lucas Gonzalez", "Duc-Hieu Tran", "Averi Nowak", "John Palowitch", "Martin Chadwick", "Ellie Talius", "Harsh Mehta", "Thibault Sellam", "Philipp Fränken", "Massimo Nicosia", "Kyle He", "Aditya Kini", "David Amos", "Sugato Basu", "Harrison Jobe", "Eleni Shaw", "Qiantong Xu", "Colin Evans", "Daisuke Ikeda", "Chaochao Yan", "Larry Jin", "Lun Wang", "Sachin Yadav", "Ilia Labzovsky", "Ramesh Sampath", "Ada Ma", "Candice Schumann", "Aditya Siddhant", "Rohin Shah", "John Youssef", "Rishabh Agarwal", "Natalie Dabney", "Alessio Tonioni", "Moran Ambar", "Jing Li", "Isabelle Guyon", "Benny Li", "David Soergel", "Boya Fang", "Georgi Karadzhov", "Cristian Udrescu", "Trieu Trinh", "Vikas Raunak", "Seb Noury", "Dee Guo", "Sonal Gupta", "Mara Finkelstein", "Denis Petek", "Lihao Liang", "Greg Billock", "Pei Sun", "David Wood", "Yiwen Song", "Xiaobin Yu", "Tatiana Matejovicova", "Regev Cohen", "Kalyan Andra", "David D'Ambrosio", "Zhiwei Deng", "Vincent Nallatamby", "Ebrahim Songhori", "Rumen Dangovski", "Andrew Lampinen", "Pankil Botadra", "Adam Hillier", "Jiawei Cao", "Nagabhushan Baddi", "Adhi Kuncoro", "Toshihiro Yoshino", "Ankit Bhagatwala", "Marcáurelio Ranzato", "Rylan Schaeffer", "Tianlin Liu", "Shuai Ye", "Obaid Sarvana", "John Nham", "Chenkai Kuang", "Isabel Gao", "Jinoo Baek", "Shubham Mittal", "Ayzaan Wahid", "Anita Gergely", "Bin Ni", "Josh Feldman", "Carrie Muir", "Pascal Lamblin", "Wolfgang Macherey", "Ethan Dyer", "Logan Kilpatrick", "Víctor Campos", "Mukul Bhutani", "Stanislav Fort", "Yanif Ahmad", "Aliaksei Severyn", "Kleopatra Chatziprimou", "Oleksandr Ferludin", "Mason Dimarco", "Aditya Kusupati", "Joe Heyward", "Dan Bahir", "Kevin Villela", "Katie Millican", "Dror Marcus", "Sanaz Bahargam", "Caglar Unlu", "Nicholas Roth", "Zichuan Wei", "Siddharth Gopal", "Deepanway Ghoshal", "Edward Lee", "Sharon Lin", "Jennie Lees", "Dayeong Lee", "Anahita Hosseini", "Connie Fan", "Seth Neel", "Marcus Wu", "Yasemin Altun", "Honglong Cai", "Enrique Piqueras", "Josh Woodward", "Alessandro Bissacco", "Salem Haykal", "Mahyar Bordbar", "Prasha Sundaram", "Sarah Hodkinson", "Daniel Toyama", "George Polovets", "Austin Myers", "Anu Sinha", "Tomer Levinboim", "Kashyap Krishnakumar", "Rachita Chhaparia", "Tatiana Sholokhova", "Nitesh Bharadwaj Gundavarapu", "Ganesh Jawahar", "Haroon Qureshi", "Jieru Hu", "Nikola Momchev", "Matthew Rahtz", "Renjie Wu", "Aishwarya P S", "Kedar Dhamdhere", "Meiqi Guo", "Umang Gupta", "Ali Eslami", "Mariano Schain", "Michiel Blokzijl", "David Welling", "Dave Orr", "Levent Bolelli", "Nicolas Perez-Nieves", "Mikhail Sirotenko", "Aman Prasad", "Arjun Kar", "Borja De Balle Pigem", "Tayfun Terzi", "Gellért Weisz", "Dipankar Ghosh", "Aditi Mavalankar", "Dhruv Madeka", "Kaspar Daugaard", "Hartwig Adam", "Viraj Shah", "Dana Berman", "Maggie Tran", "Steven Baker", "Ewa Andrejczuk", "Grishma Chole", "Ganna Raboshchuk", "Mahdi Mirzazadeh", "Thais Kagohara", "Shimu Wu", "Christian Schallhart", "Bernett Orlando", "Chen Wang", "Alban Rrustemi", "Hao Xiong", "Hao Liu", "Arpi Vezer", "Nolan Ramsden", "Shuo-yiin Chang", "Sidharth Mudgal", "Yan Li", "Nino Vieillard", "Yedid Hoshen", "Farooq Ahmad", "Ambrose Slone", "Amy Hua", "Natan Potikha", "Mirko Rossini", "Jon Stritar", "Sushant Prakash", "Zifeng Wang", "Xuanyi Dong", "Alireza Nazari", "Efrat Nehoran", "Kaan Tekelioglu", "Yinxiao Li", "Kartikeya Badola", "Tom Funkhouser", "Yuanzhen Li", "Varun Yerram", "Ramya Ganeshan", "Daniel Formoso", "Karol Langner", "Tian Shi", "Huijian Li", "Yumeya Yamamori", "Amayika Panda", "Alaa Saade", "Angelo Scorza Scarpati", "Chris Breaux", "CJ Carey", "Zongwei Zhou", "Cho-Jui Hsieh", "Sophie Bridgers", "Alena Butryna", "Nishesh Gupta", "Vaibhav Tulsyan", "Sanghyun Woo", "Evgenii Eltyshev", "Will Grathwohl", "Chanel Parks", "Seth Benjamin", "Rina Panigrahy", "Shenil Dodhia", "Daniel De Freitas", "Chris Sauer", "Will Song", "Ferran Alet", "Jackson Tolins", "Cosmin Paduraru", "Xingyi Zhou", "Brian Albert", "Zizhao Zhang", "Lei Shu", "Mudit Bansal", "Sarah Nguyen", "Amir Globerson", "Owen Xiao", "James Manyika", "Tom Hennigan", "Rong Rong", "Josip Matak", "Anton Bakalov", "Ankur Sharma", "Danila Sinopalnikov", "Andrew Pierson", "Stephen Roller", "Geoff Brown", "Mingcen Gao", "Toshiyuki Fukuzawa", "Amin Ghafouri", "Kenny Vassigh", "Iain Barr", "Zhicheng Wang", "Anna Korsun", "Rajesh Jayaram", "Lijie Ren", "Tim Zaman", "Samira Khan", "Yana Lunts", "Dan Deutsch", "Dave Uthus", "Nitzan Katz", "Masha Samsikova", "Amr Khalifa", "Nikhil Sethi", "Jiao Sun", "Luming Tang", "Uri Alon", "Xianghong Luo", "Dian Yu", "Abhishek Nayyar", "Bryce Petrini", "Will Truong", "Vincent Hellendoorn", "Nikolai Chinaev", "Chris Alberti", "Wei Wang", "Jingcao Hu", "Vahab Mirrokni", "Ananth Balashankar", "Avia Aharon", "Aahil Mehta", "Ahmet Iscen", "Joseph Kready", "Lucas Manning", "Anhad Mohananey", "Yuankai Chen", "Anshuman Tripathi", "Allen Wu", "Igor Petrovski", "Dawsen Hwang", "Martin Baeuml", "Shreyas Chandrakaladharan", "Yuan Liu", "Rey Coaguila", "Maxwell Chen", "Sally Ma", "Pouya Tafti", "Susheel Tatineni", "Terry Spitz", "Jiayu Ye", "Paul Vicol", "Mihaela Rosca", "Adrià Puigdomènech", "Zohar Yahav", "Sanjay Ghemawat", "Hanzhao Lin", "Phoebe Kirk", "Zaid Nabulsi", "Sergey Brin", "Bernd Bohnet", "Ken Caluwaerts", "Aditya Srikanth Veerubhotla", "Dan Zheng", "Zihang Dai", "Petre Petrov", "Yichong Xu", "Ramin Mehran", "Zhuo Xu", "Luisa Zintgraf", "Jiho Choi", "Spurthi Amba Hombaiah", "Romal Thoppilan", "Sashank Reddi", "Lukasz Lew", "Li Li", "Kellie Webster", "KP Sawhney", "Lampros Lamprou", "Siamak Shakeri", "Mayank Lunayach", "Jianmin Chen", "Sumit Bagri", "Alex Salcianu", "Ying Chen", "Yani Donchev", "Charlotte Magister", "Signe Nørly", "Vitor Rodrigues", "Tomas Izo", "Hila Noga", "Joe Zou", "Thomas Köppe", "Wenxuan Zhou", "Kenton Lee", "Xiangzhu Long", "Danielle Eisenbud", "Anthony Chen", "Connor Schenck", "Chi Ming To", "Peilin Zhong", "Emanuel Taropa", "Minh Truong", "Omer Levy", "Danilo Martins", "Zhiyuan Zhang", "Christopher Semturs", "Kelvin Zhang", "Alex Yakubovich", "Pol Moreno", "Lara McConnaughey", "Di Lu", "Sam Redmond", "Lotte Weerts", "Yonatan Bitton", "Tiziana Refice", "Nicolas Lacasse", "Arthur Conmy", "Corentin Tallec", "Julian Odell", "Hannah Forbes-Pollard", "Arkadiusz Socala", "Jonathan Hoech", "Pushmeet Kohli", "Alanna Walton", "Rui Wang", "Mikita Sazanovich", "Kexin Zhu", "Andrei Kapishnikov", "Rich Galt", "Matthew Denton", "Ben Murdoch", "Caitlin Sikora", "Kareem Mohamed", "Wei Wei", "Uri First", "Tim McConnell", "Luis C. Cobo", "James Qin", "Thi Avrahami", "Daniel Balle", "Yu Watanabe", "Annie Louis", "Adam Kraft", "Setareh Ariafar", "Yiming Gu", "Eugénie Rives", "Charles Yoon", "Andrei Rusu", "James Cobon-Kerr", "Chris Hahn", "Jiaming Luo", " Yuvein", " Zhu", "Niharika Ahuja", "Rodrigo Benenson", "Raphaël Lopez Kaufman", "Honglin Yu", "Lloyd Hightower", "Junlin Zhang", "Darren Ni", "Lisa Anne Hendricks", "Gabby Wang", "Gal Yona", "Lalit Jain", "Pablo Barrio", "Surya Bhupatiraju", "Siva Velusamy", "Allan Dafoe", "Sebastian Riedel", "Tara Thomas", "Zhe Yuan", "Mathias Bellaiche", "Sheena Panthaplackel", "Klemen Kloboves", "Sarthak Jauhari", "Canfer Akbulut", "Todor Davchev", "Evgeny Gladchenko", "David Madras", "Aleksandr Chuklin", "Tyrone Hill", "Quan Yuan", "Mukundan Madhavan", "Luke Leonhard", "Dylan Scandinaro", "Qihang Chen", "Ning Niu", "Arthur Douillard", "Bogdan Damoc", "Yasumasa Onoe", "Fabian Pedregosa", "Fred Bertsch", "Chas Leichner", "Joseph Pagadora", "Jonathan Malmaud", "Sameera Ponda", "Andy Twigg", "Oleksii Duzhyi", "Jingwei Shen", "Miaosen Wang", "Roopal Garg", "Jing Chen", "Utku Evci", "Jonathan Lee", "Leon Liu", "Koji Kojima", "Masa Yamaguchi", "Arunkumar Rajendran", "AJ Piergiovanni", "Vinodh Kumar Rajendran", "Marco Fornoni", "Gabriel Ibagon", "Harry Ragan", "Sadh MNM Khan", "John Blitzer", "Andrew Bunner", "Guan Sun", "Takahiro Kosakai", "Scott Lundberg", "Ndidi Elue", "Kelvin Guu", "SK Park", "Jane Park", "Arunachalam Narayanaswamy", "Chengda Wu", "Jayaram Mudigonda", "Trevor Cohn", "Hairong Mu", "Ravi Kumar", "Laura Graesser", "Yichi Zhang", "Richard Killam", "Vincent Zhuang", "Mai Giménez", "Wael Al Jishi", "Ruy Ley-Wild", "Alex Zhai", "Kazuki Osawa", "Diego Cedillo", "Jialu Liu", "Mayank Upadhyay", "Marcin Sieniek", "Roshan Sharma", "Tom Paine", "Anelia Angelova", "Sravanti Addepalli", "Carolina Parada", "Kingshuk Majumder", "Avery Lamp", "Sanjiv Kumar", "Xiang Deng", "Artiom Myaskovsky", "Tea Sabolić", "Jeffrey Dudek", "Sarah York", "Félix de Chaumont Quitry", "Jiazhong Nie", "Dee Cattle", "Alok Gunjan", "Bilal Piot", "Waleed Khawaja", "Seojin Bang", "Simon Wang", "Siavash Khodadadeh", "Raghavender R", "Praynaa Rawlani", "Richard Powell", "Kevin Lee", "Johannes Griesser", "GS Oh", "Cesar Magalhaes", "Yujia Li", "Simon Tokumine", "Hadas Natalie Vogel", "Dennis Hsu", "Arturo BC", "Disha Jindal", "Matan Cohen", "Zi Yang", "Junwei Yuan", "Dario de Cesare", "Tony Bruguier", "Jun Xu", "Monica Roy", "Alon Jacovi", "Dan Belov", "Rahul Arya", "Phoenix Meadowlark", "Shlomi Cohen-Ganor", "Wenting Ye", "Patrick Morris-Suzuki", "Praseem Banzal", "Gan Song", "Pranavaraj Ponnuramu", "Fred Zhang", "George Scrivener", "Salah Zaiem", "Alif Raditya Rochman", "Kehang Han", "Badih Ghazi", "Kate Lee", "Shahar Drath", "Daniel Suo", "Antonious Girgis", "Pradeep Shenoy", "Duy Nguyen", "Douglas Eck", "Somit Gupta", "Le Yan", "Joao Carreira", "Anmol Gulati", "Ruoxin Sang", "Daniil Mirylenka", "Emma Cooney", "Edward Chou", "Mingyang Ling", "Cindy Fan", "Ben Coleman", "Guilherme Tubone", "Ravin Kumar", "Jason Baldridge", "Felix Hernandez-Campos", "Angeliki Lazaridou", "James Besley", "Itay Yona", "Neslihan Bulut", "Quentin Wellens", "AJ Pierigiovanni", "Jasmine George", "Richard Green", "Pu Han", "Connie Tao", "Geoff Clark", "Chong You", "Abbas Abdolmaleki", "Justin Fu", "Tongzhou Chen", "Ashwin Chaugule", "Angad Chandorkar", "Altaf Rahman", "Will Thompson", "Penporn Koanantakool", "Mike Bernico", "Jie Ren", "Andrey Vlasov", "Sergei Vassilvitskii", "Maciej Kula", "Yizhong Liang", "Dahun Kim", "Yangsibo Huang", "Chengxi Ye", "Dmitry Lepikhin", "Wesley Helmholz"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-07-07", "url": "https://arxiv.org/abs/2507.06261", "pdf_url": "https://arxiv.org/pdf/2507.06261v6", "arxiv_id": "2507.06261", "doi": null, "citation_count": 3262, "influential_citation_count": 864, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 1.0}
{"id": "1b29811a0b7d1b97ab810b34992f649e02d3d425559acff3ab4268ccdeb02664", "sources": ["arxiv", "semantic_scholar"], "title": "When Does Divide and Conquer Work for Long Context LLM? A Noise Decomposition Framework", "abstract": "We investigate the challenge of applying Large Language Models (LLMs) to long texts. We propose a theoretical framework that distinguishes the failure modes of long context tasks into three categories: cross-chunk dependence (task noise), confusion that grows with context size (model noise), and the imperfect integration of partial results (aggregator noise). Under this view, we analyze when it is effective to use multi-agent chunking, i.e., dividing a lengthy sequence into smaller chunks and aggregating the processed results of each chunk. Our experiments on tasks such as retrieval, question answering, and summarization confirm both the theoretical analysis and the conditions that favor multi-agent chunking. By exploring the accelerated decay of model fidelity with input length, we also explain why, for large inputs, a weaker model configured with chunk-based processing can surpass a more advanced model like GPT4o applied in a single shot. Overall, we present a principled understanding framework and our results highlight a direct pathway to handling long contexts in LLMs with carefully managed chunking and aggregator strategies.", "authors": ["Zhen Xu", "Shang Zhu", "Jue Wang", "Junlin Wang", "Ben Athiwaratkun", "Chi Wang", "James Zou", "Ce Zhang"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-19", "url": "https://arxiv.org/abs/2506.16411", "pdf_url": "https://arxiv.org/pdf/2506.16411v2", "arxiv_id": "2506.16411", "doi": "10.48550/arXiv.2506.16411", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2113}
{"id": "20ab6778f363a0dbffd9359dde39cbc52e6b628a0937cbb68116b9b3661b9848", "sources": ["arxiv", "semantic_scholar"], "title": "LongLLaDA: Unlocking Long Context Capabilities in Diffusion LLMs", "abstract": "Large Language Diffusion Models, or diffusion LLMs, have emerged as a significant focus in NLP research, with substantial effort directed toward understanding their scalability and downstream task performance. However, their long-context capabilities remain unexplored, lacking systematic analysis or methods for context extension. In this work, we present the first systematic investigation comparing the long-context performance of diffusion LLMs and traditional auto-regressive LLMs. We first identify a unique characteristic of diffusion LLMs, unlike auto-regressive LLMs, they maintain remarkably stable perplexity during direct context extrapolation. Moreover, where auto-regressive models fail outright during the Needle-In-A-Haystack task with context exceeding their pretrained length, we discover diffusion LLMs exhibit a distinct local perception phenomenon, enabling successful retrieval from recent context segments. We explain both phenomena through the lens of Rotary Position Embedding (RoPE) scaling theory. Building on these observations, we propose LongLLaDA, a training-free method that integrates LLaDA with the NTK-based RoPE extrapolation. Our results validate that established extrapolation scaling laws remain effective for extending the context windows of diffusion LLMs. Furthermore, we identify long-context tasks where diffusion LLMs outperform auto-regressive LLMs and others where they fall short. Consequently, this study establishes the first length extrapolation method for diffusion LLMs while providing essential theoretical insights and empirical benchmarks critical for advancing future research on long-context diffusion LLMs. The code is available at https://github.com/OpenMOSS/LongLLaDA.", "authors": ["Xiaoran Liu", "Yuerong Song", "Zhigeng Liu", "Zengfeng Huang", "Qipeng Guo", "Ziwei He", "Xipeng Qiu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-17", "url": "https://arxiv.org/abs/2506.14429", "pdf_url": "https://arxiv.org/pdf/2506.14429v3", "arxiv_id": "2506.14429", "doi": "10.48550/arXiv.2506.14429", "citation_count": 35, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/OpenMOSS/LongLLaDA", "venue": "AAAI Conference on Artificial Intelligence", "quality_score": 0.3891}
{"id": "398b03f898fd31b6f91d11fc1ee72731916a4cdfe3d9dc6a9987e77b38405902", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Short Alignment for Effective Long-Context Modeling in LLMs", "abstract": "Large language models (LLMs) have exhibited impressive performance and surprising emergent properties. However, their effectiveness remains limited by the fixed context window of the transformer architecture, posing challenges for long-context modeling. Among these challenges, length generalization -- the ability to generalize to sequences longer than those seen during training -- is a classical and fundamental problem. In this work, we propose a fresh perspective on length generalization, shifting the focus from the conventional emphasis on input features such as positional encodings or data structures to the output distribution of the model. Specifically, through case studies on synthetic tasks, we highlight the critical role of \\textbf{long-short alignment} -- the consistency of output distributions across sequences of varying lengths. Extending this insight to natural language tasks, we propose a metric called Long-Short Misalignment to quantify this phenomenon, uncovering a strong correlation between the metric and length generalization performance. Building on these findings, we develop a regularization term that promotes long-short alignment during training. Extensive experiments validate the effectiveness of our approach, offering new insights for achieving more effective long-context modeling in LLMs. Code is available at https://github.com/PKU-ML/LongShortAlignment.", "authors": ["Tianqi Du", "Haotian Huang", "Yifei Wang", "Yisen Wang"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-13", "url": "https://arxiv.org/abs/2506.11769", "pdf_url": "https://arxiv.org/pdf/2506.11769v1", "arxiv_id": "2506.11769", "doi": "10.48550/arXiv.2506.11769", "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/PKU-ML/LongShortAlignment", "venue": "International Conference on Machine Learning", "quality_score": 0.2993}
{"id": "5b34f437d3b3a62e04df1d5599ddbc38be9b3bc3170f85d8923aaf66ea820d7c", "sources": ["arxiv", "semantic_scholar"], "title": "Lag-Relative Sparse Attention In Long Context Training", "abstract": "Large Language Models (LLMs) have made significant strides in natural language processing and generation, yet their ability to handle long-context input remains constrained by the quadratic complexity of attention computation and linear-increasing key-value memory footprint. To reduce computational costs and memory, key-value cache compression techniques are commonly applied at inference time, but this often leads to severe performance degradation, as models are not trained to handle compressed context. Although there are more sophisticated compression methods, they are typically unsuitable for post-training because of their incompatibility with gradient-based optimization or high computation overhead. To fill this gap with no additional parameter and little computation overhead, we propose Lag-Relative Sparse Attention(LRSA) anchored by the LagKV compression method for long context post-training. Our method performs chunk-by-chunk prefilling, which selects the top K most relevant key-value pairs in a fixed-size lagging window, allowing the model to focus on salient historical context while maintaining efficiency. Experimental results show that our approach significantly enhances the robustness of the LLM with key-value compression and achieves better fine-tuned results in the question-answer tuning task.", "authors": ["Manlai Liang", "Wanyi Huang", "Mandi Liu", "Huaijun Li", "Jinlong Li"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-13", "url": "https://arxiv.org/abs/2506.11498", "pdf_url": "https://arxiv.org/pdf/2506.11498v1", "arxiv_id": "2506.11498", "doi": "10.48550/arXiv.2506.11498", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1936}
{"id": "4708612f843032b8b41478799eeb81ebfc4e1b109c53d47621a74def05d3dcad", "sources": ["arxiv", "semantic_scholar"], "title": "CentroidKV: Efficient Long-Context LLM Inference via KV Cache Clustering", "abstract": "Large language models (LLMs) with extended context windows have become increasingly prevalent for tackling complex tasks. However, the substantial Key-Value (KV) cache required for long-context LLMs poses significant deployment challenges. Existing approaches either discard potentially critical information needed for future generations or offer limited efficiency gains due to high computational overhead. In this paper, we introduce CentroidKV, a simple yet effective framework for online KV cache clustering. Our approach is based on the observation that key states exhibit high similarity along the sequence dimension. To enable efficient clustering, we divide the sequence into chunks and propose Chunked Soft Matching, which employs an alternating partition strategy within each chunk and identifies clusters based on similarity. CentroidKV then merges the KV cache within each cluster into a single centroid. Additionally, we provide a theoretical analysis of the computational complexity and the optimality of the intra-chunk partitioning strategy. Extensive experiments across various models and long-context benchmarks demonstrate that CentroidKV achieves up to 75% reduction in KV cache memory usage while maintaining comparable model performance. Moreover, with minimal computational overhead, CentroidKV accelerates the decoding stage of inference by up to $1.92\\times$ and increases the serving throughput by up to $4\\times$.", "authors": ["Jie Hu", "Shengnan Wang", "Yutong He", "Ping Gong", "Jiawei Yi", "Juncheng Zhang", "Youhui Bai", "Renhai Chen", "Gong Zhang", "Cheng Li", "Kun Yuan"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-13", "url": "https://arxiv.org/abs/2506.11418", "pdf_url": "https://arxiv.org/pdf/2506.11418v2", "arxiv_id": "2506.11418", "doi": null, "citation_count": 7, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2258}
{"id": "9cb3f3b9ca26081fbd57ceb625491e047fbbba4ac0053e9985de04e225c5e470", "sources": ["arxiv", "semantic_scholar"], "title": "Mitigating Posterior Salience Attenuation in Long-Context LLMs with Positional Contrastive Decoding", "abstract": "While Large Language Models (LLMs) support long contexts, they struggle with performance degradation within the context window. Current solutions incur prohibitive training costs, leaving statistical behaviors and cost-effective approaches underexplored. From the decoding perspective, we identify the Posterior Salience Attenuation (PSA) phenomenon, where the salience ratio correlates with long-text performance degradation. Notably, despite the attenuation, gold tokens still occupy high-ranking positions in the decoding space. Motivated by it, we propose the training-free Positional Contrastive Decoding (PCD) that contrasts the logits derived from long-aware attention with those from designed local-aware attention, enabling the model to focus on the gains introduced by large-scale short-to-long training. Through the analysis of long-term decay simulation, we demonstrate that PCD effectively alleviates attention score degradation. Experimental results show that PCD achieves state-of-the-art performance on long-context benchmarks.", "authors": ["Zikai Xiao", "Ziyang Wang", "Wen Ma", "Yan Zhang", "Wei Shen", "Yan Wang", "Luqi Gong", "Zuozhu Liu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-10", "url": "https://arxiv.org/abs/2506.08371", "pdf_url": "https://arxiv.org/pdf/2506.08371v2", "arxiv_id": "2506.08371", "doi": "10.48550/arXiv.2506.08371", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.1902}
{"id": "860bd25d1e3a4a22b100e5e427b16084912681743af4def036b2f875a9969161", "sources": ["arxiv", "semantic_scholar"], "title": "Unable to Forget: Proactive Interference Reveals Working Memory Limits in LLMs Beyond Context Length", "abstract": "Information retrieval in Large Language Models (LLMs) is increasingly recognized as intertwined with generation capabilities rather than mere lookup. While longer contexts are often assumed to improve retrieval, the effects of intra-context interference remain understudied. To address this, we adapt the proactive interference (PI) paradigm from cognitive science, where earlier information disrupts recall of newer updates. In humans, susceptibility to such interference is inversely linked to working memory capacity. We introduce PI-LLM, an evaluation that sequentially streams semantically related key-value updates and queries only the final values. Although these final values are clearly positioned just before the query, LLM retrieval accuracy declines log-linearly toward zero as interference accumulates; errors arise from retrieving previously overwritten values. Attempts to mitigate interference via prompt engineering (e.g., instructing models to ignore earlier input) yield limited success. These findings reveal a fundamental constraint on LLMs' ability to disentangle interference and flexibly manipulate information, suggesting a working memory bottleneck beyond mere context access. This calls for approaches that strengthen models' ability to suppress irrelevant content during retrieval.", "authors": ["Chupei Wang", "Jiaqiu Vince Sun"], "categories": ["cs.CL", "cs.AI", "q-bio.NC"], "fields_of_study": ["Computer Science", "Biology"], "published_date": "2025-06-09", "url": "https://arxiv.org/abs/2506.08184", "pdf_url": "https://arxiv.org/pdf/2506.08184v3", "arxiv_id": "2506.08184", "doi": "10.48550/arXiv.2506.08184", "citation_count": 16, "influential_citation_count": 3, "has_code": true, "code_url": "https://github.com/zhuangziGiantfish/Unable-to-Forget", "venue": "arXiv.org", "quality_score": 0.3076}
{"id": "5fe3a80d0a32df7680f3462c033db1fe04c97db418e18159c857f03d61c94d99", "sources": ["arxiv", "semantic_scholar"], "title": "TracLLM: A Generic Framework for Attributing Long Context LLMs", "abstract": "Long context large language models (LLMs) are deployed in many real-world applications such as RAG, agent, and broad LLM-integrated applications. Given an instruction and a long context (e.g., documents, PDF files, webpages), a long context LLM can generate an output grounded in the provided context, aiming to provide more accurate, up-to-date, and verifiable outputs while reducing hallucinations and unsupported claims. This raises a research question: how to pinpoint the texts (e.g., sentences, passages, or paragraphs) in the context that contribute most to or are responsible for the generated output by an LLM? This process, which we call context traceback, has various real-world applications, such as 1) debugging LLM-based systems, 2) conducting post-attack forensic analysis for attacks (e.g., prompt injection attack, knowledge corruption attacks) to an LLM, and 3) highlighting knowledge sources to enhance the trust of users towards outputs generated by LLMs. When applied to context traceback for long context LLMs, existing feature attribution methods such as Shapley have sub-optimal performance and/or incur a large computational cost. In this work, we develop TracLLM, the first generic context traceback framework tailored to long context LLMs. Our framework can improve the effectiveness and efficiency of existing feature attribution methods. To improve the efficiency, we develop an informed search based algorithm in TracLLM. We also develop contribution score ensemble/denoising techniques to improve the accuracy of TracLLM. Our evaluation results show TracLLM can effectively identify texts in a long context that lead to the output of an LLM. Our code and data are at: https://github.com/Wang-Yanting/TracLLM.", "authors": ["Yanting Wang", "Wei Zou", "Runpeng Geng", "Jinyuan Jia"], "categories": ["cs.CR", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-04", "url": "https://arxiv.org/abs/2506.04202", "pdf_url": "https://arxiv.org/pdf/2506.04202v3", "arxiv_id": "2506.04202", "doi": "10.48550/arXiv.2506.04202", "citation_count": 7, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Wang-Yanting/TracLLM", "venue": "USENIX Security Symposium", "quality_score": 0.2833}
{"id": "f4e70d0328ebabee0fd513f9deae6bad16d1556d99e0557a038ebc2a94f88b89", "sources": ["arxiv", "semantic_scholar"], "title": "Context as Memory: Scene-Consistent Interactive Long Video Generation with Memory Retrieval", "abstract": "Recent advances in interactive video generation have shown promising results, yet existing approaches struggle with scene-consistent memory capabilities in long video generation due to limited use of historical context. In this work, we propose Context-as-Memory, which utilizes historical context as memory for video generation. It includes two simple yet effective designs: (1) storing context in frame format without additional post-processing; (2) conditioning by concatenating context and frames to be predicted along the frame dimension at the input, requiring no external control modules. Furthermore, considering the enormous computational overhead of incorporating all historical context, we propose the Memory Retrieval module to select truly relevant context frames by determining FOV (Field of View) overlap between camera poses, which significantly reduces the number of candidate frames without substantial information loss. Experiments demonstrate that Context-as-Memory achieves superior memory capabilities in interactive long video generation compared to SOTAs, even generalizing effectively to open-domain scenarios not seen during training. The link of our project page is https://context-as-memory.github.io/.", "authors": ["Jiwen Yu", "Jianhong Bai", "Yiran Qin", "Quande Liu", "Xintao Wang", "Pengfei Wan", "Di Zhang", "Xihui Liu"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-03", "url": "https://arxiv.org/abs/2506.03141", "pdf_url": "https://arxiv.org/pdf/2506.03141v2", "arxiv_id": "2506.03141", "doi": "10.1145/3757377.3763833", "citation_count": 119, "influential_citation_count": 18, "has_code": false, "code_url": null, "venue": "ACM SIGGRAPH Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia", "quality_score": 0.6394}
{"id": "3f419277194fc40f91880e5439576d08e83783c91eed8f76f67509ed65793cf9", "sources": ["arxiv", "semantic_scholar"], "title": "Beyond In-Context Learning: Aligning Long-form Generation of Large Language Models via Task-Inherent Attribute Guidelines", "abstract": "In-context learning (ICL) is an important yet not fully understood ability of pre-trained large language models (LLMs). It can greatly enhance task performance using a few examples, termed demonstrations, without fine-tuning. Although effective in question answering, ICL often underperforms in long-form generation tasks such as summarization. Under appropriately realistic assumptions, we empirically and theoretically show that ICL demonstrations alone are insufficient to teach LLMs the task language and format distributions for generation. We argue for explicit exposure to the task distributions and hypothesize that defining them by prompting enhances model performance. To this end, we present LongGuide, which efficiently generates two parallel streams of guidelines capturing task language and format properties: (i) Metric Guidelines (MGs) that instruct models to optimize self-evaluated metrics; and (ii) Output Constraint Guidelines (OCGs) that constrain generation at both token and sentence levels. LongGuide automatically selects the best combination of guidelines, improving both strong open- and closed-source LLMs by over 5% in both zero- and few-shot settings. We show that LongGuide is generalizable, learnable by weak models to enhance strong ones, and integrates synergistically with automatic prompt optimizers.", "authors": ["Do Xuan Long", "Duong Ngoc Yen", "Do Xuan Trong", "Luu Anh Tuan", "Kenji Kawaguchi", "Shafiq Joty", "Min-Yen Kan", "Nancy F. Chen"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-06-02", "url": "https://arxiv.org/abs/2506.01265", "pdf_url": "https://arxiv.org/pdf/2506.01265v1", "arxiv_id": "2506.01265", "doi": "10.48550/arXiv.2506.01265", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.181}
{"id": "d23575b7c15d83cb19f894f7f6b0e3dce43b371280cb7ffdb2a6984930289ad3", "sources": ["arxiv", "semantic_scholar"], "title": "Sentinel: Decoding Context Utilization via Attention Probing for Efficient LLM Context Compression", "abstract": "Retrieval-augmented generation (RAG) often suffers from long and noisy retrieved contexts. Existing context compression methods typically rely on heuristic relevance estimation or supervised compression models rather than on how LLMs utilize retrieved context during inference. We propose Sentinel, a lightweight sentence-level compression framework that decodes inference-time contextual utilization behaviors from head-wise attention patterns of frozen LLMs. To ground supervision in retrieval-dependent answering behavior, Sentinel trains a lightweight probe using QA examples where the model succeeds only when retrieved context is available. Sentinel performs compression using only a single non-autoregressive forward pass without dedicated compression training or autoregressive scoring. Empirically, we find that effective contextual utilization signals remain accessible even in compact proxy models. On LongBench, Sentinel with a 0.5B proxy model achieves up to 5$\\times$ compression while attaining question-answering performance competitive with compression methods built on 7B-scale models. Despite being trained only on English QA data, Sentinel also generalizes effectively to Chinese and out-of-domain settings.", "authors": ["Yong Zhang", "Heng Li", "Yanwen Huang", "Ning Cheng", "Yang Guo", "Yun Zhu", "Yanmeng Wang", "Shaojun Wang", "Jing Xiao"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-29", "url": "https://arxiv.org/abs/2505.23277", "pdf_url": "https://arxiv.org/pdf/2505.23277v3", "arxiv_id": "2505.23277", "doi": null, "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1123}
{"id": "872dd7e249a71d40adab080547523030d44636a32dd467b778bbc34378bbd763", "sources": ["arxiv", "semantic_scholar"], "title": "Bayesian Attention Mechanism: A Probabilistic Framework for Positional Encoding and Context Length Extrapolation", "abstract": "Transformer-based language models rely on positional encoding (PE) to handle token order and support context length extrapolation. However, existing PE methods lack theoretical clarity and rely on limited evaluation metrics to substantiate their extrapolation claims. We propose the Bayesian Attention Mechanism (BAM), a theoretical framework that formulates positional encoding as a prior within a probabilistic model. BAM unifies existing methods (e.g., NoPE and ALiBi) and motivates a new Generalized Gaussian positional prior that substantially improves long-context generalization. Empirically, BAM enables accurate information retrieval at $500\\times$ the training context length, outperforming previous state-of-the-art context length generalization in long context retrieval accuracy while maintaining comparable perplexity and introducing minimal additional parameters.", "authors": ["Arthur S. Bianchessi", "Yasmin C. Aguirre", "Rodrigo C. Barros", "Lucas S. Kupssinskü"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-28", "url": "https://arxiv.org/abs/2505.22842", "pdf_url": "https://arxiv.org/pdf/2505.22842v4", "arxiv_id": "2505.22842", "doi": "10.48550/arXiv.2505.22842", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1753}
{"id": "5c1bcb73bfb8d41ce08c711137fc2a3b9e5336d06682f0fb149055e386a8404b", "sources": ["arxiv", "semantic_scholar"], "title": "FIER: Fine-Grained and Efficient KV Cache Retrieval for Long-context LLM Inference", "abstract": "The Key-Value (KV) cache reading latency increases significantly with context lengths, hindering the efficiency of long-context LLM inference. To address this, previous works propose retaining a small fraction of KV cache based on token importance. For example, KV eviction uses static heuristics to retain tokens, while KV retrieval dynamically selects query-relevant tokens for more adaptive cache management. However, we observe that important tokens are often sparsely distributed across the long context. This sparsity makes existing page-level KV retrieval inaccurate, as each page may include irrelevant tokens and miss critical ones. In this work, we propose Fier, a \\underline{Fi}ne-Grained and \\underline{E}fficient KV cache \\underline{R}etrieval method. Fier uses 1-bit quantized keys to estimate the importance of each token, resulting in efficient and precise retrieval. Experiments show that Fier matches full KV performance using only 11\\% of the cache budget across various long-context tasks, reducing decoding latency by 1.2$\\times$ to 1.5$\\times$.Code is available at https://github.com/SimWangArizona/FIER", "authors": ["Dongwei Wang", "Zijie Liu", "Song Wang", "Yuxin Ren", "Jianing Deng", "Jingtong Hu", "Tianlong Chen", "Huanrui Yang"], "categories": ["cs.DB"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-28", "url": "https://arxiv.org/abs/2508.08256", "pdf_url": "https://arxiv.org/pdf/2508.08256v2", "arxiv_id": "2508.08256", "doi": "10.48550/arXiv.2508.08256", "citation_count": 3, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/SimWangArizona/FIER", "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.2709}
{"id": "5f5985212926e8a6fa4647a90cfc75568b22724dceb45cf850e10698460b048b", "sources": ["arxiv", "semantic_scholar"], "title": "Scaling External Knowledge Input Beyond Context Windows of LLMs via Multi-Agent Collaboration", "abstract": "With the rapid advancement of post-training techniques for reasoning and information seeking, large language models (LLMs) can incorporate a large quantity of retrieved knowledge to solve complex tasks. However, the limited context window of LLMs obstructs scaling the amount of external knowledge input, prohibiting further improvement. Existing context window extension methods inevitably cause information loss. LLM-based multi-agent methods emerge as a new paradigm to handle massive input in a distributional manner, where we identify two core bottlenecks in existing agent orchestration designs. In this work, we develop a multi-agent framework, \\textbf{\\ExtAgents}, to overcome the bottlenecks and enable better scalability in inference-time knowledge integration without longer-context training. Benchmarked with our enhanced multi-hop question answering test, \\textbf{$\\boldsymbol{\\infty}$Bench+}, and other public test sets including long survey generation, \\ExtAgents significantly enhances the performance over existing non-training methods with the same amount of external knowledge input, regardless of whether it falls \\emph{within or exceeds the context window}. Moreover, the method maintains efficiency due to high parallelism. We believe further study in the coordination of LLM agents on increasing external knowledge input could benefit real-world applications.", "authors": ["Zijun Liu", "Zhennan Wan", "Peng Li", "Ming Yan", "Fei Huang", "Yang Liu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-27", "url": "https://arxiv.org/abs/2505.21471", "pdf_url": "https://arxiv.org/pdf/2505.21471v2", "arxiv_id": "2505.21471", "doi": "10.48550/arXiv.2505.21471", "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/THUNLP-MT/ExtAgents", "venue": "arXiv.org", "quality_score": 0.2692}
{"id": "f68b5dbda741f364a531113aed4f2098ac76cdbb79db8bb85cfab841e2974066", "sources": ["arxiv", "semantic_scholar"], "title": "HoPE: Hybrid of Position Embedding for Long Context Vision-Language Models", "abstract": "Vision-Language Models (VLMs) have made significant progress in multimodal tasks. However, their performance often deteriorates in long-context scenarios, particularly long videos. While Rotary Position Embedding (RoPE) has been widely adopted for length generalization in Large Language Models (LLMs), extending vanilla RoPE to capture the intricate spatial-temporal dependencies in videos remains an unsolved challenge. Existing methods typically allocate different frequencies within RoPE to encode 3D positional information. However, these allocation strategies mainly rely on heuristics, lacking in-depth theoretical analysis. In this paper, we first study how different allocation strategies impact the long-context capabilities of VLMs. Our analysis reveals that current multimodal RoPEs fail to reliably capture semantic similarities over extended contexts. To address this issue, we propose HoPE, a Hybrid of Position Embedding designed to improve the long-context capabilities of VLMs. HoPE introduces a hybrid frequency allocation strategy for reliable semantic modeling over arbitrarily long contexts, and a dynamic temporal scaling mechanism to facilitate robust learning and flexible inference across diverse context lengths. Extensive experiments across four video benchmarks on long video understanding and retrieval tasks demonstrate that HoPE consistently outperforms existing methods, confirming its effectiveness. Our code is available at https://github.com/hrlics/HoPE.", "authors": ["Haoran Li", "Yingjie Qin", "Baoyuan Ou", "Lai Xu", "Ruiwen Xu"], "categories": ["cs.LG", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-26", "url": "https://arxiv.org/abs/2505.20444", "pdf_url": "https://arxiv.org/pdf/2505.20444v2", "arxiv_id": "2505.20444", "doi": null, "citation_count": 4, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/hrlics/HoPE", "venue": null, "quality_score": 0.2045}
{"id": "3675dbd527b1b963fa038336c79e20c98ad95b2e878255b92e7b9b1cc95383c0", "sources": ["arxiv", "semantic_scholar"], "title": "What Really Matters in Many-Shot Attacks? An Empirical Study of Long-Context Vulnerabilities in LLMs", "abstract": "We investigate long-context vulnerabilities in Large Language Models (LLMs) through Many-Shot Jailbreaking (MSJ). Our experiments utilize context length of up to 128K tokens. Through comprehensive analysis with various many-shot attack settings with different instruction styles, shot density, topic, and format, we reveal that context length is the primary factor determining attack effectiveness. Critically, we find that successful attacks do not require carefully crafted harmful content. Even repetitive shots or random dummy text can circumvent model safety measures, suggesting fundamental limitations in long-context processing capabilities of LLMs. The safety behavior of well-aligned models becomes increasingly inconsistent with longer contexts. These findings highlight significant safety gaps in context expansion capabilities of LLMs, emphasizing the need for new safety mechanisms.", "authors": ["Sangyeop Kim", "Yohan Lee", "Yongwoo Song", "Kimin Lee"], "categories": ["cs.CL", "cs.CR"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-26", "url": "https://arxiv.org/abs/2505.19773", "pdf_url": "https://arxiv.org/pdf/2505.19773v1", "arxiv_id": "2505.19773", "doi": "10.48550/arXiv.2505.19773", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.1747}
{"id": "8e9c212d0a7075849489c46123498a8495675e116bbcf26f298af6fc2956480c", "sources": ["arxiv", "semantic_scholar"], "title": "Meaning Is Not A Metric: Using LLMs to make cultural context legible at scale", "abstract": "This position paper argues that large language models (LLMs) can make cultural context, and therefore human meaning, legible at an unprecedented scale in AI-based sociotechnical systems. We argue that such systems have previously been unable to represent human meaning because they rely on thin descriptions (numerical representations that enforce standardization and therefore strip human activity of the cultural context which gives it meaning). By contrast, scholars in the humanities and qualitative social sciences have developed frameworks for representing meaning through thick description (verbal representations that accommodate heterogeneity and retain contextual information needed to represent human meaning). The verbal capabilities of LLMs now provide a means of at least partially automating the generation and processing of thick descriptions, offering new ways to deploy them at scale. We argue that the problem of rendering human meaning legible is not just about selecting better metrics but about developing new representational formats based on thick description. We frame this as a crucial direction for the application of generative AI and identify five key challenges: preserving context, maintaining interpretive pluralism, integrating perspectives based on lived experience and critical distance, distinguishing qualitative content from quantitative magnitude, and acknowledging meaning as dynamic rather than static.", "authors": ["Cody Kommers", "Drew Hemment", "Maria Antoniak", "Joel Z. Leibo", "Hoyt Long", "Emily Robinson", "Adam Sobey"], "categories": ["cs.CL", "cs.AI", "cs.CY"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-23", "url": "https://arxiv.org/abs/2505.23785", "pdf_url": "https://arxiv.org/pdf/2505.23785v2", "arxiv_id": "2505.23785", "doi": "10.48550/arXiv.2505.23785", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2386}
{"id": "5752b8bbf6f8ece7bf7b315a09324080fbc95960765b645d0c8edfa0f8000f69", "sources": ["arxiv", "semantic_scholar"], "title": "Longer Context, Deeper Thinking: Uncovering the Role of Long-Context Ability in Reasoning", "abstract": "Recent language models exhibit strong reasoning capabilities, yet the influence of long-context capacity on reasoning remains underexplored. In this work, we hypothesize that current limitations in reasoning stem, in part, from insufficient long-context capacity, motivated by empirical observations such as (1) higher context window length often leads to stronger reasoning performance, and (2) failed reasoning cases resemble failed long-context cases. To test this hypothesis, we examine whether enhancing a model's long-context ability before Supervised Fine-Tuning (SFT) leads to improved reasoning performance. Specifically, we compared models with identical architectures and fine-tuning data but varying levels of long-context capacity. Our results reveal a consistent trend: models with stronger long-context capacity achieve significantly higher accuracy on reasoning benchmarks after SFT. Notably, these gains persist even on tasks with short input lengths, indicating that long-context training offers generalizable benefits for reasoning performance. These findings suggest that long-context modeling is not just essential for processing lengthy inputs, but also serves as a critical foundation for reasoning. We advocate for treating long-context capacity as a first-class objective in the design of future language models.", "authors": ["Wang Yang", "Zirui Liu", "Hongye Jin", "Qingyu Yin", "Vipin Chaudhary", "Xiaotian Han"], "categories": ["cs.AI", "cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-22", "url": "https://arxiv.org/abs/2505.17315", "pdf_url": "https://arxiv.org/pdf/2505.17315v2", "arxiv_id": "2505.17315", "doi": "10.48550/arXiv.2505.17315", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2113}
{"id": "da6eed3c778e2e903ea2a4ece254242dc982085e72f8746794c4468b49210287", "sources": ["arxiv", "semantic_scholar"], "title": "Too Long, Didn't Model: Decomposing LLM Long-Context Understanding With Novels", "abstract": "Although the context length of large language models (LLMs) has increased to millions of tokens, evaluating their effectiveness beyond needle-in-a-haystack approaches has proven difficult. We argue that novels provide a case study of subtle, complicated structure and long-range semantic dependencies often over 128k tokens in length. Inspired by work on computational novel analysis, we release the Too Long, Didn't Model (TLDM) benchmark, which tests a model's ability to report plot summary, storyworld configuration, and elapsed narrative time. We find that none of seven tested frontier LLMs retain stable understanding beyond 64k tokens. Our results suggest language model developers must look beyond \"lost in the middle\" benchmarks when evaluating model performance in complex long-context scenarios. To aid in further development we release the TLDM benchmark together with reference code and data.", "authors": ["Sil Hamilton", "Rebecca M. M. Hicke", "Matthew Wilkens", "David Mimno"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-20", "url": "https://arxiv.org/abs/2505.14925", "pdf_url": "https://arxiv.org/pdf/2505.14925v1", "arxiv_id": "2505.14925", "doi": "10.48550/arXiv.2505.14925", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2258}
{"id": "9420cb19f58122bac0846c7bcade8f4e51a2dfa5942a9823633a1a352106f7cb", "sources": ["arxiv", "semantic_scholar"], "title": "Sense and Sensitivity: Examining the Influence of Semantic Recall on Long Context Code Reasoning", "abstract": "Large language models (LLMs) are increasingly deployed for understanding large codebases, but whether they understand operational semantics of long code context or rely on pattern matching shortcuts remains unclear. We distinguish between lexical recall (retrieving code verbatim) and semantic recall (understanding operational semantics). Evaluating 10 state-of-the-art LLMs, we find that while frontier models achieve near-perfect, position-independent lexical recall, semantic recall degrades severely when code is centrally positioned in long contexts. We introduce semantic recall sensitivity to measure whether tasks require understanding of code's operational semantics vs. permit pattern matching shortcuts. Through a novel counterfactual measurement method, we show that models rely heavily on pattern matching shortcuts to solve existing code understanding benchmarks. We propose a new task SemTrace, which achieves high semantic recall sensitivity through unpredictable operations; LLMs' accuracy exhibits severe positional effects, with median accuracy drops of 92.73% versus CRUXEval's 53.36% as the relevant code snippet approaches the middle of the input code context. Our findings suggest current evaluations substantially underestimate semantic recall failures in long context code understanding.", "authors": ["Adam Štorek", "Mukur Gupta", "Samira Hajizadeh", "Prashast Srivastava", "Suman Jana"], "categories": ["cs.CL", "cs.LG", "cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-19", "url": "https://arxiv.org/abs/2505.13353", "pdf_url": "https://arxiv.org/pdf/2505.13353v4", "arxiv_id": "2505.13353", "doi": null, "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1505}
{"id": "56cd8554f32ebbb130ceeb7a741e3a3a6525d63520936af8b35c155fd599a6e9", "sources": ["arxiv", "semantic_scholar"], "title": "PSC: Extending Context Window of Large Language Models via Phase Shift Calibration", "abstract": "Rotary Position Embedding (RoPE) is an efficient position encoding approach and is widely utilized in numerous large language models (LLMs). Recently, a lot of methods have been put forward to further expand the context window based on RoPE. The core concept of those methods is to predefine or search for a set of factors to rescale the base frequencies of RoPE. Nevertheless, it is quite a challenge for existing methods to predefine an optimal factor due to the exponential search space. In view of this, we introduce PSC (Phase Shift Calibration), a small module for calibrating the frequencies predefined by existing methods. With the employment of PSC, we demonstrate that many existing methods can be further enhanced, like PI, YaRN, and LongRoPE. We conducted extensive experiments across multiple models and tasks. The results demonstrate that (1) when PSC is enabled, the comparative reductions in perplexity increase as the context window size is varied from 16k, to 32k, and up to 64k. (2) Our approach is broadly applicable and exhibits robustness across a variety of models and tasks. The code can be found at https://github.com/WNQzhu/PSC.", "authors": ["Wenqiao Zhu", "Chao Xu", "Lulu Wang", "Jun Wu"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-18", "url": "https://arxiv.org/abs/2505.12423", "pdf_url": "https://arxiv.org/pdf/2505.12423v1", "arxiv_id": "2505.12423", "doi": "10.18653/v1/2024.emnlp-main.341", "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/WNQzhu/PSC", "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.2532}
{"id": "a1a489970a982505b373a591296923832227c76dfb2b03e219a8b95c62bf885a", "sources": ["arxiv", "semantic_scholar"], "title": "SoLoPO: Unlocking Long-Context Capabilities in LLMs via Short-to-Long Preference Optimization", "abstract": "Despite advances in pretraining with extended context sizes, large language models (LLMs) still face challenges in effectively utilizing real-world long-context information, primarily due to insufficient long-context alignment caused by data quality issues, training inefficiencies, and the lack of well-designed optimization objectives. To address these limitations, we propose a framework named \\textbf{S}h\\textbf{o}rt-to-\\textbf{Lo}ng \\textbf{P}reference \\textbf{O}ptimization (\\textbf{SoLoPO}), decoupling long-context preference optimization (PO) into two components: short-context PO and short-to-long reward alignment (SoLo-RA), supported by both theoretical and empirical evidence. Specifically, short-context PO leverages preference pairs sampled from short contexts to enhance the model's contextual knowledge utilization ability. Meanwhile, SoLo-RA explicitly encourages reward score consistency for the responses when conditioned on both short and long contexts that contain identical task-relevant information. This facilitates transferring the model's ability to handle short contexts into long-context scenarios. SoLoPO is compatible with mainstream preference optimization algorithms, while substantially improving the efficiency of data construction and training processes. Experimental results show that SoLoPO enhances all these algorithms with respect to stronger length and domain generalization abilities across various long-context benchmarks, while achieving notable improvements in both computational and memory efficiency.", "authors": ["Huashan Sun", "Shengyi Liao", "Yansen Han", "Yu Bai", "Yang Gao", "Cheng Fu", "Weizhou Shen", "Fanqi Wan", "Ming Yan", "Ji Zhang", "Fei Huang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-16", "url": "https://arxiv.org/abs/2505.11166", "pdf_url": "https://arxiv.org/pdf/2505.11166v3", "arxiv_id": "2505.11166", "doi": "10.48550/arXiv.2505.11166", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2113}
{"id": "c5ad7a5bc2dd9a819dde932e589bd011bc36451f3a9053d29e5ef9baecbc3dc2", "sources": ["arxiv", "semantic_scholar"], "title": "Learning Long-Context Diffusion Policies via Past-Token Prediction", "abstract": "Reasoning over long sequences of observations and actions is essential for many robotic tasks. Yet, learning effective long-context policies from demonstrations remains challenging. As context length increases, training becomes increasingly expensive due to rising memory demands, and policy performance often degrades as a result of spurious correlations. Recent methods typically sidestep these issues by truncating context length, discarding historical information that may be critical for subsequent decisions. In this paper, we propose an alternative approach that explicitly regularizes the retention of past information. We first revisit the copycat problem in imitation learning and identify an opposite challenge in recent diffusion policies: rather than over-relying on prior actions, they often fail to capture essential dependencies between past and future actions. To address this, we introduce Past-Token Prediction (PTP), an auxiliary task in which the policy learns to predict past action tokens alongside future ones. This regularization significantly improves temporal modeling in the policy head, with minimal reliance on visual representations. Building on this observation, we further introduce a multistage training strategy: pre-train the visual encoder with short contexts, and fine-tune the policy head using cached long-context embeddings. This strategy preserves the benefits of PTP while greatly reducing memory and computational overhead. Finally, we extend PTP into a self-verification mechanism at test time, enabling the policy to score and select candidates consistent with past actions during inference. Experiments across four real-world and six simulated tasks demonstrate that our proposed method improves the performance of long-context diffusion policies by 3x and accelerates policy training by more than 10x.", "authors": ["Marcel Torne", "Andy Tang", "Yuejiang Liu", "Chelsea Finn"], "categories": ["cs.RO", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-14", "url": "https://arxiv.org/abs/2505.09561", "pdf_url": "https://arxiv.org/pdf/2505.09561v2", "arxiv_id": "2505.09561", "doi": "10.48550/arXiv.2505.09561", "citation_count": 23, "influential_citation_count": 7, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4515}
{"id": "9c9094b5ceaad35ac3c7cd323ca91834d5a703540ab8ce01f974817a5cc9f642", "sources": ["arxiv", "semantic_scholar"], "title": "LongCodeBench: Evaluating Coding LLMs at 1M Context Windows", "abstract": "Context lengths for models have grown rapidly, from thousands to millions of tokens in just a few years. The extreme context sizes of modern long-context models have made it difficult to construct realistic long-context benchmarks -- not only due to the cost of collecting million-context tasks but also in identifying realistic scenarios that require significant contexts. We identify code comprehension and repair as a natural testbed and challenge task for long-context models and introduce LongCodeBench (LCB), a benchmark to test LLM coding abilities in long-context scenarios. Our benchmark tests both the comprehension and repair capabilities of LCLMs in realistic and important settings by drawing from real-world GitHub issues and constructing QA (LongCodeQA) and bug fixing (LongSWE-Bench) tasks. We carefully stratify the complexity of our benchmark, enabling us to evaluate models across different scales -- ranging from Qwen2.5 14B Instruct to Google's flagship Gemini model. We find that long-context remains a weakness for all models, with performance drops such as from 29% to 3% for Claude 3.5 Sonnet, or from 70.2% to 40% for Qwen2.5. The LCB dataset is available publicly at https://huggingface.co/datasets/Steefano/LCB and the codebase to replicate the work on this paper at https://github.com/Zteefano/long-code-bench.", "authors": ["Stefano Rando", "Luca Romani", "Alessio Sampieri", "Luca Franco", "John Yang", "Yuta Kyuragi", "Fabio Galasso", "Tatsunori Hashimoto"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-12", "url": "https://arxiv.org/abs/2505.07897", "pdf_url": "https://arxiv.org/pdf/2505.07897v3", "arxiv_id": "2505.07897", "doi": "10.48550/arXiv.2505.07897", "citation_count": 30, "influential_citation_count": 6, "has_code": true, "code_url": "https://github.com/Zteefano/long-code-bench", "venue": "arXiv.org", "quality_score": 0.4225}
{"id": "4639d2eaf809e8edc0ba4bd13d6c4a012152a5d57e6b4fa77f28df4bdfcbc507", "sources": ["arxiv", "semantic_scholar"], "title": "Overflow Prevention Enhances Long-Context Recurrent LLMs", "abstract": "A recent trend in LLMs is developing recurrent sub-quadratic models that improve long-context processing efficiency. We investigate leading large long-context models, focusing on how their fixed-size recurrent memory affects their performance. Our experiments reveal that, even when these models are trained for extended contexts, their use of long contexts remains underutilized. Specifically, we demonstrate that a chunk-based inference procedure, which identifies and processes only the most relevant portion of the input can mitigate recurrent memory failures and be effective for many long-context tasks: On LongBench, our method improves the overall performance of Falcon3-Mamba-Inst-7B by 14%, Falcon-Mamba-Inst-7B by 28%, RecurrentGemma-IT-9B by 50%, and RWKV6-Finch-7B by 51%. Surprisingly, this simple approach also leads to state-of-the-art results in the challenging LongBench v2 benchmark, showing competitive performance with equivalent size Transformers. Furthermore, our findings raise questions about whether recurrent models genuinely exploit long-range dependencies, as our single-chunk strategy delivers stronger performance - even in tasks that presumably require cross-context relations.", "authors": ["Assaf Ben-Kish", "Itamar Zimerman", "M. Jehanzeb Mirza", "Lior Wolf", "James Glass", "Leonid Karlinsky", "Raja Giryes"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-12", "url": "https://arxiv.org/abs/2505.07793", "pdf_url": "https://arxiv.org/pdf/2505.07793v2", "arxiv_id": "2505.07793", "doi": "10.48550/arXiv.2505.07793", "citation_count": 3, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/assafbk/OPRM", "venue": "arXiv.org", "quality_score": 0.2426}
{"id": "8548c290bd3f447f3967f98234e1b8dee6e82047c9e9daaa3856de4b15331314", "sources": ["arxiv", "semantic_scholar"], "title": "Breaking Quadratic Barriers: A Non-Attention LLM for Ultra-Long Context Horizons", "abstract": "We present a novel non attention based architecture for large language models (LLMs) that efficiently handles very long context windows, on the order of hundreds of thousands to potentially millions of tokens. Unlike traditional Transformer designs, which suffer from quadratic memory and computation overload due to the nature of the self attention mechanism, our model avoids token to token attention entirely. Instead, it combines the following complementary components: State Space blocks (inspired by S4) that learn continuous time convolution kernels and scale near linearly with sequence length, Multi Resolution Convolution layers that capture local context at different dilation levels, a lightweight Recurrent Supervisor to maintain a global hidden state across sequential chunks, and Retrieval Augmented External Memory that stores and retrieves high-level chunk embeddings without reintroducing quadratic operations.", "authors": ["Andrew Kiruluta", "Preethi Raju", "Priscilla Burity"], "categories": ["cs.LG", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-09", "url": "https://arxiv.org/abs/2506.01963", "pdf_url": "https://arxiv.org/pdf/2506.01963v1", "arxiv_id": "2506.01963", "doi": "10.48550/arXiv.2506.01963", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1535}
{"id": "79fcbcfd5017c403a2369e4b78e10f310f187a5a1874291c10d7e188f069551f", "sources": ["arxiv", "semantic_scholar"], "title": "RetroInfer: A Vector Storage Engine for Scalable Long-Context LLM Inference", "abstract": "Recent large language models (LLMs) are rapidly extending their context windows, yet inference throughput lags due to increasing GPU memory and bandwidth demands. This is because the key-value (KV) cache, an intermediate structure storing token representations, grows linearly with context length and requires an iterative linear scan for attention computation. A promising direction to accelerate long-context inference is to exploit attention's inherent sparsity by offloading the KV cache to CPU memory and retrieving only a small subset of tokens important to the current generation step. However, prior sparse attention approaches struggle to balance accuracy and retrieval cost due to varying sparsity patterns and inefficient GPU-CPU memory management. We present RetroInfer, a vector storage engine that realizes a sparsity-based KV cache for long-context inference. RetroInfer introduces an Attention-aWare VEctor index (wave index), which fundamentally improves the tradeoff between attention accuracy and retrieval cost through tripartite attention approximation, accuracy-bound attention estimation, and segmented clustering. We also design the wave buffer, a GPU-CPU buffer manager that assigns computation and manages data across heterogeneous hardware. We evaluate RetroInfer across a range of models and workloads, demonstrating up to 4.4X decoding throughput over full attention at 120K context and up to 12.2X over sparse attention baselines at 1 million tokens -- all while preserving full-attention-level accuracy.", "authors": ["Yaoqi Chen", "Jinkai Zhang", "Baotong Lu", "Qianxi Zhang", "Chengruidong Zhang", "Jing Liu", "Jingjia Luo", "Di Liu", "Huiqiang Jiang", "Qi Chen", "Bailu Ding", "Xiao Yan", "Jiawei Jiang", "Chen Chen", "Mingxing Zhang", "Cheng Li", "Yuqing Yang", "Fan Yang", "Mao Yang"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-05", "url": "https://arxiv.org/abs/2505.02922", "pdf_url": "https://arxiv.org/pdf/2505.02922v3", "arxiv_id": "2505.02922", "doi": "10.14778/3796195.3796212", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Proceedings of the VLDB Endowment", "quality_score": 0.1505}
{"id": "2bbc4c69a04cb03901dab526d8d3a79ffd7d7cbeaad0f0891d2b95b3ec251ca5", "sources": ["arxiv", "semantic_scholar"], "title": "FreqKV: Key-Value Compression in Frequency Domain for Context Window Extension", "abstract": "Existing key-value (KV) cache compression methods for large language models (LLMs) often rely on token eviction, which risks losing critical local information in both long prefilling and decoding scenarios. When extrapolating beyond the pretrained context length, their performance degrades sharply on long-context benchmarks. Motivated by the observation in the frequency domain that the context information is concentrated in the low-frequency components, we propose FreqKV, a parameter-free and architecture-agnostic approach. It iteratively compresses the increasing KV cache in the frequency domain, allowing models to process lengthy contexts efficiently. With minimal training at 8K length, FreqKV extends the context window of LLaMA-2-7B up to 256K tokens while maintaining stable perplexity. Extensive experiments across prefilling and decoding demonstrate that FreqKV enables robust context window extension and consistently outperforms existing KV cache compression methods on LLaMA-2 and LLaMA-3, highlighting its effectiveness for both understanding and generation in long contexts.", "authors": ["Jushi Kai", "Yixuan Wang", "Boyi Zeng", "Haoli Bai", "Bo Jiang", "Ziwei He", "Zhouhan Lin"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-05-01", "url": "https://arxiv.org/abs/2505.00570", "pdf_url": "https://arxiv.org/pdf/2505.00570v3", "arxiv_id": "2505.00570", "doi": null, "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1747}
{"id": "24778953e69e00127060f17089643a42b288b017853b7fa8cd3ceb9a5c436976", "sources": ["arxiv", "semantic_scholar"], "title": "LongFuncEval: Measuring the effectiveness of long context models for function calling", "abstract": "Multiple recent studies have documented large language models' (LLMs) performance on calling external tools/functions. Others focused on LLMs' abilities to handle longer context lengths. At the intersection of these areas lies another interesting problem: LLMs' abilities to accurately perform function calls in long context settings. Particularly, when calling tools, LLMs are encumbered by three predominant challenges: (1) a large catalog of tools, (2) long responses from the tool APIs, and (3) long multi-turn conversations. These challenges are particularly relevant to enterprise applications of LLMs which engage in multi-turn conversations with users to complete complex tasks that require a large catalog of complex tools. The literature contains multiple investigations of long context challenges such as lost in the middle or needle in the haystack for natural language tasks. In this paper, we make the first attempt to comprehensively study the long context understanding capabilities of these models in the tool calling setup. We modify existing benchmarks for challenge 1 and 3, and create a new evaluation set for challenge 2 to enable this analysis. We gradually increase the input context length and also vary the position of the answer in the input. When evaluated with several long context models, we observe a performance drop of 7% to 85% as the number of tools increases, a 7% to 91% degradation in answer retrieval as the tool responses length increases, and 13% and 40% degradation for as multi-turn conversations get longer. Our study shows that LLMs still struggle with long context in tool calling settings, motivating future research to drive further LLM improvements.", "authors": ["Kiran Kate", "Tejaswini Pedapati", "Kinjal Basu", "Yara Rizk", "Vijil Chenthamarakshan", "Subhajit Chaudhury", "Mayank Agarwal", "Ibrahim Abdelaziz"], "categories": ["cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-30", "url": "https://arxiv.org/abs/2505.10570", "pdf_url": "https://arxiv.org/pdf/2505.10570v1", "arxiv_id": "2505.10570", "doi": "10.48550/arXiv.2505.10570", "citation_count": 13, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2865}
{"id": "aff57d07c92316be3fefdcec79cf1ecd55d3a36fdb6d79cfeaf6034d43bfe3e1", "sources": ["arxiv", "semantic_scholar"], "title": "Towards Long Context Hallucination Detection", "abstract": "Large Language Models (LLMs) have demonstrated remarkable performance across various tasks. However, they are prone to contextual hallucination, generating information that is either unsubstantiated or contradictory to the given context. Although many studies have investigated contextual hallucinations in LLMs, addressing them in long-context inputs remains an open problem. In this work, we take an initial step toward solving this problem by constructing a dataset specifically designed for long-context hallucination detection. Furthermore, we propose a novel architecture that enables pre-trained encoder models, such as BERT, to process long contexts and effectively detect contextual hallucinations through a decomposition and aggregation mechanism. Our experimental results show that the proposed architecture significantly outperforms previous models of similar size as well as LLM-based models across various metrics, while providing substantially faster inference.", "authors": ["Siyi Liu", "Kishaloy Halder", "Zheng Qi", "Wei Xiao", "Nikolaos Pappas", "Phu Mon Htut", "Neha Anna John", "Yassine Benajiba", "Dan Roth"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-28", "url": "https://arxiv.org/abs/2504.19457", "pdf_url": "https://arxiv.org/pdf/2504.19457v1", "arxiv_id": "2504.19457", "doi": "10.48550/arXiv.2504.19457", "citation_count": 18, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "North American Chapter of the Association for Computational Linguistics", "quality_score": 0.3197}
{"id": "f3c139da0f4b9d0d155a147d9e41150dcffc44cc7b2d5ddd0490494a58145200", "sources": ["arxiv", "semantic_scholar"], "title": "llm-jp-modernbert: A ModernBERT Model Trained on a Large-Scale Japanese Corpus with Long Context Length", "abstract": "Encoder-only transformer models like BERT are widely adopted as a pre-trained backbone for tasks like sentence classification and retrieval. However, pretraining of encoder models with large-scale corpora and long contexts has been relatively underexplored compared to decoder-only transformers. In this work, we present llm-jp-modernbert, a ModernBERT model trained on a publicly available, massive Japanese corpus with a context length of 8192 tokens. While our model does not surpass existing baselines on downstream tasks, it achieves good results on fill-mask test evaluations. We also analyze the effect of context length expansion through pseudo-perplexity experiments. Furthermore, we investigate sentence embeddings in detail, analyzing their transitions during training and comparing them with those from other existing models, confirming similar trends with models sharing the same architecture. To support reproducibility and foster the development of long-context BERT, we release our model, along with the training and evaluation code.", "authors": ["Issa Sugiura", "Kouta Nakayama", "Yusuke Oda"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-22", "url": "https://arxiv.org/abs/2504.15544", "pdf_url": "https://arxiv.org/pdf/2504.15544v1", "arxiv_id": "2504.15544", "doi": "10.48550/arXiv.2504.15544", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2386}
{"id": "b40f7d3e1a4365004f2a6bbbfbb54a07beb1fef63644b2d618bc5cd7aa2d80c8", "sources": ["arxiv", "semantic_scholar"], "title": "SlimPipe: Memory-Thrifty and Efficient Pipeline Parallelism for Long-Context LLM Training", "abstract": "Pipeline Parallelism (PP) serves as a crucial technique for training Large Language Models (LLMs), owing to its capability to alleviate memory pressure from model states with relatively low communication overhead. However, in long-context scenarios, existing pipeline parallelism methods fail to address the substantial activation memory pressure, primarily due to the peak memory consumption resulting from the accumulation of activations across multiple microbatches. Moreover, these approaches inevitably introduce considerable pipeline bubbles, further hindering efficiency. To tackle these challenges, we propose SlimPipe, a novel approach to fine-grained pipeline parallelism that employs uniform sequence slicing coupled with one-forward-one-backward (1F1B) schedule. It reduces the accumulated activations from several microbatches to just one, which is split into several slices. Although the slices are evenly partitioned, the computation cost is not equal across slices due to causal attention. We develop a sophisticated workload redistribution technique to address this load imbalance. SlimPipe achieves (1) near-zero memory overhead and (2) minimal pipeline bubbles simultaneously. The effectiveness of SlimPipe has been proven by thorough testing with diverse model architectures, context window sizes, and SlimPipe-specific configurations. For example, on the Llama 70B model, compared to state-of-the-art methods, SlimPipe significantly boosts the Model FLOPs Utilization (MFU) to up to $1.57\\times$ for a context length of 512K. More notably, for a context length of 2048K, it maintains over 45% utilization on 256 NVIDIA Hopper 80GB GPUs, while other approaches either suffer significant performance drops or fail entirely due to memory constraints.", "authors": ["Zhouyang Li", "Yuliang Liu", "Wei Zhang", "Tailing Yuan", "Bin Chen", "Chengru Song", "Di Zhang"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-20", "url": "https://arxiv.org/abs/2504.14519", "pdf_url": "https://arxiv.org/pdf/2504.14519v1", "arxiv_id": "2504.14519", "doi": "10.1145/3712285.3759855", "citation_count": 4, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Software Composition", "quality_score": 0.1747}
{"id": "735d73da5fa06b3934641bcb71e10ffbb50c3a8f242b9a7f9fb48edc981d9dac", "sources": ["arxiv", "semantic_scholar"], "title": "Can LLMs reason over extended multilingual contexts? Towards long-context evaluation beyond retrieval and haystacks", "abstract": "Existing multilingual long-context benchmarks, often based on the popular needle-in-a-haystack test, primarily evaluate a model's ability to locate specific information buried within irrelevant texts. However, such a retrieval-centric approach is myopic and inherently limited, as successful recall alone does not indicate a model's capacity to reason over extended contexts. Moreover, these benchmarks are susceptible to data leakage, short-circuiting, and risk making the evaluation a priori identifiable. To address these limitations, we introduce MLRBench, a new synthetic benchmark for multilingual long-context reasoning. Unlike existing benchmarks, MLRBench goes beyond surface-level retrieval by including tasks that assess multi-hop inference, aggregation, and epistemic reasoning. Spanning seven languages, MLRBench is designed to be parallel, resistant to leakage, and scalable to arbitrary context lengths. Our extensive experiments with an open-weight large language model (LLM) reveal a pronounced gap between high- and low-resource languages, particularly for tasks requiring the model to aggregate multiple facts or predict the absence of information. We also find that, in multilingual settings, LLMs effectively utilize less than 30% of their claimed context length. Although off-the-shelf Retrieval Augmented Generation helps alleviate this to a certain extent, it does not solve the long-context problem. We open-source MLRBench to enable future research in improved evaluation and training of multilingual LLMs.", "authors": ["Amey Hengle", "Prasoon Bajpai", "Soham Dan", "Tanmoy Chakraborty"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-17", "url": "https://arxiv.org/abs/2504.12845", "pdf_url": "https://arxiv.org/pdf/2504.12845v1", "arxiv_id": "2504.12845", "doi": "10.48550/arXiv.2504.12845", "citation_count": 4, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1983}
{"id": "395deacee509919ac5e78830e4033391166983613c5e33336c65617075fc98c4", "sources": ["arxiv", "semantic_scholar"], "title": "Scaling Instruction-Tuned LLMs to Million-Token Contexts via Hierarchical Synthetic Data Generation", "abstract": "Large Language Models (LLMs) struggle with long-context reasoning, not only due to the quadratic scaling of computational complexity with sequence length but also because of the scarcity and expense of annotating long-context data. There has been barely any open-source work that systematically ablates long-context data, nor is there any openly available instruction tuning dataset with contexts surpassing 100K tokens. To bridge this gap, we introduce a novel post-training synthetic data generation strategy designed to efficiently extend the context window of LLMs while preserving their general task performance. Our approach scalably extends to arbitrarily long context lengths, unconstrained by the length of available real-world data, which effectively addresses the scarcity of raw long-context data. Through a step-by-step rotary position embedding (RoPE) scaling training strategy, we demonstrate that our model, with a context length of up to 1M tokens, performs well on the RULER benchmark and InfiniteBench and maintains robust performance on general language tasks.", "authors": ["Linda He", "Jue Wang", "Maurice Weber", "Shang Zhu", "Ben Athiwaratkun", "Ce Zhang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-17", "url": "https://arxiv.org/abs/2504.12637", "pdf_url": "https://arxiv.org/pdf/2504.12637v1", "arxiv_id": "2504.12637", "doi": "10.48550/arXiv.2504.12637", "citation_count": 4, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.1983}
{"id": "5b5cfbbc7cda14338513dae4beb70ccf173b5b0698e526044f7d97138061de60", "sources": ["arxiv", "semantic_scholar"], "title": "AlayaDB: The Data Foundation for Efficient and Effective Long-context LLM Inference", "abstract": "AlayaDB is a cutting-edge vector database system natively architected for efficient and effective long-context inference for Large Language Models (LLMs) at AlayaDB AI. Specifically, it decouples the KV cache and attention computation from the LLM inference systems, and encapsulates them into a novel vector database system. For the Model as a Service providers (MaaS), AlayaDB consumes fewer hardware resources and offers higher generation quality for various workloads with different kinds of Service Level Objectives (SLOs), when comparing with the existing alternative solutions (e.g., KV cache disaggregation, retrieval-based sparse attention). The crux of AlayaDB is that it abstracts the attention computation and cache management for LLM inference into a query processing procedure, and optimizes the performance via a native query optimizer. In this work, we demonstrate the effectiveness of AlayaDB via (i) three use cases from our industry partners, and (ii) extensive experimental results on LLM inference benchmarks.", "authors": ["Yangshen Deng", "Zhengxin You", "Long Xiang", "Qilong Li", "Peiqi Yuan", "Zhaoyang Hong", "Yitao Zheng", "Wanting Li", "Runzhong Li", "Haotian Liu", "Kyriakos Mouratidis", "Man Lung Yiu", "Huan Li", "Qiaomu Shen", "Rui Mao", "Bo Tang"], "categories": ["cs.AI", "cs.DB", "cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-14", "url": "https://arxiv.org/abs/2504.10326", "pdf_url": "https://arxiv.org/pdf/2504.10326v1", "arxiv_id": "2504.10326", "doi": "10.1145/3722212.3724428", "citation_count": 11, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2698}
{"id": "1df3f1ff9e326469952fa98a6c44aa986304846ce91a96accae9c6da4751b64d", "sources": ["arxiv", "semantic_scholar"], "title": "Multimodal Long Video Modeling Based on Temporal Dynamic Context", "abstract": "Recent advances in Large Language Models (LLMs) have led to significant breakthroughs in video understanding. However, existing models still struggle with long video processing due to the context length constraint of LLMs and the vast amount of information within the video. Although some recent methods are designed for long video understanding, they often lose crucial information during token compression and struggle with additional modality like audio. In this work, we propose a dynamic long video encoding method utilizing the temporal relationship between frames, named Temporal Dynamic Context (TDC). Firstly, we segment the video into semantically consistent scenes based on inter-frame similarities, then encode each frame into tokens using visual-audio encoders. Secondly, we propose a novel temporal context compressor to reduce the number of tokens within each segment. Specifically, we employ a query-based Transformer to aggregate video, audio, and instruction text tokens into a limited set of temporal context tokens. Finally, we feed the static frame tokens and the temporal context tokens into the LLM for video understanding. Furthermore, to handle extremely long videos, we propose a training-free chain-of-thought strategy that progressively extracts answers from multiple video segments. These intermediate answers serve as part of the reasoning process and contribute to the final answer. We conduct extensive experiments on general video understanding and audio-video understanding benchmarks, where our method demonstrates strong performance. The code and models are available at https://github.com/Hoar012/TDC-Video.", "authors": ["Haoran Hao", "Jiaming Han", "Yiyuan Zhang", "Xiangyu Yue"], "categories": ["cs.CV", "cs.AI", "cs.CL", "cs.LG", "cs.MM"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-14", "url": "https://arxiv.org/abs/2504.10443", "pdf_url": "https://arxiv.org/pdf/2504.10443v1", "arxiv_id": "2504.10443", "doi": "10.48550/arXiv.2504.10443", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Hoar012/TDC-Video", "venue": "arXiv.org", "quality_score": 0.193}
{"id": "04f500a534711a0f199742716e69fb272ab7af034fc6698021158d9df7d80781", "sources": ["arxiv", "semantic_scholar"], "title": "Long Context In-Context Compression by Getting to the Gist of Gisting", "abstract": "Long context processing is critical for the adoption of LLMs, but existing methods often introduce architectural complexity that hinders their practical adoption. Gisting, an in-context compression method with no architectural modification to the decoder transformer, is a promising approach due to its simplicity and compatibility with existing frameworks. While effective for short instructions, we demonstrate that gisting struggles with longer contexts, with significant performance drops even at minimal compression rates. Surprisingly, a simple average pooling baseline consistently outperforms gisting. We analyze the limitations of gisting, including information flow interruptions, capacity limitations and the inability to restrict its attention to subsets of the context. Motivated by theoretical insights into the performance gap between gisting and average pooling, and supported by extensive experimentation, we propose GistPool, a new in-context compression method. GistPool preserves the simplicity of gisting, while significantly boosting its performance on long context compression tasks.", "authors": ["Aleksandar Petrov", "Mark Sandler", "Andrey Zhmoginov", "Nolan Miller", "Max Vladymyrov"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-11", "url": "https://arxiv.org/abs/2504.08934", "pdf_url": "https://arxiv.org/pdf/2504.08934v1", "arxiv_id": "2504.08934", "doi": "10.48550/arXiv.2504.08934", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2386}
{"id": "7b4f2c0de8430aa9df55780bb5ede504e5fdff13ad8dcce4d2830f6ceaa2e046", "sources": ["arxiv", "semantic_scholar"], "title": "From 128K to 4M: Efficient Training of Ultra-Long Context Large Language Models", "abstract": "Long-context capabilities are essential for a wide range of applications, including document and video understanding, in-context learning, and inference-time scaling, all of which require models to process and reason over long sequences of text and multimodal data. In this work, we introduce a efficient training recipe for building ultra-long context LLMs from aligned instruct model, pushing the boundaries of context lengths from 128K to 1M, 2M, and 4M tokens. Our approach leverages efficient continued pretraining strategies to extend the context window and employs effective instruction tuning to maintain the instruction-following and reasoning abilities. Our UltraLong-8B, built on Llama3.1-Instruct with our recipe, achieves state-of-the-art performance across a diverse set of long-context benchmarks. Importantly, models trained with our approach maintain competitive performance on standard benchmarks, demonstrating balanced improvements for both long and short context tasks. We further provide an in-depth analysis of key design choices, highlighting the impacts of scaling strategies and data composition. Our findings establish a robust framework for efficiently scaling context lengths while preserving general model capabilities. We release all model weights at: https://ultralong.github.io/.", "authors": ["Chejian Xu", "Wei Ping", "Peng Xu", "Zihan Liu", "Boxin Wang", "Mohammad Shoeybi", "Bo Li", "Bryan Catanzaro"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-08", "url": "https://arxiv.org/abs/2504.06214", "pdf_url": "https://arxiv.org/pdf/2504.06214v1", "arxiv_id": "2504.06214", "doi": "10.48550/arXiv.2504.06214", "citation_count": 13, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2865}
{"id": "1177aec0b47763537910f8b8245e2741d23048d20a22f565e3d1ac5a571f224f", "sources": ["arxiv", "semantic_scholar"], "title": "Clinical ModernBERT: An efficient and long context encoder for biomedical text", "abstract": "We introduce Clinical ModernBERT, a transformer based encoder pretrained on large scale biomedical literature, clinical notes, and medical ontologies, incorporating PubMed abstracts, MIMIC IV clinical data, and medical codes with their textual descriptions. Building on ModernBERT the current state of the art natural language text encoder featuring architectural upgrades such as rotary positional embeddings (RoPE), Flash Attention, and extended context length up to 8,192 tokens our model adapts these innovations specifically for biomedical and clinical domains. Clinical ModernBERT excels at producing semantically rich representations tailored for long context tasks. We validate this both by analyzing its pretrained weights and through empirical evaluation on a comprehensive suite of clinical NLP benchmarks.", "authors": ["Simon A. Lee", "Anthony Wu", "Jeffrey N. Chiang"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-04", "url": "https://arxiv.org/abs/2504.03964", "pdf_url": "https://arxiv.org/pdf/2504.03964v1", "arxiv_id": "2504.03964", "doi": "10.48550/arXiv.2504.03964", "citation_count": 41, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4058}
{"id": "c508aa8afc65069cf89789c167ee5db9da0ad8916ec9026ffe923fb9ff57165a", "sources": ["arxiv", "semantic_scholar"], "title": "InfiniteICL: Breaking the Limit of Context Window Size via Long Short-term Memory Transformation", "abstract": "In-context learning (ICL) is critical for large language models (LLMs), but its effectiveness is constrained by finite context windows, particularly in ultra-long contexts. To overcome this, we introduce InfiniteICL, a framework that parallels context and parameters in LLMs with short- and long-term memory in human cognitive systems, focusing on transforming temporary context knowledge into permanent parameter updates. This approach significantly reduces memory usage, maintains robust performance across varying input lengths, and theoretically enables infinite context integration through the principles of context knowledge elicitation, selection, and consolidation. Evaluations demonstrate that our method reduces context length by 90% while achieving 103% average performance of full-context prompting across fact recall, grounded reasoning, and skill acquisition tasks. When conducting sequential multi-turn transformations on complex, real-world contexts (with length up to 2M tokens), our approach surpasses full-context prompting while using only 0.4% of the original contexts. These findings highlight InfiniteICL's potential to enhance the scalability and efficiency of LLMs by breaking the limitations of conventional context window sizes.", "authors": ["Bowen Cao", "Deng Cai", "Wai Lam"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-02", "url": "https://arxiv.org/abs/2504.01707", "pdf_url": "https://arxiv.org/pdf/2504.01707v2", "arxiv_id": "2504.01707", "doi": "10.48550/arXiv.2504.01707", "citation_count": 15, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.301}
{"id": "7843775b13ae08048ad1f602b0d409450e6390361e298092f9087589722dfcd6", "sources": ["arxiv", "semantic_scholar"], "title": "On the Consistency of Multilingual Context Utilization in Retrieval-Augmented Generation", "abstract": "Retrieval-augmented generation (RAG) with large language models (LLMs) has demonstrated strong performance in multilingual question-answering (QA) tasks by leveraging relevant passages retrieved from corpora. In multilingual RAG (mRAG), the retrieved passages can be written in languages other than that of the query entered by the user, making it challenging for LLMs to effectively utilize the provided information. Recent research suggests that retrieving passages from multilingual corpora can improve RAG performance, particularly for low-resource languages. However, the extent to which LLMs can leverage different kinds of multilingual contexts to generate accurate answers, *independently from retrieval quality*, remains understudied. In this paper, we conduct an extensive assessment of LLMs' ability to (i) make consistent use of a relevant passage regardless of its language, (ii) respond in the expected language, and (iii) focus on the relevant passage even when multiple `distracting' passages in different languages are provided in the context. Our experiments with four LLMs across three QA datasets covering a total of 48 languages reveal a surprising ability of LLMs to extract the relevant information from passages in a different language than the query, but a much weaker ability to formulate a full answer in the correct language. Our analysis, based on both accuracy and feature attribution techniques, further shows that distracting passages negatively impact answer quality regardless of their language. However, distractors in the query language exert a slightly stronger influence. Taken together, our findings deepen the understanding of how LLMs utilize context in mRAG systems, providing directions for future improvements.", "authors": ["Jirui Qi", "Raquel Fernández", "Arianna Bisazza"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-04-01", "url": "https://arxiv.org/abs/2504.00597", "pdf_url": "https://arxiv.org/pdf/2504.00597v4", "arxiv_id": "2504.00597", "doi": "10.18653/v1/2025.mrl-main.15", "citation_count": 6, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Betswish/mRAG-Context-Consistency", "venue": null, "quality_score": 0.2113}
{"id": "c1888bed0e12191641d4f5dc0ac93cc3bc2948b1543494f192d5df4ac5f7d8d5", "sources": ["arxiv", "semantic_scholar"], "title": "Cocktail: Chunk-Adaptive Mixed-Precision Quantization for Long-Context LLM Inference", "abstract": "Recently, large language models (LLMs) have been able to handle longer and longer contexts. However, a context that is too long may cause intolerant inference latency and GPU memory usage. Existing methods propose mixed-precision quantization to the key-value (KV) cache in LLMs based on token granularity, which is time-consuming in the search process and hardware inefficient during computation. This paper introduces a novel approach called Cocktail, which employs chunk-adaptive mixed-precision quantization to optimize the KV cache. Cocktail consists of two modules: chunk-level quantization search and chunk-level KV cache computation. Chunk-level quantization search determines the optimal bitwidth configuration of the KV cache chunks quickly based on the similarity scores between the corresponding context chunks and the query, maintaining the model accuracy. Furthermore, chunk-level KV cache computation reorders the KV cache chunks before quantization, avoiding the hardware inefficiency caused by mixed-precision quantization in inference computation. Extensive experiments demonstrate that Cocktail outperforms state-of-the-art KV cache quantization methods on various models and datasets.", "authors": ["Wei Tao", "Bin Zhang", "Xiaoyang Qu", "Jiguang Wan", "Jianzong Wang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-30", "url": "https://arxiv.org/abs/2503.23294", "pdf_url": "https://arxiv.org/pdf/2503.23294v1", "arxiv_id": "2503.23294", "doi": "10.23919/DATE64628.2025.10992912", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Design, Automation and Test in Europe", "quality_score": 0.2258}
{"id": "ff6e9f95d1602f61e7dace252804e523e9e5716cb371c9a8aa5cfa82b5145552", "sources": ["arxiv", "semantic_scholar"], "title": "Context in object detection: a systematic literature review", "abstract": "Context is an important factor in computer vision as it offers valuable information to clarify and analyze visual data. Utilizing the contextual information inherent in an image or a video can improve the precision and effectiveness of object detectors. For example, where recognizing an isolated object might be challenging, context information can improve comprehension of the scene. This study explores the impact of various context-based approaches to object detection. Initially, we investigate the role of context in object detection and survey it from several perspectives. We then review and discuss the most recent context-based object detection approaches and compare them. Finally, we conclude by addressing research questions and identifying gaps for further studies. More than 265 publications are included in this survey, covering different aspects of context in different categories of object detection, including general object detection, video object detection, small object detection, camouflaged object detection, zero-shot, one-shot, and few-shot object detection. This literature review presents a comprehensive overview of the latest advancements in context-based object detection, providing valuable contributions such as a thorough understanding of contextual information and effective methods for integrating various context types into object detection, thus benefiting researchers.", "authors": ["Mahtab Jamali", "Paul Davidsson", "Reza Khoshkangini", "Martin Georg Ljungqvist", "Radu-Casian Mihailescu"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-29", "url": "https://arxiv.org/abs/2503.23249", "pdf_url": "https://arxiv.org/pdf/2503.23249v1", "arxiv_id": "2503.23249", "doi": "10.1007/s10462-025-11186-x", "citation_count": 25, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Artificial Intelligence Review", "quality_score": 0.3537}
{"id": "15d3de9662a1d5674c2ffcdd1ce5a516bc5ad8cdc85f44e3cb71dcec21fc68fc", "sources": ["arxiv", "semantic_scholar"], "title": "A Comprehensive Survey on Long Context Language Modeling", "abstract": "Efficient processing of long contexts has been a persistent pursuit in Natural Language Processing. With the growing number of long documents, dialogues, and other textual data, it is important to develop Long Context Language Models (LCLMs) that can process and analyze extensive inputs in an effective and efficient way. In this paper, we present a comprehensive survey on recent advances in long-context modeling for large language models. Our survey is structured around three key aspects: how to obtain effective and efficient LCLMs, how to train and deploy LCLMs efficiently, and how to evaluate and analyze LCLMs comprehensively. For the first aspect, we discuss data strategies, architectural designs, and workflow approaches oriented with long context processing. For the second aspect, we provide a detailed examination of the infrastructure required for LCLM training and inference. For the third aspect, we present evaluation paradigms for long-context comprehension and long-form generation, as well as behavioral analysis and mechanism interpretability of LCLMs. Beyond these three key aspects, we thoroughly explore the diverse application scenarios where existing LCLMs have been deployed and outline promising future development directions. This survey provides an up-to-date review of the literature on long-context LLMs, which we wish to serve as a valuable resource for both researchers and engineers. An associated GitHub repository collecting the latest papers and repos is available at: \\href{https://github.com/LCLM-Horizon/A-Comprehensive-Survey-For-Long-Context-Language-Modeling}{\\color[RGB]{175,36,67}{LCLM-Horizon}}.", "authors": ["Jiaheng Liu", "Dawei Zhu", "Zhiqi Bai", "Yancheng He", "Huanxuan Liao", "Haoran Que", "Zekun Wang", "Chenchen Zhang", "Ge Zhang", "Jiebin Zhang", "Yuanxing Zhang", "Zhuo Chen", "Hangyu Guo", "Shilong Li", "Ziqiang Liu", "Yong Shan", "Yifan Song", "Jiayi Tian", "Wenhao Wu", "Zhejian Zhou", "Ruijie Zhu", "Junlan Feng", "Yang Gao", "Shizhu He", "Zhoujun Li", "Tianyu Liu", "Fanyu Meng", "Wenbo Su", "Yingshui Tan", "Zili Wang", "Jian Yang", "Wei Ye", "Bo Zheng", "Wangchunshu Zhou", "Wenhao Huang", "Sujian Li", "Zhaoxiang Zhang"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-20", "url": "https://arxiv.org/abs/2503.17407", "pdf_url": "https://arxiv.org/pdf/2503.17407v2", "arxiv_id": "2503.17407", "doi": "10.48550/arXiv.2503.17407", "citation_count": 115, "influential_citation_count": 5, "has_code": true, "code_url": "https://github.com/LCLM-Horizon/A-Comprehensive-Survey-For-Long-Context-Language-Modeling}{\\color[RGB", "venue": "arXiv.org", "quality_score": 0.5161}
{"id": "07744d0e3f05f5cd94b38f3c2df832779a4ca2f9e611adefdc568f1629311fe9", "sources": ["arxiv", "semantic_scholar"], "title": "SkyLadder: Better and Faster Pretraining via Context Window Scheduling", "abstract": "Recent advancements in LLM pretraining have featured ever-expanding context windows to process longer sequences. However, our pilot study reveals that models pretrained with shorter context windows consistently outperform their long-context counterparts under a fixed token budget. This finding motivates us to explore an optimal context window scheduling strategy to better balance long-context capability with pretraining efficiency. To this end, we propose SkyLadder, a simple yet effective approach that implements a short-to-long context window transition. SkyLadder preserves strong standard benchmark performance, while matching or exceeding baseline results on long context tasks. Through extensive experiments, we pre-train 1B-parameter models (up to 32K context) and 3B-parameter models (8K context) on 100B tokens, demonstrating that SkyLadder yields consistent gains of up to 3.7% on common benchmarks, while achieving up to 22% faster training speeds compared to baselines. The code is at https://github.com/sail-sg/SkyLadder.", "authors": ["Tongyao Zhu", "Qian Liu", "Haonan Wang", "Shiqi Chen", "Xiangming Gu", "Tianyu Pang", "Min-Yen Kan"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-19", "url": "https://arxiv.org/abs/2503.15450", "pdf_url": "https://arxiv.org/pdf/2503.15450v2", "arxiv_id": "2503.15450", "doi": "10.48550/arXiv.2503.15450", "citation_count": 4, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/sail-sg/SkyLadder", "venue": "arXiv.org", "quality_score": 0.1747}
{"id": "a999dbaafa94390d114ff2260f4199987d3cc6b7c837d5be7f03a5ec446c00ae", "sources": ["arxiv", "semantic_scholar"], "title": "aiXcoder-7B-v2: Training LLMs to Fully Utilize the Long Context in Repository-level Code Completion", "abstract": "Large Language Models (LLMs) have shown promising results in repository-level code completion, which completes code based on the in-file and cross-file context of a repository. The cross-file context typically contains different types of information (e.g., relevant APIs and similar code) and is lengthy. In this paper, we found that LLMs struggle to fully utilize the information in the cross-file context. We hypothesize that one of the root causes of the limitation is the misalignment between pre-training (i.e., relying on nearby context) and repo-level code completion (i.e., frequently attending to long-range cross-file context). To address the above misalignment, we propose Code Long-context Alignment - COLA, a purely data-driven approach to explicitly teach LLMs to focus on the cross-file context. Specifically, COLA constructs a large-scale repo-level code completion dataset - COLA-132K, where each sample contains the long cross-file context (up to 128K tokens) and requires generating context-aware code (i.e., cross-file API invocations and code spans similar to cross-file context). Through a two-stage training pipeline upon COLA-132K, LLMs learn the capability of finding relevant information in the cross-file context, thus aligning LLMs with repo-level code completion. We apply COLA to multiple popular LLMs (e.g., aiXcoder-7B) and extensive experiments on COLA-132K and a public benchmark - CrossCodeEval. Our experiments yield the following results. 1) Effectiveness. COLA substantially improves the performance of multiple LLMs in repo-level code completion. For example, it improves aiXcoder-7B by up to 19.7% in exact match. 2) Generalizability. The capability learned by COLA can generalize to new languages. 3) Enhanced Context Utilization Capability. We design two probing experiments, which show COLA improves the capability of LLMs in utilizing the information in cross-file context.", "authors": ["Jia Li", "Hao Zhu", "Huanyu Liu", "Xianjie Shi", "He Zong", "Yihong Dong", "Kechi Zhang", "Siyuan Jiang", "Zhi Jin", "Ge Li"], "categories": ["cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-19", "url": "https://arxiv.org/abs/2503.15301", "pdf_url": "https://arxiv.org/pdf/2503.15301v2", "arxiv_id": "2503.15301", "doi": "10.1109/ASE63991.2025.00125", "citation_count": 13, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Automated Software Engineering", "quality_score": 0.2865}
{"id": "aca5628509dbcbe4be5527e94dad9a2a7d251b06c4aabadc324c6f94f4f7b07b", "sources": ["arxiv", "semantic_scholar"], "title": "A Survey on Transformer Context Extension: Approaches and Evaluation", "abstract": "Large language models (LLMs) based on Transformer have been widely applied in the filed of natural language processing (NLP), demonstrating strong performance, particularly in handling short text tasks. However, when it comes to long context scenarios, the performance of LLMs degrades due to some challenges. To alleviate this phenomenon, there is a number of work proposed recently. In this survey, we first list the challenges of applying pre-trained LLMs to process long contexts. Then systematically review the approaches related to long context and propose our taxonomy categorizing them into four main types: positional encoding, context compression, retrieval augmented, and attention pattern. In addition to the approaches, we focus on the evaluation of long context, organizing relevant data, tasks, and metrics based on existing long context benchmarks. Finally, we summarize unresolved issues in the long context domain and put forward our views on future developments.", "authors": ["Yijun Liu", "Jinzheng Yu", "Yang Xu", "Zhongyang Li", "Qingfu Zhu"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-17", "url": "https://arxiv.org/abs/2503.13299", "pdf_url": "https://arxiv.org/pdf/2503.13299v2", "arxiv_id": "2503.13299", "doi": "10.48550/arXiv.2503.13299", "citation_count": 15, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.301}
{"id": "57d7487f9566701a3230409c7cbf6a3d53fcdb9008e90902b5a0b9d388ab8453", "sources": ["arxiv", "semantic_scholar"], "title": "Atlas: Multi-Scale Attention Improves Long Context Image Modeling", "abstract": "Efficiently modeling massive images is a long-standing challenge in machine learning. To this end, we introduce Multi-Scale Attention (MSA). MSA relies on two key ideas, (i) multi-scale representations (ii) bi-directional cross-scale communication. MSA creates O(log N) scales to represent the image across progressively coarser features and leverages cross-attention to propagate information across scales. We then introduce Atlas, a novel neural network architecture based on MSA. We demonstrate that Atlas significantly improves the compute-performance tradeoff of long-context image modeling in a high-resolution variant of ImageNet 100. At 1024px resolution, Atlas-B achieves 91.04% accuracy, comparable to ConvNext-B (91.92%) while being 4.3x faster. Atlas is 2.95x faster and 7.38% better than FasterViT, 2.25x faster and 4.96% better than LongViT. In comparisons against MambaVision-S, we find Atlas-S achieves 5%, 16% and 32% higher accuracy at 1024px, 2048px and 4096px respectively, while obtaining similar runtimes. Code for reproducing our experiments and pretrained models is available at https://github.com/yalalab/atlas.", "authors": ["Kumar Krishna Agrawal", "Long Lian", "Longchao Liu", "Natalia Harguindeguy", "Boyi Li", "Alexander Bick", "Maggie Chung", "Trevor Darrell", "Adam Yala"], "categories": ["cs.CV", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-16", "url": "https://arxiv.org/abs/2503.12355", "pdf_url": "https://arxiv.org/pdf/2503.12355v1", "arxiv_id": "2503.12355", "doi": "10.48550/arXiv.2503.12355", "citation_count": 5, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/yalalab/atlas", "venue": "arXiv.org", "quality_score": 0.1945}
{"id": "778f942b9c33a112839b4f4b2280ab68b99358554e4ffcf8bc95419efd770df5", "sources": ["arxiv", "semantic_scholar"], "title": "StFT: Spatio-temporal Fourier Transformer for Long-term Dynamics Prediction", "abstract": "Simulating the long-term dynamics of multi-scale and multi-physics systems poses a significant challenge in understanding complex phenomena across science and engineering. The complexity arises from the intricate interactions between scales and the interplay of diverse physical processes, which manifest in PDEs through coupled, nonlinear terms that govern the evolution of multiple physical fields across scales. Neural operators have shown potential in short-term prediction of such complex spatio-temporal dynamics; however, achieving stable high-fidelity predictions and providing robust uncertainty quantification over extended time horizons remains an open and unsolved area of research. These limitations often lead to stability degradation with rapid error accumulation, particularly in long-term forecasting of systems characterized by multi-scale behaviors involving dynamics of different orders. To address these challenges, we propose an autoregressive Spatio-temporal Fourier Transformer (StFT), in which each transformer block is designed to learn the system dynamics at a distinct scale through a dual-path architecture that integrates frequency-domain and spatio-temporal representations. By leveraging a structured hierarchy of \\ours blocks, the resulting model explicitly captures the underlying dynamics across both macro- and micro- spatial scales. Furthermore, a generative residual correction mechanism is introduced to learn a probabilistic refinement temporally while simultaneously quantifying prediction uncertainties, enhancing both the accuracy and reliability of long-term probabilistic forecasting. Evaluations conducted on three benchmark datasets (plasma, fluid, and atmospheric dynamics) demonstrate the advantages of our approach over state-of-the-art ML methods.", "authors": ["Da Long", "Shandian Zhe", "Samuel Williams", "Leonid Oliker", "Zhe Bai"], "categories": ["cs.LG", "eess.SP"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2025-03-14", "url": "https://arxiv.org/abs/2503.11899", "pdf_url": "https://arxiv.org/pdf/2503.11899v2", "arxiv_id": "2503.11899", "doi": null, "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.0753}
{"id": "c73908a0afe37f2b45d82b77f9865189cfb864f6c7e0ef67deff4de3202c8485", "sources": ["arxiv", "semantic_scholar"], "title": "Long Context Tuning for Video Generation", "abstract": "Recent advances in video generation can produce realistic, minute-long single-shot videos with scalable diffusion transformers. However, real-world narrative videos require multi-shot scenes with visual and dynamic consistency across shots. In this work, we introduce Long Context Tuning (LCT), a training paradigm that expands the context window of pre-trained single-shot video diffusion models to learn scene-level consistency directly from data. Our method expands full attention mechanisms from individual shots to encompass all shots within a scene, incorporating interleaved 3D position embedding and an asynchronous noise strategy, enabling both joint and auto-regressive shot generation without additional parameters. Models with bidirectional attention after LCT can further be fine-tuned with context-causal attention, facilitating auto-regressive generation with efficient KV-cache. Experiments demonstrate single-shot models after LCT can produce coherent multi-shot scenes and exhibit emerging capabilities, including compositional generation and interactive shot extension, paving the way for more practical visual content creation. See https://guoyww.github.io/projects/long-context-video/ for more details.", "authors": ["Yuwei Guo", "Ceyuan Yang", "Ziyan Yang", "Zhibei Ma", "Zhijie Lin", "Zhenheng Yang", "Dahua Lin", "Lu Jiang"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-13", "url": "https://arxiv.org/abs/2503.10589", "pdf_url": "https://arxiv.org/pdf/2503.10589v1", "arxiv_id": "2503.10589", "doi": "10.1109/ICCV51701.2025.01605", "citation_count": 95, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "IEEE International Conference on Computer Vision", "quality_score": 0.4956}
{"id": "25d6b918bf7f16c4e539a31b8ab0d4a3e72cb2ea1d8cfe041df8bc4abfbdb6b5", "sources": ["arxiv", "semantic_scholar"], "title": "Weighted Tensor Decompositions for Context-aware Collaborative Filtering", "abstract": "Over recent years it has become well accepted that user interest is not static or immutable. There are a variety of contextual factors, such as time of day, the weather or the user's mood, that influence the current interests of the user. Modelling approaches need to take these factors into account if they want to succeed at finding the most relevant content to recommend given the situation. A popular method for context-aware recommendation is to encode context attributes as extra dimensions of the classic user-item interaction matrix, effectively turning it into a tensor, followed by applying the appropriate tensor decomposition methods to learn missing values. However, unlike with matrix factorization, where all decompositions are essentially a product of matrices, there exist many more options for decomposing tensors by combining vector, matrix and tensor products. We study the most successful decomposition methods that use weighted square loss and categorize them based on their tensor structure and regularization strategy. Additionally, we further extend the pool of methods by filling in the missing combinations. In this paper we provide an overview of the properties of the different decomposition methods, such as their complexity, scalability, and modelling capacity. These benefits are then contrasted with the performances achieved in offline experiments to gain more insight into which method to choose depending on a specific situation and constraints.", "authors": ["Joey De Pauw", "Bart Goethals"], "categories": ["cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-11", "url": "https://arxiv.org/abs/2503.08393", "pdf_url": "https://arxiv.org/pdf/2503.08393v2", "arxiv_id": "2503.08393", "doi": "10.48550/arXiv.2503.08393", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1193}
{"id": "32835bee7e2df09686da56149dcb98445fb5d0abe09139d23cd982ced3b3d272", "sources": ["arxiv", "semantic_scholar"], "title": "Layer-Specific Scaling of Positional Encodings for Superior Long-Context Modeling", "abstract": "Although large language models (LLMs) have achieved significant progress in handling long-context inputs, they still suffer from the ``lost-in-the-middle'' problem, where crucial information in the middle of the context is often underrepresented or lost. Our extensive experiments reveal that this issue may arise from the rapid long-term decay in Rotary Position Embedding (RoPE). To address this problem, we propose a layer-specific positional encoding scaling method that assigns distinct scaling factors to each layer, slowing down the decay rate caused by RoPE to make the model pay more attention to the middle context. A specially designed genetic algorithm is employed to efficiently select the optimal scaling factors for each layer by incorporating Bezier curves to reduce the search space. Through comprehensive experimentation, we demonstrate that our method significantly alleviates the ``lost-in-the-middle'' problem. Our approach results in an average accuracy improvement of up to 20% on the Key-Value Retrieval dataset. Furthermore, we show that layer-specific interpolation, as opposed to uniform interpolation across all layers, enhances the model's extrapolation capabilities when combined with PI and Dynamic-NTK positional encoding schemes.", "authors": ["Zhenghua Wang", "Yiran Ding", "Changze Lv", "Zhibo Xu", "Tianlong Li", "Tianyuan Shi", "Xiaoqing Zheng", "Xuanjing Huang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-06", "url": "https://arxiv.org/abs/2503.04355", "pdf_url": "https://arxiv.org/pdf/2503.04355v1", "arxiv_id": "2503.04355", "doi": "10.48550/arXiv.2503.04355", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0802}
{"id": "04bb6014c0b5d9fe20ec75264d4fde34909e2192400e7850369a4130426f75b1", "sources": ["arxiv", "semantic_scholar"], "title": "Shifting Long-Context LLMs Research from Input to Output", "abstract": "Recent advancements in long-context Large Language Models (LLMs) have primarily concentrated on processing extended input contexts, resulting in significant strides in long-context comprehension. However, the equally critical aspect of generating long-form outputs has received comparatively less attention. This paper advocates for a paradigm shift in NLP research toward addressing the challenges of long-output generation. Tasks such as novel writing, long-term planning, and complex reasoning require models to understand extensive contexts and produce coherent, contextually rich, and logically consistent extended text. These demands highlight a critical gap in current LLM capabilities. We underscore the importance of this under-explored domain and call for focused efforts to develop foundational LLMs tailored for generating high-quality, long-form outputs, which hold immense potential for real-world applications.", "authors": ["Yuhao Wu", "Yushi Bai", "Zhiqing Hu", "Shangqing Tu", "Ming Shan Hee", "Juanzi Li", "Roy Ka-Wei Lee"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-06", "url": "https://arxiv.org/abs/2503.04723", "pdf_url": "https://arxiv.org/pdf/2503.04723v2", "arxiv_id": "2503.04723", "doi": "10.48550/arXiv.2503.04723", "citation_count": 16, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3076}
{"id": "3446ff4844894828dbe72286dfc5a9ea8e257d1b2cd703b807b494fdf67a86ed", "sources": ["arxiv", "semantic_scholar"], "title": "LADM: Long-context Training Data Selection with Attention-based Dependency Measurement for LLMs", "abstract": "Long-context modeling has drawn more and more attention in the area of Large Language Models (LLMs). Continual training with long-context data becomes the de-facto method to equip LLMs with the ability to process long inputs. However, it still remains an open challenge to measure the quality of long-context training data. To address this issue, we propose a Long-context data selection framework with Attention-based Dependency Measurement (LADM), which can efficiently identify high-quality long-context data from a large-scale, multi-domain pre-training corpus. LADM leverages the retrieval capabilities of the attention mechanism to capture contextual dependencies, ensuring a comprehensive quality measurement of long-context data. Experimental results show that our LADM framework significantly boosts the performance of LLMs on multiple long-context tasks with only 1B tokens for continual training.", "authors": ["Jianghao Chen", "Junhong Wu", "Yangyifan Xu", "Jiajun Zhang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-04", "url": "https://arxiv.org/abs/2503.02502", "pdf_url": "https://arxiv.org/pdf/2503.02502v3", "arxiv_id": "2503.02502", "doi": "10.48550/arXiv.2503.02502", "citation_count": 12, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/ZNLP/LADM", "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.2785}
{"id": "d70d2061490daf367ca77be548735b1120ea23f56c50b7ab4b6db85129f50e7e", "sources": ["arxiv", "semantic_scholar"], "title": "Understanding the formation of Saturn's regular moons in the context of giant planet moons formation scenarios", "abstract": "This article explores the different formation scenarios of the Kronian moons system in the context of a highly dissipative Saturn, with the objective of identifying the most likely of these scenarios. First, we review the diversity of objects - moons and rings - orbiting solar system giant planets, and the diversity of their architectures, which formation scenarios must reproduce. We then identify in this broader context the specific features of the Saturn system, such as the particularly large spectrum of its moon masses, the uniqueness of Titan and the presence of both dense and tenuous rings, before discussing the applicability of the different giant planet moon formation scenarios to the Saturn case. We discuss each of the most relevant scenarios and their respective merits. Finally, we tentatively propose a \"favorite\" scenario and we identify the key observations to be made by future space missions and/or Earth-based telescopic observations to validate this scenario or possibly alternative ones.", "authors": ["Michel Blanc", "Aurélien Crida", "Yuhito Shibaike", "Sebastien Charnoz", "Maryame El Moutamid", "Paul Estrada", "Olivier Mousis", "Julien Salmon", "Antoine Schneeberger", "Pierre Vernazza"], "categories": ["astro-ph.EP"], "fields_of_study": ["Medicine", "Physics"], "published_date": "2025-03-03", "url": "https://arxiv.org/abs/2503.01351", "pdf_url": "https://arxiv.org/pdf/2503.01351v1", "arxiv_id": "2503.01351", "doi": "10.1007/s11214-025-01156-8", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Space Science Reviews", "quality_score": 0.2258}
{"id": "e309720ef64a0d85743df660153304749b6ccee1bd5dade630a371594c0c958c", "sources": ["arxiv", "semantic_scholar"], "title": "U-NIAH: Unified RAG and LLM Evaluation for Long Context Needle-In-A-Haystack", "abstract": "Recent advancements in Large Language Models (LLMs) have expanded their context windows to unprecedented lengths, sparking debates about the necessity of Retrieval-Augmented Generation (RAG). To address the fragmented evaluation paradigms and limited cases in existing Needle-in-a-Haystack (NIAH), this paper introduces U-NIAH, a unified framework that systematically compares LLMs and RAG methods in controlled long context settings. Our framework extends beyond traditional NIAH by incorporating multi-needle, long-needle, and needle-in-needle configurations, along with different retrieval settings, while leveraging the synthetic Starlight Academy dataset-a fictional magical universe-to eliminate biases from pre-trained knowledge. Through extensive experiments, we investigate three research questions: (1) performance trade-offs between LLMs and RAG, (2) error patterns in RAG, and (3) RAG's limitations in complex settings. Our findings show that RAG significantly enhances smaller LLMs by mitigating the \"lost-in-the-middle\" effect and improving robustness, achieving an 82.58% win-rate over LLMs. However, we observe that retrieval noise and reverse chunk ordering degrade performance, while surprisingly, advanced reasoning LLMs exhibit reduced RAG compatibility due to sensitivity to semantic distractors. We identify typical error patterns including omission due to noise, hallucination under high noise critical condition, and self-doubt behaviors. Our work not only highlights the complementary roles of RAG and LLMs, but also provides actionable insights for optimizing deployments. Code: https://github.com/Tongji-KGLLM/U-NIAH.", "authors": ["Yunfan Gao", "Yun Xiong", "Wenlong Wu", "Zijing Huang", "Bohan Li", "Haofen Wang"], "categories": ["cs.CL", "cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2025-03-01", "url": "https://arxiv.org/abs/2503.00353", "pdf_url": "https://arxiv.org/pdf/2503.00353v1", "arxiv_id": "2503.00353", "doi": "10.1145/3786609", "citation_count": 19, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/Tongji-KGLLM/U-NIAH", "venue": null, "quality_score": 0.3253}
{"id": "02b339afa8f8823abfa199f5864f771da2a0d210c1b336c959494fff19eb0c97", "sources": ["arxiv", "semantic_scholar"], "title": "LongRoPE2: Near-Lossless LLM Context Window Scaling", "abstract": "LongRoPE2 is a novel approach that extends the effective context window of pre-trained large language models (LLMs) to the target length, while preserving the performance on the original shorter context window. This is achieved by three contributions: (1) a hypothesis that insufficient training in higher RoPE dimensions contributes to the persistent out-of-distribution (OOD) issues observed in existing methods; (2) an effective RoPE rescaling algorithm that adopts evolutionary search guided by \"needle-driven\" perplexity to address the insufficient training problem; (3) a mixed context window training approach that fine-tunes model weights to adopt rescaled RoPE for long-context sequences while preserving the short-context performance with the original RoPE. Extensive experiments on LLaMA3-8B and Phi3-mini-3.8B across various benchmarks validate the hypothesis and demonstrate the effectiveness of LongRoPE2. Remarkably, LongRoPE2 extends LLaMA3-8B to achieve a 128K effective context length while retaining over 98.5% of short-context performance, using only 10B tokens -- 80x fewer than Meta's approach, which fails to reach the target effective context length. Code will be available at https://github.com/microsoft/LongRoPE.", "authors": ["Ning Shang", "Li Lyna Zhang", "Siyuan Wang", "Gaokai Zhang", "Gilsinia Lopez", "Fan Yang", "Weizhu Chen", "Mao Yang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-27", "url": "https://arxiv.org/abs/2502.20082", "pdf_url": "https://arxiv.org/pdf/2502.20082v1", "arxiv_id": "2502.20082", "doi": "10.48550/arXiv.2502.20082", "citation_count": 16, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/microsoft/LongRoPE", "venue": "International Conference on Machine Learning", "quality_score": 0.3076}
{"id": "27dd55bb86758a445f506be407177a7536af7729ae5d5a5622f20b4579490eb3", "sources": ["arxiv", "semantic_scholar"], "title": "Thus Spake Long-Context Large Language Model", "abstract": "Long context is an important topic in Natural Language Processing (NLP), running through the development of NLP architectures, and offers immense opportunities for Large Language Models (LLMs), giving LLMs the lifelong learning potential akin to humans. Unfortunately, the pursuit of a long context is accompanied by numerous obstacles. Nevertheless, long context remains a core competitive advantage for LLMs. In the past two years, the context length of LLMs has achieved a breakthrough extension to millions of tokens. Moreover, research on long-context LLMs has expanded beyond length extrapolation to a comprehensive focus on architecture, infrastructure, training, and evaluation technologies. Inspired by the symphonic poem, Thus Spake Zarathustra, we draw an analogy between the journey of extending the context of LLM and the attempts of humans to transcend their mortality. In this survey, we will illustrate how LLM struggles between the tremendous need for a longer context and its equal need to accept the fact that it is ultimately finite. To achieve this, we give a global picture of the lifecycle of long-context LLMs from four perspectives: architecture, infrastructure, training, and evaluation, showcasing the full spectrum of long-context technologies. At the end of this survey, we will present 10 unanswered questions currently faced by long-context LLMs. We hope this survey can serve as a systematic introduction to research on long-context LLMs. Video: https://www.bilibili.com/video/BV11h9AYoEYj. Github: https://github.com/OpenMOSS/Thus-Spake-Long-Context-LLM.", "authors": ["Xiaoran Liu", "Ruixiao Li", "Mianqiu Huang", "Zhigeng Liu", "Yuerong Song", "Qipeng Guo", "Siyang He", "Qiqi Wang", "Linlin Li", "Qun Liu", "Ziwei He", "Yaqian Zhou", "Xuanjing Huang", "Xipeng Qiu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-24", "url": "https://arxiv.org/abs/2502.17129", "pdf_url": "https://arxiv.org/pdf/2502.17129v2", "arxiv_id": "2502.17129", "doi": "10.48550/arXiv.2502.17129", "citation_count": 10, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/OpenMOSS/Thus-Spake-Long-Context-LLM", "venue": "arXiv.org", "quality_score": 0.2603}
{"id": "8565ed8c20617afa200bf3a1f48139c7b8df3fd5dfbb1d93d259d472c40e6fb0", "sources": ["arxiv", "semantic_scholar"], "title": "LongSafety: Evaluating Long-Context Safety of Large Language Models", "abstract": "As Large Language Models (LLMs) continue to advance in understanding and generating long sequences, new safety concerns have been introduced through the long context. However, the safety of LLMs in long-context tasks remains under-explored, leaving a significant gap in both evaluation and improvement of their safety. To address this, we introduce LongSafety, the first comprehensive benchmark specifically designed to evaluate LLM safety in open-ended long-context tasks. LongSafety encompasses 7 categories of safety issues and 6 user-oriented long-context tasks, with a total of 1,543 test cases, averaging 5,424 words per context. Our evaluation towards 16 representative LLMs reveals significant safety vulnerabilities, with most models achieving safety rates below 55%. Our findings also indicate that strong safety performance in short-context scenarios does not necessarily correlate with safety in long-context tasks, emphasizing the unique challenges and urgency of improving long-context safety. Moreover, through extensive analysis, we identify challenging safety issues and task types for long-context models. Furthermore, we find that relevant context and extended input sequences can exacerbate safety risks in long-context scenarios, highlighting the critical need for ongoing attention to long-context safety challenges. Our code and data are available at https://github.com/thu-coai/LongSafety.", "authors": ["Yida Lu", "Jiale Cheng", "Zhexin Zhang", "Shiyao Cui", "Cunxiang Wang", "Xiaotao Gu", "Yuxiao Dong", "Jie Tang", "Hongning Wang", "Minlie Huang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-24", "url": "https://arxiv.org/abs/2502.16971", "pdf_url": "https://arxiv.org/pdf/2502.16971v1", "arxiv_id": "2502.16971", "doi": "10.48550/arXiv.2502.16971", "citation_count": 8, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/thu-coai/LongSafety", "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.2386}
{"id": "31960dfaeacdb5e957db064678fec882e7cf2fa3139001d3d3fdd58a77b2491b", "sources": ["arxiv", "semantic_scholar"], "title": "SQLong: Enhanced NL2SQL for Longer Contexts with LLMs", "abstract": "Open-weight large language models (LLMs) have significantly advanced performance in the Natural Language to SQL (NL2SQL) task. However, their effectiveness diminishes when dealing with large database schemas, as the context length increases. To address this limitation, we present SQLong, a novel and efficient data augmentation framework designed to enhance LLM performance in long-context scenarios for the NL2SQL task. SQLong generates augmented datasets by extending existing database schemas with additional synthetic CREATE TABLE commands and corresponding data rows, sampled from diverse schemas in the training data. This approach effectively simulates long-context scenarios during finetuning and evaluation. Through experiments on the Spider and BIRD datasets, we demonstrate that LLMs finetuned with SQLong-augmented data significantly outperform those trained on standard datasets. These imply SQLong's practical implementation and its impact on improving NL2SQL capabilities in real-world settings with complex database schemas.", "authors": ["Dai Quoc Nguyen", "Cong Duy Vu Hoang", "Duy Vu", "Gioacchino Tangari", "Thanh Tien Vu", "Don Dharmasiri", "Yuan-Fang Li", "Long Duong"], "categories": ["cs.CL", "cs.AI", "cs.LG", "cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-23", "url": "https://arxiv.org/abs/2502.16747", "pdf_url": "https://arxiv.org/pdf/2502.16747v2", "arxiv_id": "2502.16747", "doi": "10.48550/arXiv.2502.16747", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1945}
{"id": "792f3673115d07a3b8160336a2b020020c14425929fef64605c2a9563834703c", "sources": ["arxiv", "semantic_scholar"], "title": "WildLong: Synthesizing Realistic Long-Context Instruction Data at Scale", "abstract": "Large language models (LLMs) with extended context windows enable tasks requiring extensive information integration but are limited by the scarcity of high-quality, diverse datasets for long-context instruction tuning. Existing data synthesis methods focus narrowly on objectives like fact retrieval and summarization, restricting their generalizability to complex, real-world tasks. WildLong extracts meta-information from real user queries, models co-occurrence relationships via graph-based methods, and employs adaptive generation to produce scalable data. It extends beyond single-document tasks to support multi-document reasoning, such as cross-document comparison and aggregation. Our models, finetuned on 150K instruction-response pairs synthesized using WildLong, surpasses existing open-source long-context-optimized models across benchmarks while maintaining strong performance on short-context tasks without incorporating supplementary short-context data. By generating a more diverse and realistic long-context instruction dataset, WildLong enhances LLMs' ability to generalize to complex, real-world reasoning over long contexts, establishing a new paradigm for long-context data synthesis.", "authors": ["Jiaxi Li", "Xingxing Zhang", "Xun Wang", "Xiaolong Huang", "Li Dong", "Liang Wang", "Si-Qing Chen", "Wei Lu", "Furu Wei"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-23", "url": "https://arxiv.org/abs/2502.16684", "pdf_url": "https://arxiv.org/pdf/2502.16684v1", "arxiv_id": "2502.16684", "doi": "10.48550/arXiv.2502.16684", "citation_count": 6, "influential_citation_count": 1, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2113}
{"id": "617caacf31e8e69a51e1bf52fcdb07f95d99a95fd117e335c6e241c4f66c318c", "sources": ["arxiv", "semantic_scholar"], "title": "Generalizing From Short to Long: Effective Data Synthesis for Long-Context Instruction Tuning", "abstract": "Long-context modelling for large language models (LLMs) has been a key area of recent research because many real world use cases require reasoning over longer inputs such as documents. The focus of research into modelling long context has been on how to model position and there has been little investigation into other important aspects of language modelling such as instruction tuning. Long context training examples are challenging and expensive to create and use. In this paper, we investigate how to design instruction data for the post-training phase of a long context pre-trained model: how much and what type of context is needed for optimal and efficient post-training. Our controlled study reveals that models instruction-tuned on short contexts can effectively generalize to longer ones, while also identifying other critical factors such as instruction difficulty and context composition. Based on these findings, we propose context synthesis, a novel data synthesis framework that leverages off-the-shelf LLMs to generate extended background contexts for high-quality instruction-answer pairs. Experiment results on the document-level benchmark (LongBench) demonstrate that our proposed approach outperforms previous instruction synthesis approaches and comes close to the performance of human-annotated long-context instruction data. The project will be available at: https://github.com/NJUNLP/context-synthesis.", "authors": ["Wenhao Zhu", "Pinzhen Chen", "Hanxu Hu", "Shujian Huang", "Fei Yuan", "Jiajun Chen", "Alexandra Birch"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-21", "url": "https://arxiv.org/abs/2502.15592", "pdf_url": "https://arxiv.org/pdf/2502.15592v1", "arxiv_id": "2502.15592", "doi": "10.48550/arXiv.2502.15592", "citation_count": 7, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/NJUNLP/context-synthesis", "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "fe0b5a91f4a8705ecc8bb3d7bd01e64cb242e19775379a2b1a65a1dfa6bc1c15", "sources": ["arxiv", "semantic_scholar"], "title": "Self-Taught Agentic Long Context Understanding", "abstract": "Answering complex, long-context questions remains a major challenge for large language models (LLMs) as it requires effective question clarifications and context retrieval. We propose Agentic Long-Context Understanding (AgenticLU), a framework designed to enhance an LLM's understanding of such queries by integrating targeted self-clarification with contextual grounding within an agentic workflow. At the core of AgenticLU is Chain-of-Clarifications (CoC), where models refine their understanding through self-generated clarification questions and corresponding contextual groundings. By scaling inference as a tree search where each node represents a CoC step, we achieve 97.8% answer recall on NarrativeQA with a search depth of up to three and a branching factor of eight. To amortize the high cost of this search process to training, we leverage the preference pairs for each step obtained by the CoC workflow and perform two-stage model finetuning: (1) supervised finetuning to learn effective decomposition strategies, and (2) direct preference optimization to enhance reasoning quality. This enables AgenticLU models to generate clarifications and retrieve relevant context effectively and efficiently in a single inference pass. Extensive experiments across seven long-context tasks demonstrate that AgenticLU significantly outperforms state-of-the-art prompting methods and specialized long-context LLMs, achieving robust multi-hop reasoning while sustaining consistent performance as context length grows.", "authors": ["Yufan Zhuang", "Xiaodong Yu", "Jialian Wu", "Ximeng Sun", "Ze Wang", "Jiang Liu", "Yusheng Su", "Jingbo Shang", "Zicheng Liu", "Emad Barsoum"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-21", "url": "https://arxiv.org/abs/2502.15920", "pdf_url": "https://arxiv.org/pdf/2502.15920v2", "arxiv_id": "2502.15920", "doi": "10.48550/arXiv.2502.15920", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.1945}
{"id": "0291d1e470918ff9087231da0b3fc2079fdfe086ca638e96c61c7095a55afb96", "sources": ["arxiv", "semantic_scholar"], "title": "LIFT: A Novel Framework for Enhancing Long-Context Understanding of LLMs via Long Input Fine-Tuning", "abstract": "Long context understanding remains challenging for large language models due to their limited context windows. This paper introduces Long Input Fine-Tuning (LIFT), a novel framework for long-context modeling that can enhance the long-context performance of arbitrary short-context LLMs by dynamically adapting their parameters to the given long input. Importantly, rather than endlessly extending the context window size to accommodate increasingly longer inputs in context, LIFT stores and absorbs the long input in parameters. By fine-tuning the long input into model parameters, LIFT allows short-context LLMs to answer questions even when the required information is not provided in the context during inference, avoiding the quadratic complexity w.r.t. input length of a normal long context model. Furthermore, LIFT does not simply perform continued pretraining on new, long contexts, but leverages carefully designed LLM-generated synthetic tasks to enhance the comprehension of long contexts, moving beyond mere memorization. To accommodate the additional cost of fine-tuning, we design a highly optimized pipeline that reduces the Time to First Token (TTFT) to less than 10 seconds for 8k context. We further provide a comprehensive analysis of LIFT's strengths and limitations in long-context understanding, discuss its feasibility for large-scale real-world deployment, and highlight valuable directions for future research.", "authors": ["Yansheng Mao", "Yufei Xu", "Jiaqi Li", "Fanxu Meng", "Haotong Yang", "Zilong Zheng", "Xiyuan Wang", "Muhan Zhang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-20", "url": "https://arxiv.org/abs/2502.14644", "pdf_url": "https://arxiv.org/pdf/2502.14644v5", "arxiv_id": "2502.14644", "doi": null, "citation_count": 7, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2258}
{"id": "26f5e94752568b1c058be3de6af1d9dfccc87462e018528e0e29ffb0b2ecbc1e", "sources": ["arxiv", "semantic_scholar"], "title": "Activation-aware Probe-Query: Effective Key-Value Retrieval for Long-Context LLMs Inference", "abstract": "Recent advances in large language models (LLMs) have showcased exceptional performance in long-context tasks, while facing significant inference efficiency challenges with limited GPU memory. Existing solutions first proposed the sliding-window approach to accumulate a set of historical \\textbf{key-value} (KV) pairs for reuse, then further improvements selectively retain its subsets at each step. However, due to the sparse attention distribution across a long context, it is hard to identify and recall relevant KV pairs, as the attention is distracted by massive candidate pairs. Additionally, we found it promising to select representative tokens as probe-Query in each sliding window to effectively represent the entire context, which is an approach overlooked by existing methods. Thus, we propose \\textbf{ActQKV}, a training-free, \\textbf{Act}ivation-aware approach that dynamically determines probe-\\textbf{Q}uery and leverages it to retrieve the relevant \\textbf{KV} pairs for inference. Specifically, ActQKV monitors a token-level indicator, Activation Bias, within each context window, enabling the proper construction of probe-Query for retrieval at pre-filling stage. To accurately recall the relevant KV pairs and minimize the irrelevant ones, we design a dynamic KV cut-off mechanism guided by information density across layers at the decoding stage. Experiments on the Long-Bench and $\\infty$ Benchmarks demonstrate its state-of-the-art performance with competitive inference quality and resource efficiency.", "authors": ["Qingfa Xiao", "Jiachuan Wang", "Haoyang Li", "Cheng Deng", "Jiaqi Tang", "Shuangyin Li", "Yongqi Zhang", "Jun Wang", "Lei Chen"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-19", "url": "https://arxiv.org/abs/2502.13542", "pdf_url": "https://arxiv.org/pdf/2502.13542v1", "arxiv_id": "2502.13542", "doi": "10.48550/arXiv.2502.13542", "citation_count": 2, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1505}
{"id": "091118f742aa060b0c349a8d59282fa49f9482527a5480be0f97d5a546f077b0", "sources": ["arxiv", "semantic_scholar"], "title": "LongPO: Long Context Self-Evolution of Large Language Models through Short-to-Long Preference Optimization", "abstract": "Large Language Models (LLMs) have demonstrated remarkable capabilities through pretraining and alignment. However, superior short-context LLMs may underperform in long-context scenarios due to insufficient long-context alignment. This alignment process remains challenging due to the impracticality of human annotation for extended contexts and the difficulty in balancing short- and long-context performance. To address these challenges, we introduce LongPO, that enables short-context LLMs to self-evolve to excel on long-context tasks by internally transferring short-context capabilities. LongPO harnesses LLMs to learn from self-generated short-to-long preference data, comprising paired responses generated for identical instructions with long-context inputs and their compressed short-context counterparts, respectively. This preference reveals capabilities and potentials of LLMs cultivated during short-context alignment that may be diminished in under-aligned long-context scenarios. Additionally, LongPO incorporates a short-to-long KL constraint to mitigate short-context performance decline during long-context alignment. When applied to Mistral-7B-Instruct-v0.2 from 128K to 512K context lengths, LongPO fully retains short-context performance and largely outperforms naive SFT and DPO in both long- and short-context tasks. Specifically, LongPO-trained models can achieve results on long-context benchmarks comparable to, or even surpassing, those of superior LLMs (e.g., GPT-4-128K) that involve extensive long-context annotation and larger parameter scales. Our code is available at https://github.com/DAMO-NLP-SG/LongPO.", "authors": ["Guanzheng Chen", "Xin Li", "Michael Qizhe Shieh", "Lidong Bing"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-19", "url": "https://arxiv.org/abs/2502.13922", "pdf_url": "https://arxiv.org/pdf/2502.13922v3", "arxiv_id": "2502.13922", "doi": "10.48550/arXiv.2502.13922", "citation_count": 24, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/DAMO-NLP-SG/LongPO", "venue": "International Conference on Learning Representations", "quality_score": 0.3495}
{"id": "cebf17bcd0b9a83af058be900c3859a0659abdf0956ef3a082d8429af69c4b79", "sources": ["arxiv", "semantic_scholar"], "title": "Facilitating Long Context Understanding via Supervised Chain-of-Thought Reasoning", "abstract": "Recent advances in Large Language Models (LLMs) have enabled them to process increasingly longer sequences, ranging from 2K to 2M tokens and even beyond. However, simply extending the input sequence length does not necessarily lead to effective long-context understanding. In this study, we integrate Chain-of-Thought (CoT) reasoning into LLMs in a supervised manner to facilitate effective long-context understanding. To achieve this, we introduce LongFinanceQA, a synthetic dataset in the financial domain designed to improve long-context reasoning. Unlike existing long-context synthetic data, LongFinanceQA includes intermediate CoT reasoning before the final conclusion, which encourages LLMs to perform explicit reasoning, improving accuracy and interpretability in long-context understanding. To generate synthetic CoT reasoning, we propose Property-based Agentic Inference (PAI), an agentic framework that simulates human-like reasoning steps, including property extraction, retrieval, and summarization. We evaluate PAI's reasoning capabilities by assessing GPT-4o-mini w/ PAI on the Loong benchmark, outperforming standard GPT-4o-mini by 20.0%. Furthermore, we fine-tune LLaMA-3.1-8B-Instruct on LongFinanceQA, achieving a 28.0% gain on Loong's financial subset.", "authors": ["Jingyang Lin", "Andy Wong", "Tian Xia", "Shenghua He", "Hui Wei", "Mei Han", "Jiebo Luo"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-18", "url": "https://arxiv.org/abs/2502.13127", "pdf_url": "https://arxiv.org/pdf/2502.13127v2", "arxiv_id": "2502.13127", "doi": "10.48550/arXiv.2502.13127", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.1945}
{"id": "a93d6eefc19953c4b8627c83ecb33c59e052eb7ce53fbb769559c3115fdab322", "sources": ["arxiv", "semantic_scholar"], "title": "A Dual-Stage Time-Context Network for Speech-Based Alzheimer's Disease Detection", "abstract": "Alzheimer's disease (AD) is a progressive neurodegenerative disorder that leads to irreversible cognitive decline in memory and communication. Early detection of AD through speech analysis is crucial for delaying disease progression. However, existing methods mainly use pre-trained acoustic models for feature extraction but have limited ability to model both local and global patterns in long-duration speech. In this letter, we introduce a Dual-Stage Time-Context Network (DSTC-Net) for speech-based AD detection, integrating local acoustic features with global conversational context in long-duration recordings.We first partition each long-duration recording into fixed-length segments to reduce computational overhead and preserve local temporal details.Next, we feed these segments into an Intra-Segment Temporal Attention (ISTA) module, where a bidirectional Long Short-Term Memory (BiLSTM) network with frame-level attention extracts enhanced local features.Subsequently, a Cross-Segment Context Attention (CSCA) module applies convolution-based context modeling and adaptive attention to unify global patterns across all segments.Extensive experiments on the ADReSSo dataset show that our DSTC-Net outperforms state-of-the-art models, reaching 83.10% accuracy and 83.15% F1.", "authors": ["Yifan Gao", "Long Guo", "Hong Liu"], "categories": ["cs.SD"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-18", "url": "https://arxiv.org/abs/2502.13064", "pdf_url": "https://arxiv.org/pdf/2502.13064v1", "arxiv_id": "2502.13064", "doi": "10.1109/LSP.2026.3655341", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Signal Processing Letters", "quality_score": 0.0753}
{"id": "e91dcf1f18abe9db213be23abc4b3ac4cfdf181a19f8a17ded587ef7ff1f37a9", "sources": ["arxiv", "semantic_scholar"], "title": "Emulating Retrieval Augmented Generation via Prompt Engineering for Enhanced Long Context Comprehension in LLMs", "abstract": "This paper addresses the challenge of comprehending very long contexts in Large Language Models (LLMs) by proposing a method that emulates Retrieval Augmented Generation (RAG) through specialized prompt engineering and chain-of-thought (CoT) reasoning. While recent LLMs support over 100,000 tokens in a single prompt, simply enlarging context windows has not guaranteed robust multi-hop reasoning when key details are scattered across massive input. Our approach treats the model as both the retriever and the reasoner: it first tags relevant segments within a long passage, then employs a stepwise CoT workflow to integrate these pieces of evidence. This single-pass method thereby reduces reliance on an external retriever, yet maintains focus on crucial segments. We evaluate our approach on selected tasks from BABILong, which interleaves standard bAbI QA problems with large amounts of distractor text. Compared to baseline (no retrieval) and naive RAG pipelines, our approach more accurately handles multi-fact questions such as object location tracking, counting, and indefinite knowledge. Furthermore, we analyze how prompt structure, including the order of question, relevant-text tags, and overall instructions, significantly affects performance. These findings underscore that optimized prompt engineering, combined with guided reasoning, can enhance LLMs' long-context comprehension and serve as a lightweight alternative to traditional retrieval pipelines.", "authors": ["Joon Park", "Kyohei Atarashi", "Koh Takeuchi", "Hisashi Kashima"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-18", "url": "https://arxiv.org/abs/2502.12462", "pdf_url": "https://arxiv.org/pdf/2502.12462v1", "arxiv_id": "2502.12462", "doi": "10.48550/arXiv.2502.12462", "citation_count": 11, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2698}
{"id": "1ea03ad1e7c207b1c52fd64b43675af746d88cbf2fc2f01c716afeabb9fb5be8", "sources": ["arxiv", "semantic_scholar"], "title": "Infinite Retrieval: Attention Enhanced LLMs in Long-Context Processing", "abstract": "Limited by the context window size of Large Language Models(LLMs), handling various tasks with input tokens exceeding the upper limit has been challenging, whether it is a simple direct retrieval task or a complex multi-hop reasoning task. Although various methods have been proposed to enhance the long-context processing capabilities of LLMs, they either incur substantial post-training costs, or require additional tool modules(e.g.,RAG), or have not shown significant improvement in realistic tasks. Our work observes the correlation between the attention distribution and generated answers across each layer, and establishes the attention allocation aligns with retrieval-augmented capabilities through experiments. Drawing on the above insights, we propose a novel method InfiniRetri that leverages the LLMs's own attention information to enable accurate retrieval across inputs of infinitely length. Our evaluations indicate that InfiniRetri achieves 100% accuracy in the Needle-In-a-Haystack(NIH) test over 1M tokens using a 0.5B parameter model, surpassing other method or larger models and setting a new state-of-the-art(SOTA). Moreover, our method achieves significant performance improvements on real-world benchmarks, with a maximum 288% improvement. In addition, InfiniRetri can be applied to any Transformer-based LLMs without additional training and substantially reduces inference latency and compute overhead in long texts. In summary, our comprehensive studies show InfiniRetri's potential for practical applications and creates a paradigm for retrievaling information using LLMs own capabilities under infinite-length tokens. Code will be released in link.", "authors": ["Xiaoju Ye", "Zhichun Wang", "Jingyuan Wang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-18", "url": "https://arxiv.org/abs/2502.12962", "pdf_url": "https://arxiv.org/pdf/2502.12962v1", "arxiv_id": "2502.12962", "doi": "10.48550/arXiv.2502.12962", "citation_count": 7, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "86b018a4775b526fb8f97627af089b5c26247c475fbde3ce9fb5db57a7715e55", "sources": ["arxiv", "semantic_scholar"], "title": "MoBA: Mixture of Block Attention for Long-Context LLMs", "abstract": "Scaling the effective context length is essential for advancing large language models (LLMs) toward artificial general intelligence (AGI). However, the quadratic increase in computational complexity inherent in traditional attention mechanisms presents a prohibitive overhead. Existing approaches either impose strongly biased structures, such as sink or window attention which are task-specific, or radically modify the attention mechanism into linear approximations, whose performance in complex reasoning tasks remains inadequately explored. In this work, we propose a solution that adheres to the ``less structure'' principle, allowing the model to determine where to attend autonomously, rather than introducing predefined biases. We introduce Mixture of Block Attention (MoBA), an innovative approach that applies the principles of Mixture of Experts (MoE) to the attention mechanism. This novel architecture demonstrates superior performance on long-context tasks while offering a key advantage: the ability to seamlessly transition between full and sparse attention, enhancing efficiency without the risk of compromising performance. MoBA has already been deployed to support Kimi's long-context requests and demonstrates significant advancements in efficient attention computation for LLMs. Our code is available at https://github.com/MoonshotAI/MoBA.", "authors": ["Enzhe Lu", "Zhejun Jiang", "Jingyuan Liu", "Yulun Du", "Tao Jiang", "Chao Hong", "Shaowei Liu", "Weiran He", "Enming Yuan", "Yuzhi Wang", "Zhiqi Huang", "Huan Yuan", "Suting Xu", "Xinran Xu", "Guokun Lai", "Yanru Chen", "Huabin Zheng", "Junjie Yan", "Jianlin Su", "Yuxin Wu", "Neo Y. Zhang", "Zhilin Yang", "Xinyu Zhou", "Mingxing Zhang", "Jiezhong Qiu"], "categories": ["cs.LG", "cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-18", "url": "https://arxiv.org/abs/2502.13189", "pdf_url": "https://arxiv.org/pdf/2502.13189v1", "arxiv_id": "2502.13189", "doi": "10.48550/arXiv.2502.13189", "citation_count": 162, "influential_citation_count": 19, "has_code": true, "code_url": "https://github.com/MoonshotAI/MoBA", "venue": "arXiv.org", "quality_score": 0.6505}
{"id": "24657a4edc7991729a6f26c21009721da24eaa2bffdb957a03dc0883a4c8e27b", "sources": ["arxiv", "semantic_scholar"], "title": "Does RAG Really Perform Bad For Long-Context Processing?", "abstract": "The efficient processing of long context poses a serious challenge for large language models (LLMs). Recently, retrieval-augmented generation (RAG) has emerged as a promising strategy for this problem, as it enables LLMs to make selective use of the long context for efficient computation. However, existing RAG approaches lag behind other long-context processing methods due to inherent limitations on inaccurate retrieval and fragmented contexts. To address these challenges, we introduce RetroLM, a novel RAG framework for long-context processing. Unlike traditional methods, RetroLM employs KV-level retrieval augmentation, where it partitions the LLM's KV cache into contiguous pages and retrieves the most crucial ones for efficient computation. This approach enhances robustness to retrieval inaccuracy, facilitates effective utilization of fragmented contexts, and saves the cost from repeated computation. Building on this framework, we further develop a specialized retriever for precise retrieval of critical pages and conduct unsupervised post-training to optimize the model's ability to leverage retrieved information. We conduct comprehensive evaluations with a variety of benchmarks, including LongBench, InfiniteBench, and RULER, where RetroLM significantly outperforms existing long-context LLMs and efficient long-context processing methods, particularly in tasks requiring intensive reasoning or extremely long-context comprehension.", "authors": ["Kun Luo", "Zheng Liu", "Peitian Zhang", "Hongjin Qian", "Jun Zhao", "Kang Liu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-17", "url": "https://arxiv.org/abs/2502.11444", "pdf_url": "https://arxiv.org/pdf/2502.11444v1", "arxiv_id": "2502.11444", "doi": "10.48550/arXiv.2502.11444", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2113}
{"id": "93837aacd35a3eedc41d461f264d2ef56e2fa6881c76ed4cfbeb4e9e912cb371", "sources": ["arxiv", "semantic_scholar"], "title": "LaRA: Benchmarking Retrieval-Augmented Generation and Long-Context LLMs -- No Silver Bullet for LC or RAG Routing", "abstract": "Effectively incorporating external knowledge into Large Language Models (LLMs) is crucial for enhancing their capabilities and addressing real-world needs. Retrieval-Augmented Generation (RAG) offers an effective method for achieving this by retrieving the most relevant fragments into LLMs. However, the advancements in context window size for LLMs offer an alternative approach, raising the question of whether RAG remains necessary for effectively handling external knowledge. Several existing studies provide inconclusive comparisons between RAG and long-context (LC) LLMs, largely due to limitations in the benchmark designs. In this paper, we present LaRA, a novel benchmark specifically designed to rigorously compare RAG and LC LLMs. LaRA encompasses 2326 test cases across four practical QA task categories and three types of naturally occurring long texts. Through systematic evaluation of seven open-source and four proprietary LLMs, we find that the optimal choice between RAG and LC depends on a complex interplay of factors, including the model's parameter size, long-text capabilities, context length, task type, and the characteristics of the retrieved chunks. Our findings provide actionable guidelines for practitioners to effectively leverage both RAG and LC approaches in developing and deploying LLM applications. Our code and dataset is provided at: \\href{https://github.com/Alibaba-NLP/LaRA}{\\textbf{https://github.com/Alibaba-NLP/LaRA}}.", "authors": ["Kuan Li", "Liwen Zhang", "Yong Jiang", "Pengjun Xie", "Fei Huang", "Shuai Wang", "Minhao Cheng"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-14", "url": "https://arxiv.org/abs/2502.09977", "pdf_url": "https://arxiv.org/pdf/2502.09977v2", "arxiv_id": "2502.09977", "doi": "10.48550/arXiv.2502.09977", "citation_count": 26, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/Alibaba-NLP/LaRA}{\\textbf{https://github.com/Alibaba-NLP/LaRA}}", "venue": "International Conference on Machine Learning", "quality_score": 0.3578}
{"id": "9d38ca47969c6bdadf2a9e08fa28b75f70fdc2885b5ccd0feb13a96a8c6e0928", "sources": ["arxiv", "semantic_scholar"], "title": "Diversity Enhances an LLM's Performance in RAG and Long-context Task", "abstract": "The rapid advancements in large language models (LLMs) have highlighted the challenge of context window limitations, primarily due to the quadratic time complexity of the self-attention mechanism (\\(O(N^2)\\), where \\(N\\) denotes the context window length). This constraint impacts tasks such as retrieval-augmented generation (RAG) in question answering (Q\\&A) and long context summarization. A common approach involves selecting content with the highest similarity to the query; however, this often leads to redundancy and the exclusion of diverse yet relevant information. Building on principles from Maximal Marginal Relevance (MMR) and Farthest Point Sampling (FPS), we integrate diversity into the content selection process. Our findings reveal that incorporating diversity substantially increases the recall of selecting relevant sentences or chunks before LLM-based Q\\&A and summarization. These results highlight the importance of maintaining diversity in future LLM applications to further improve summarization and Q\\&A outcomes.", "authors": ["Zhichao Wang", "Bin Bi", "Yanqi Luo", "Sitaram Asur", "Claire Na Cheng"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-13", "url": "https://arxiv.org/abs/2502.09017", "pdf_url": "https://arxiv.org/pdf/2502.09017v2", "arxiv_id": "2502.09017", "doi": "10.48550/arXiv.2502.09017", "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2603}
{"id": "4b605057f0e3d91b6c98ad97b7a097d3128666c8b031ff7567befdaac8c2bfef", "sources": ["arxiv", "semantic_scholar"], "title": "LCIRC: A Recurrent Compression Approach for Efficient Long-form Context and Query Dependent Modeling in LLMs", "abstract": "While large language models (LLMs) excel in generating coherent and contextually rich outputs, their capacity to efficiently handle long-form contexts is limited by fixed-length position embeddings. Additionally, the computational cost of processing long sequences increases quadratically, making it challenging to extend context length. To address these challenges, we propose Long-form Context Injection with Recurrent Compression (LCIRC), a method that enables the efficient processing long-form sequences beyond the model's length limit through recurrent compression without retraining the entire model. We further introduce query dependent context modeling, which selectively compresses query-relevant information, ensuring that the model retains the most pertinent content. Our empirical results demonstrate that Query Dependent LCIRC (QD-LCIRC) significantly improves LLM's ability to manage extended contexts, making it well-suited for tasks that require both comprehensive context understanding and query relevance.", "authors": ["Sumin An", "Junyoung Sung", "Wonpyo Park", "Chanjun Park", "Paul Hongsuck Seo"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-10", "url": "https://arxiv.org/abs/2502.06139", "pdf_url": "https://arxiv.org/pdf/2502.06139v2", "arxiv_id": "2502.06139", "doi": "10.48550/arXiv.2502.06139", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "North American Chapter of the Association for Computational Linguistics", "quality_score": 0.0753}
{"id": "35b73696f04f0258ef0d2877df3c24a8709a63b39d4eef5539c94e4de87249f9", "sources": ["arxiv", "semantic_scholar"], "title": "Exploiting Sparsity for Long Context Inference: Million Token Contexts on Commodity GPUs", "abstract": "There is growing demand for performing inference with hundreds of thousands of input tokens on trained transformer models. Inference at this extreme scale demands significant computational resources, hindering the application of transformers at long contexts on commodity (i.e not data center scale) hardware. To address the inference time costs associated with running self-attention based transformer language models on long contexts and enable their adoption on widely available hardware, we propose a tunable mechanism that reduces the cost of the forward pass by attending to only the most relevant tokens at every generation step using a top-k selection mechanism. We showcase the efficiency gains afforded by our method by performing inference on context windows up to 1M tokens using approximately 16GB of GPU RAM. Our experiments reveal that models are capable of handling the sparsity induced by the reduced number of keys and values. By attending to less than 2% of input tokens, we achieve over 95% of model performance on common benchmarks (RULER, AlpacaEval, and Open LLM Leaderboard).", "authors": ["Ryan Synk", "Monte Hoover", "John Kirchenbauer", "Neel Jain", "Alex Stein", "Manli Shu", "Josue Melendez Sanchez", "Ramani Duraiswami", "Tom Goldstein"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-10", "url": "https://arxiv.org/abs/2502.06766", "pdf_url": "https://arxiv.org/pdf/2502.06766v2", "arxiv_id": "2502.06766", "doi": "10.48550/arXiv.2502.06766", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1747}
{"id": "59f0611cdb97d75c72adbaedbc01952b8d965b1d1a387a47cb3b0ddfc8a79b73", "sources": ["arxiv", "semantic_scholar"], "title": "APE: Faster and Longer Context-Augmented Generation via Adaptive Parallel Encoding", "abstract": "Context-augmented generation (CAG) techniques, including RAG and ICL, require the efficient combination of multiple contexts to generate responses to user queries. Directly inputting these contexts as a sequence introduces a considerable computational burden by re-encoding the combined selection of contexts for every request. To address this, we explore the promising potential of parallel encoding to independently pre-compute and cache each context's KV states. This approach enables the direct loading of cached states during inference while accommodating more contexts through position reuse across contexts. However, due to misalignments in attention distribution, directly applying parallel encoding results in a significant performance drop. To enable effective and efficient CAG, we propose Adaptive Parallel Encoding ($\\textbf{APE}$), which brings shared prefix, attention temperature, and scaling factor to align the distribution of parallel encoding with sequential encoding. Results on RAG and ICL tasks demonstrate that APE can preserve 98% and 93% sequential encoding performance using the same inputs while outperforming parallel encoding by 3.6% and 7.9%, respectively. It also scales to many-shot CAG, effectively encoding hundreds of contexts in parallel. Efficiency evaluation shows that APE can achieve an end-to-end 4.5$\\times$ speedup by reducing 28$\\times$ prefilling time for a 128K-length context.", "authors": ["Xinyu Yang", "Tianqi Chen", "Beidi Chen"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-08", "url": "https://arxiv.org/abs/2502.05431", "pdf_url": "https://arxiv.org/pdf/2502.05431v2", "arxiv_id": "2502.05431", "doi": "10.48550/arXiv.2502.05431", "citation_count": 24, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.4225}
{"id": "c12fe1c7b0e6469789af4751fb04104d6148562af3184c4de0e8fd644ef214ea", "sources": ["arxiv", "semantic_scholar"], "title": "Long-VITA: Scaling Large Multi-modal Models to 1 Million Tokens with Leading Short-Context Accuracy", "abstract": "We introduce Long-VITA, a simple yet effective large multi-modal model for long-context visual-language understanding tasks. It is adept at concurrently processing and analyzing modalities of image, video, and text over 4K frames or 1M tokens while delivering advanced performances on short-context multi-modal tasks. We propose an effective multi-modal training schema that starts with large language models and proceeds through vision-language alignment, general knowledge learning, and two sequential stages of long-sequence fine-tuning. We further implement context-parallelism distributed inference and logits-masked language modeling head to scale Long-VITA to infinitely long inputs of images and texts during model inference. Regarding training data, Long-VITA is built on a mix of 17M samples from public datasets only and demonstrates state-of-the-art performance on various multi-modal benchmarks, compared against recent cutting-edge models with internal data. Long-VITA is fully open-source and reproducible.. By leveraging our inference designs, Long-VITA models achieve a remarkable 2x prefill speedup and 4x context length extension in a single node with 8 GPUs. We hope Long-VITA can serve as a competitive baseline and offer valuable insights for the open-source community in advancing long-context multi-modal understanding.", "authors": ["Yunhang Shen", "Chaoyou Fu", "Shaoqi Dong", "Xiong Wang", "Yi-Fan Zhang", "Peixian Chen", "Mengdan Zhang", "Haoyu Cao", "Ke Li", "Shaohui Lin", "Xiawu Zheng", "Yan Zhang", "Yiyi Zhou", "Ran He", "Caifeng Shan", "Rongrong Ji", "Xing Sun"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-07", "url": "https://arxiv.org/abs/2502.05177", "pdf_url": "https://arxiv.org/pdf/2502.05177v3", "arxiv_id": "2502.05177", "doi": "10.48550/arXiv.2502.05177", "citation_count": 37, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/VITA-MLLM/Long-VITA", "venue": "arXiv.org", "quality_score": 0.3949}
{"id": "47cf312f85c40b83e5dc32d773f8b11ba8c38440c969022d57114f922ed54dd9", "sources": ["arxiv", "semantic_scholar"], "title": "Wavelet-based Positional Representation for Long Context", "abstract": "In the realm of large-scale language models, a significant challenge arises when extrapolating sequences beyond the maximum allowable length. This is because the model's position embedding mechanisms are limited to positions encountered during training, thus preventing effective representation of positions in longer sequences. We analyzed conventional position encoding methods for long contexts and found the following characteristics. (1) When the representation dimension is regarded as the time axis, Rotary Position Embedding (RoPE) can be interpreted as a restricted wavelet transform using Haar-like wavelets. However, because it uses only a fixed scale parameter, it does not fully exploit the advantages of wavelet transforms, which capture the fine movements of non-stationary signals using multiple scales (window sizes). This limitation could explain why RoPE performs poorly in extrapolation. (2) Previous research as well as our own analysis indicates that Attention with Linear Biases (ALiBi) functions similarly to windowed attention, using windows of varying sizes. However, it has limitations in capturing deep dependencies because it restricts the receptive field of the model. From these insights, we propose a new position representation method that captures multiple scales (i.e., window sizes) by leveraging wavelet transforms without limiting the model's attention field. Experimental results show that this new method improves the performance of the model in both short and long contexts. In particular, our method allows extrapolation of position information without limiting the model's attention field.", "authors": ["Yui Oka", "Taku Hasegawa", "Kyosuke Nishida", "Kuniko Saito"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-04", "url": "https://arxiv.org/abs/2502.02004", "pdf_url": "https://arxiv.org/pdf/2502.02004v1", "arxiv_id": "2502.02004", "doi": "10.48550/arXiv.2502.02004", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.1945}
{"id": "f5443cec2d919d9b26efb963b71a9c538ffaeb63ac18f6087cb17180794e29b8", "sources": ["arxiv", "semantic_scholar"], "title": "VideoRAG: Retrieval-Augmented Generation with Extreme Long-Context Videos", "abstract": "Retrieval-Augmented Generation (RAG) has demonstrated remarkable success in enhancing Large Language Models (LLMs) through external knowledge integration, yet its application has primarily focused on textual content, leaving the rich domain of multi-modal video knowledge predominantly unexplored. This paper introduces VideoRAG, the first retrieval-augmented generation framework specifically designed for processing and understanding extremely long-context videos. Our core innovation lies in its dual-channel architecture that seamlessly integrates (i) graph-based textual knowledge grounding for capturing cross-video semantic relationships, and (ii) multi-modal context encoding for efficiently preserving visual features. This novel design empowers VideoRAG to process unlimited-length videos by constructing precise knowledge graphs that span multiple videos while maintaining semantic dependencies through specialized multi-modal retrieval paradigms. Through comprehensive empirical evaluation on our proposed LongerVideos benchmark-comprising over 160 videos totaling 134+ hours across lecture, documentary, and entertainment categories-VideoRAG demonstrates substantial performance compared to existing RAG alternatives and long video understanding methods. The source code of VideoRAG implementation and the benchmark dataset are openly available at: https://github.com/HKUDS/VideoRAG.", "authors": ["Xubin Ren", "Lingrui Xu", "Long Xia", "Shuaiqiang Wang", "Dawei Yin", "Chao Huang"], "categories": ["cs.IR", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2025-02-03", "url": "https://arxiv.org/abs/2502.01549", "pdf_url": "https://arxiv.org/pdf/2502.01549v1", "arxiv_id": "2502.01549", "doi": "10.1145/3770854.3783944", "citation_count": 61, "influential_citation_count": 9, "has_code": true, "code_url": "https://github.com/HKUDS/VideoRAG", "venue": "Knowledge Discovery and Data Mining", "quality_score": 0.5}
{"id": "5708a250c6fd847f284b49d18e69b85107261dd6c77dbf9bc7ef806c141d4456", "sources": ["arxiv", "semantic_scholar"], "title": "SEAL: Scaling to Emphasize Attention for Long-Context Retrieval", "abstract": "While many advanced LLMs are designed to handle long sequence data, we can still observe notable quality degradation even within the sequence limit. In this work, we introduce a novel approach called Scaling to Emphasize Attention for Long-context retrieval (SEAL), which enhances the retrieval performance of large language models (LLMs) over long contexts. We observe that specific attention heads are closely tied to long-context retrieval, showing positive or negative correlation with retrieval scores, and adjusting the strength of these heads boosts the quality of LLMs in long context by a large margin. Built on this insight, we propose a learning-based mechanism that leverages generated data to emphasize these heads. By applying SEAL, we achieve significant improvements in long-context retrieval performance across various tasks and models. Additionally, when combined with existing training-free context extension techniques, SEAL extends the contextual limits of LLMs while maintaining highly reliable outputs.", "authors": ["Changhun Lee", "Minsang Seok", "Jun-gyu Jin", "Younghyun Cho", "Eunhyeok Park"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2025-01-25", "url": "https://arxiv.org/abs/2501.15225", "pdf_url": "https://arxiv.org/pdf/2501.15225v2", "arxiv_id": "2501.15225", "doi": "10.48550/arXiv.2501.15225", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.1505}
{"id": "ff1f5a020878676c06fff3d4eb99d402a35ecbe02a229d67d1e860f199dbf3f5", "sources": ["arxiv", "semantic_scholar"], "title": "LongReason: A Synthetic Long-Context Reasoning Benchmark via Context Expansion", "abstract": "Large language models (LLMs) have demonstrated remarkable progress in understanding long-context inputs. However, benchmarks for evaluating the long-context reasoning abilities of LLMs fall behind the pace. Existing benchmarks often focus on a narrow range of tasks or those that do not demand complex reasoning. To address this gap and enable a more comprehensive evaluation of the long-context reasoning capabilities of current LLMs, we propose a new synthetic benchmark, LongReason, which is constructed by synthesizing long-context reasoning questions from a varied set of short-context reasoning questions through context expansion. LongReason consists of 794 multiple-choice reasoning questions with diverse reasoning patterns across three task categories: reading comprehension, logical inference, and mathematical word problems. We evaluate 21 LLMs on LongReason, revealing that most models experience significant performance drops as context length increases. Our further analysis shows that even state-of-the-art LLMs still have significant room for improvement in providing robust reasoning across different tasks. We have open-sourced LongReason under https://huggingface.co/datasets/lz1bytedance/LongReason to support the comprehensive evaluation of LLMs' long-context reasoning capabilities.", "authors": ["Zhan Ling", "Kang Liu", "Kai Yan", "Yifan Yang", "Weijian Lin", "Ting-Han Fan", "Lingfeng Shen", "Zhengyin Du", "Jiecao Chen"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-01-25", "url": "https://arxiv.org/abs/2501.15089", "pdf_url": "https://arxiv.org/pdf/2501.15089v3", "arxiv_id": "2501.15089", "doi": "10.48550/arXiv.2501.15089", "citation_count": 37, "influential_citation_count": 5, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3949}
{"id": "c22061c285ca52d789ee303ac813abc2dd58107cc28cef0077f398fd8636180f", "sources": ["arxiv", "semantic_scholar"], "title": "Serving Long-Context LLMs at the Mobile Edge: Test-Time Reinforcement Learning-based Model Caching and Inference Offloading", "abstract": "Large Language Models (LLMs) can perform zero-shot learning on unseen tasks and few-shot learning on complex reasoning tasks. However, resource-limited mobile edge networks struggle to support long-context LLM serving for LLM agents during multi-round interactions with users. Unlike stateless computation offloading and static service offloading in edge computing, optimizing LLM serving at edge servers is challenging because LLMs continuously learn from context which raises accuracy, latency, and resource consumption dynamics. In this paper, we propose a joint model caching and inference offloading framework that utilizes test-time deep reinforcement learning (T2DRL) to optimize deployment and execution strategies for long-context LLM serving. In this framework, we analyze the performance convergence and design an optimization problem considering the utilization of context windows in LLMs. Furthermore, the T2DRL algorithm can learn in both the training phase and the testing phase to proactively manage cached models and service requests and adapt to context changes and usage patterns during execution. To further enhance resource allocation efficiency, we propose a double Dutch auction (DDA) mechanism, which dynamically matches supply and demand while maximizing social welfare. Finally, experimental results demonstrate that the T2DRL algorithm can reduce system costs by at least 30% compared to baselines while guaranteeing the performance of LLM agents in real-world perception and reasoning tasks.", "authors": ["Minrui Xu", "Dusit Niyato", "Christopher G. Brinton"], "categories": ["cs.NI"], "fields_of_study": ["Computer Science"], "published_date": "2025-01-24", "url": "https://arxiv.org/abs/2501.14205", "pdf_url": "https://arxiv.org/pdf/2501.14205v1", "arxiv_id": "2501.14205", "doi": "10.1109/GLOBECOM59602.2025.11432543", "citation_count": 15, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Global Communications Conference", "quality_score": 0.301}
{"id": "5fa9cba618738ae56d921e1fe9b32be4628071883de3e224a1bd53b8515ecae7", "sources": ["arxiv", "semantic_scholar"], "title": "NExtLong: Toward Effective Long-Context Training without Long Documents", "abstract": "Large language models (LLMs) with extended context windows have made significant strides yet remain a challenge due to the scarcity of long documents. Existing methods tend to synthesize long-context data but lack a clear mechanism to reinforce the long-range dependency modeling. To address this limitation, we propose NExtLong, a novel framework for synthesizing long-context data through Negative document Extension. NExtLong decomposes a document into multiple meta-chunks and extends the context by interleaving hard negative distractors retrieved from pretraining corpora. This approach compels the model to discriminate long-range dependent context from distracting content, enhancing its ability to model long-range dependencies. Extensive experiments demonstrate that NExtLong achieves significant performance improvements on the HELMET and RULER benchmarks compared to existing long-context synthesis approaches and leading models, which are trained on non-synthetic long documents. These findings highlight NExtLong's ability to reduce reliance on non-synthetic long documents, making it an effective framework for developing advanced long-context LLMs.", "authors": ["Chaochen Gao", "Xing Wu", "Zijia Lin", "Debing Zhang", "Songlin Hu"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-01-22", "url": "https://arxiv.org/abs/2501.12766", "pdf_url": "https://arxiv.org/pdf/2501.12766v2", "arxiv_id": "2501.12766", "doi": "10.48550/arXiv.2501.12766", "citation_count": 18, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.3197}
{"id": "8efc99e1aa146b630e5046457f1d0fb4bf5645b0fb2fb754398129145cdd5445", "sources": ["arxiv", "semantic_scholar"], "title": "Is Long Context All You Need? Leveraging LLM's Extended Context for NL2SQL", "abstract": "Large Language Models (LLMs) have demonstrated impressive capabilities across a range of natural language processing tasks. In particular, improvements in reasoning abilities and the expansion of context windows have opened new avenues for leveraging these powerful models. NL2SQL is challenging in that the natural language question is inherently ambiguous, while the SQL generation requires a precise understanding of complex data schema and semantics. One approach to this semantic ambiguous problem is to provide more and sufficient contextual information. In this work, we explore the performance and the latency trade-offs of the extended context window (a.k.a., long context) offered by Google's state-of-the-art LLM (\\textit{gemini-1.5-pro}). We study the impact of various contextual information, including column example values, question and SQL query pairs, user-provided hints, SQL documentation, and schema. To the best of our knowledge, this is the first work to study how the extended context window and extra contextual information can help NL2SQL generation with respect to both accuracy and latency cost. We show that long context LLMs are robust and do not get lost in the extended contextual information. Additionally, our long-context NL2SQL pipeline based on Google's \\textit{gemini-pro-1.5} achieve strong performances on various benchmark datasets without finetuning and expensive self-consistency based techniques.", "authors": ["Yeounoh Chung", "Gaurav T. Kakkar", "Yu Gan", "Brenton Milne", "Fatma Ozcan"], "categories": ["cs.DB", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-01-21", "url": "https://arxiv.org/abs/2501.12372", "pdf_url": "https://arxiv.org/pdf/2501.12372v6", "arxiv_id": "2501.12372", "doi": "10.14778/3742728.3742761", "citation_count": 32, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "Proceedings of the VLDB Endowment", "quality_score": 0.3796}
{"id": "ec30204430e669b134b71225c9506fda5944b0cd07c49578b2ed32302c714869", "sources": ["arxiv", "semantic_scholar"], "title": "Gravitational physics in the context of Indian astronomy: A vision document", "abstract": "Contributions from the Indian gravity community have played a significant role in shaping several branches of astronomy and astrophysics. This document reviews some of the most important contributions and presents a vision for gravity research in the context of astronomy \\& astrophysics in India. This is an expanded version of one of the chapters in the recently released Vision Document of the Astronomical Society of India.", "authors": ["P. Ajith", "K. G. Arun", "Sukanta Bose", "Sumanta Chakraborty", "Shantanu Desai", "A. Gopakumar", "Sanved Kolekar", "Rajesh Nayak", "Archana Pai", "Sudipta Sarkar", "Jasjeet Singh Bagla", "Patrick Das Gupta", "Rahul Kashyap", "Prashant Kocherlakota", "Prayush Kumar", "Banibrata Mukhopadhyay"], "categories": ["astro-ph.IM", "astro-ph.HE", "gr-qc"], "fields_of_study": ["Physics"], "published_date": "2025-01-08", "url": "https://arxiv.org/abs/2501.04333", "pdf_url": "https://arxiv.org/pdf/2501.04333v1", "arxiv_id": "2501.04333", "doi": "10.1007/s12036-024-10031-x", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Journal of astrophysics and astronomy", "quality_score": 0.0753}
{"id": "3bd27b48b42722dd43fe05d54c49c56658b9e3bff7024b3ff75f6233ffbb2ae8", "sources": ["arxiv", "semantic_scholar"], "title": "AdaSkip: Adaptive Sublayer Skipping for Accelerating Long-Context LLM Inference", "abstract": "Long-context large language models (LLMs) inference is increasingly critical, motivating a number of studies devoted to alleviating the substantial storage and computational costs in such scenarios. Layer-wise skipping methods are promising optimizations but rarely explored in long-context inference. We observe that existing layer-wise skipping strategies have several limitations when applied in long-context inference, including the inability to adapt to model and context variability, disregard for sublayer significance, and inapplicability for the prefilling phase. This paper proposes \\sysname, an adaptive sublayer skipping method specifically designed for long-context inference. \\sysname adaptively identifies less important layers by leveraging on-the-fly similarity information, enables sublayer-wise skipping, and accelerates both the prefilling and decoding phases. The effectiveness of \\sysname is demonstrated through extensive experiments on various long-context benchmarks and models, showcasing its superior inference performance over existing baselines.", "authors": ["Zhuomin He", "Yizhen Yao", "Pengfei Zuo", "Bin Gao", "Qinya Li", "Zhenzhe Zheng", "Fan Wu"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2025-01-04", "url": "https://arxiv.org/abs/2501.02336", "pdf_url": "https://arxiv.org/pdf/2501.02336v1", "arxiv_id": "2501.02336", "doi": "10.48550/arXiv.2501.02336", "citation_count": 13, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "AAAI Conference on Artificial Intelligence", "quality_score": 0.301}
{"id": "b66ad0c92550cc0159e90c8cd82749b9341ef05a3e12627b5a7866191e6c2be4", "sources": ["arxiv", "semantic_scholar"], "title": "Adjoint sharding for very long context training of state space models", "abstract": "Despite very fast progress, efficiently training large language models (LLMs) in very long contexts remains challenging. Existing methods fall back to training LLMs with short contexts (a maximum of a few thousands tokens in training) and use inference time techniques when evaluating on long contexts (above 1M tokens context window at inference). As opposed to long-context-inference, training on very long context input prompts is quickly limited by GPU memory availability and by the prohibitively long training times it requires on state-of-the-art hardware. Meanwhile, many real-life applications require not only inference but also training/fine-tuning with long context on specific tasks. Such applications include, for example, augmenting the context with various sources of raw reference information for fact extraction, fact summarization, or fact reconciliation tasks. We propose adjoint sharding, a novel technique that comprises sharding gradient calculation during training to reduce memory requirements by orders of magnitude, making training on very long context computationally tractable. Adjoint sharding is based on the adjoint method and computes equivalent gradients to backpropagation. We also propose truncated adjoint sharding to speed up the algorithm while maintaining performance. We provide a distributed version, and a paralleled version of adjoint sharding to further speed up training. Empirical results show the proposed adjoint sharding algorithm reduces memory usage by up to 3X with a 1.27B parameter large language model on 1M context length training. This allows to increase the maximum context length during training or fine-tuning of a 1.27B parameter model from 35K tokens to above 100K tokens on a training infrastructure composed of five AWS P4 instances.", "authors": ["Xingzi Xu", "Amir Tavanaei", "Kavosh Asadi", "Karim Bouyarmane"], "categories": ["cs.LG", "cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2025-01-01", "url": "https://arxiv.org/abs/2501.00692", "pdf_url": "https://arxiv.org/pdf/2501.00692v1", "arxiv_id": "2501.00692", "doi": "10.48550/arXiv.2501.00692", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0069}
{"id": "1610cce0fb473c9ade63e084ee99469759b229d7c203e6d53fdb34dba58e0840", "sources": ["arxiv", "semantic_scholar"], "title": "PIMphony: Overcoming Bandwidth and Capacity Inefficiency in PIM-based Long-Context LLM Inference System", "abstract": "The expansion of long-context Large Language Models (LLMs) creates significant memory system challenges. While Processing-in-Memory (PIM) is a promising accelerator, we identify that it suffers from critical inefficiencies when scaled to long contexts: severe channel underutilization, performance-limiting I/O bottlenecks, and massive memory waste from static KV cache management. In this work, we propose PIMphony, a PIM orchestrator that systematically resolves these issues with three co-designed techniques. First, Token-Centric PIM Partitioning (TCP) ensures high channel utilization regardless of batch size. Second, Dynamic PIM Command Scheduling (DCS) mitigates the I/O bottleneck by overlapping data movement and computation. Finally, a Dynamic PIM Access (DPA) controller enables dynamic memory management to eliminate static memory waste. Implemented via an MLIR-based compiler and evaluated on a cycle-accurate simulator, PIMphony significantly improves throughput for long-context LLM inference (up to 72B parameters and 1M context length). Our evaluations show performance boosts of up to 11.3x on PIM-only systems and 8.4x on xPU+PIM systems, enabling more efficient deployment of LLMs in real-world long-context applications.", "authors": ["Hyucksung Kwon", "Kyungmo Koo", "Janghyeon Kim", "Woongkyu Lee", "Minjae Lee", "Gyeonggeun Jung", "Hyungdeok Lee", "Yousub Jung", "Jaehan Park", "Yosub Song", "Byeongsu Yang", "Haerang Choi", "Guhyun Kim", "Jongsoon Won", "Woojae Shin", "Changhyun Kim", "Gyeongcheol Shin", "Yongkee Kwon", "Ilkon Kim", "Euicheol Lim", "John Kim", "Jungwook Choi"], "categories": ["cs.AR", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-28", "url": "https://arxiv.org/abs/2412.20166", "pdf_url": "https://arxiv.org/pdf/2412.20166v3", "arxiv_id": "2412.20166", "doi": "10.1109/HPCA68181.2026.11408592", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Symposium on High-Performance Computer Architecture", "quality_score": 0.2113}
{"id": "29fb3c8bad55bf56081bb6fddb4af2746bebd405a72550e427ccb952f872f8ed", "sources": ["arxiv", "semantic_scholar"], "title": "Long Context vs. RAG for LLMs: An Evaluation and Revisits", "abstract": "Extending context windows (i.e., Long Context, LC) and using retrievers to selectively access relevant information (i.e., Retrieval-Augmented Generation, RAG) are the two main strategies to enable LLMs to incorporate extremely long external contexts. This paper revisits recent studies on this topic, highlighting their key insights and discrepancies. We then provide a more comprehensive evaluation by filtering out questions answerable without external context, identifying the most effective retrieval methods, and expanding the datasets. We show that LC generally outperforms RAG in question-answering benchmarks, especially for Wikipedia-based questions. Summarization-based retrieval performs comparably to LC, while chunk-based retrieval lags behind. However, RAG has advantages in dialogue-based and general question queries. These insights underscore the trade-offs between RAG and LC strategies, offering guidance for future optimization of LLMs with external knowledge sources. We also provide an in-depth discussion on this topic, highlighting the overlooked importance of context relevance in existing studies.", "authors": ["Xinze Li", "Yixin Cao", "Yubo Ma", "Aixin Sun"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-27", "url": "https://arxiv.org/abs/2501.01880", "pdf_url": "https://arxiv.org/pdf/2501.01880v1", "arxiv_id": "2501.01880", "doi": "10.48550/arXiv.2501.01880", "citation_count": 29, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3693}
{"id": "8a00c0c65a10d97905cf48ff9ee58923e8db89dcd4fc7c35ede71860d1752bb8", "sources": ["arxiv", "semantic_scholar"], "title": "PRISM: Efficient Long-Range Reasoning With Short-Context LLMs", "abstract": "Long-range tasks demand reasoning over long inputs. However, existing solutions are limited, e.g., long-context models require large compute budgets, parameter-efficient fine-tuning (PEFT) needs training data, and retrieval-augmented generation (RAG) entails complex task-specific designs. Though in-context approaches overcome many of these issues, methods with short-context LLMs are inefficient, trading context for processing more tokens. We introduce PRISM, a highly token-efficient in-context method based on structured schemas that outperforms baselines on diverse tasks with 4x shorter contexts. This approach produces concise outputs and efficiently leverages key-value (KV) caches to reduce costs by up to 54%. PRISM scales down to tiny contexts without increasing costs or sacrificing quality, and generalizes to new tasks with minimal effort by generating schemas from task descriptions.", "authors": ["Dulhan Jayalath", "James Bradley Wendt", "Nicholas Monath", "Sandeep Tata", "Beliz Gunel"], "categories": ["cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-25", "url": "https://arxiv.org/abs/2412.18914", "pdf_url": "https://arxiv.org/pdf/2412.18914v3", "arxiv_id": "2412.18914", "doi": "10.18653/v1/2025.emnlp-main.517", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.1747}
{"id": "3cd7344f2ec7b5c04cd280dd32d22bcd3bdff4ac7b599c788e1af917ad9fe891", "sources": ["arxiv", "semantic_scholar"], "title": "Revisiting In-Context Learning with Long Context Language Models", "abstract": "In-Context Learning (ICL) is a technique by which language models make predictions based on examples provided in their input context. Previously, their context window size imposed a limit on the number of examples that can be shown, making example selection techniques crucial for identifying the maximally effective set of examples. However, the recent advent of Long Context Language Models (LCLMs) has significantly increased the number of examples that can be included in context, raising an important question of whether ICL performance in a many-shot regime is still sensitive to the method of sample selection. To answer this, we revisit these approaches in the context of LCLMs through extensive experiments on 18 datasets spanning 4 tasks. Surprisingly, we observe that sophisticated example selection techniques do not yield significant improvements over a simple random sample selection method. Instead, we discover that the advent of LCLMs has fundamentally shifted the challenge of ICL from that of selecting the most effective examples to that of collecting sufficient examples to fill the context window. Specifically, in certain datasets, including all available examples does not fully utilize the context window; however, by augmenting the examples in context with a simple data augmentation approach, we substantially improve ICL performance by 5%.", "authors": ["Jinheon Baek", "Sun Jae Lee", "Prakhar Gupta", "Geunseob Oh", "Siddharth Dalmia", "Prateek Kolhar"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-22", "url": "https://arxiv.org/abs/2412.16926", "pdf_url": "https://arxiv.org/pdf/2412.16926v3", "arxiv_id": "2412.16926", "doi": "10.48550/arXiv.2412.16926", "citation_count": 11, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.301}
{"id": "90968f759d76c448e46137cadadeb15d96e80de55570b864fc83e9b9ff73ff64", "sources": ["arxiv", "semantic_scholar"], "title": "Systematic Evaluation of Long-Context LLMs on Financial Concepts", "abstract": "Long-context large language models (LC LLMs) promise to increase reliability of LLMs in real-world tasks requiring processing and understanding of long input documents. However, this ability of LC LLMs to reliably utilize their growing context windows remains under investigation. In this work, we evaluate the performance of state-of-the-art GPT-4 suite of LC LLMs in solving a series of progressively challenging tasks, as a function of factors such as context length, task difficulty, and position of key information by creating a real world financial news dataset. Our findings indicate that LC LLMs exhibit brittleness at longer context lengths even for simple tasks, with performance deteriorating sharply as task complexity increases. At longer context lengths, these state-of-the-art models experience catastrophic failures in instruction following resulting in degenerate outputs. Our prompt ablations also reveal unfortunate continued sensitivity to both the placement of the task instruction in the context window as well as minor markdown formatting. Finally, we advocate for more rigorous evaluation of LC LLMs by employing holistic metrics such as F1 (rather than recall) and reporting confidence intervals, thereby ensuring robust and conclusive findings.", "authors": ["Lavanya Gupta", "Saket Sharma", "Yiyun Zhao"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-19", "url": "https://arxiv.org/abs/2412.15386", "pdf_url": "https://arxiv.org/pdf/2412.15386v1", "arxiv_id": "2412.15386", "doi": "10.18653/v1/2024.emnlp-industry.88", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.1945}
{"id": "c9c4246c364e7bce50a5404c48255d40175e102f1549247673624fbc69db962f", "sources": ["arxiv", "semantic_scholar"], "title": "MultiverSeg: Scalable Interactive Segmentation of Biomedical Imaging Datasets with In-Context Guidance", "abstract": "Medical researchers and clinicians often need to perform novel segmentation tasks on a set of related images. Existing methods for segmenting a new dataset are either interactive, requiring substantial human effort for each image, or require an existing set of previously labeled images. We introduce a system, MultiverSeg, that enables practitioners to rapidly segment an entire new dataset without requiring access to any existing labeled data from that task or domain. Along with the image to segment, the model takes user interactions such as clicks, bounding boxes or scribbles as input, and predicts a segmentation. As the user segments more images, those images and segmentations become additional inputs to the model, providing context. As the context set of labeled images grows, the number of interactions required to segment each new image decreases. We demonstrate that MultiverSeg enables users to interactively segment new datasets efficiently, by amortizing the number of interactions per image to achieve an accurate segmentation. Compared to using a state-of-the-art interactive segmentation method, MultiverSeg reduced the total number of clicks by 36% and scribble steps by 25% to achieve 90% Dice on sets of images from unseen tasks. We release code and model weights at https://multiverseg.csail.mit.edu", "authors": ["Hallee E. Wong", "Jose Javier Gonzalez Ortiz", "John Guttag", "Adrian V. Dalca"], "categories": ["cs.CV", "cs.LG", "eess.IV"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2024-12-19", "url": "https://arxiv.org/abs/2412.15058", "pdf_url": "https://arxiv.org/pdf/2412.15058v2", "arxiv_id": "2412.15058", "doi": "10.1109/ICCV51701.2025.01949", "citation_count": 9, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "IEEE International Conference on Computer Vision", "quality_score": 0.25}
{"id": "7d9e562cc6dacfb6d5c30515f3a58c53ac4289c742958c61766c108eecfc361b", "sources": ["arxiv", "semantic_scholar"], "title": "Dehallucinating Parallel Context Extension for Retrieval-Augmented Generation", "abstract": "Large language models (LLMs) are susceptible to generating hallucinated information, despite the integration of retrieval-augmented generation (RAG). Parallel context extension (PCE) is a line of research attempting to effectively integrating parallel (unordered) contexts, while it still suffers from hallucinations when adapted to RAG scenarios. In this paper, we propose DePaC (Dehallucinating Parallel Context Extension), which alleviates the hallucination problem with context-aware negative training and information-calibrated aggregation. DePaC is designed to alleviate two types of in-context hallucination: fact fabrication (i.e., LLMs present claims that are not supported by the contexts) and fact omission (i.e., LLMs fail to present claims that can be supported by the contexts). Specifically, (1) for fact fabrication, we apply the context-aware negative training that fine-tunes the LLMs with negative supervisions, thus explicitly guiding the LLMs to refuse to answer when contexts are not related to questions; (2) for fact omission, we propose the information-calibrated aggregation which prioritizes context windows with higher information increment from their contexts. The experimental results on nine RAG tasks demonstrate that DePaC significantly alleviates the two types of hallucination and consistently achieves better performances on these tasks.", "authors": ["Zexiong Ma", "Shengnan An", "Zeqi Lin", "Yanzhen Zou", "Jian-Guang Lou", "Bing Xie"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-19", "url": "https://arxiv.org/abs/2412.14905", "pdf_url": "https://arxiv.org/pdf/2412.14905v1", "arxiv_id": "2412.14905", "doi": "10.48550/arXiv.2412.14905", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0}
{"id": "57c22dc2cc298d5c601882c0fd128023d6c1ca6be69a8013488014f406daab5e", "sources": ["arxiv", "semantic_scholar"], "title": "Sliding Windows Are Not the End: Exploring Full Ranking with Long-Context Large Language Models", "abstract": "Large Language Models (LLMs) have shown exciting performance in listwise passage ranking. Due to the limited input length, existing methods often adopt the sliding window strategy. Such a strategy, though effective, is inefficient as it involves repetitive and serialized processing, which usually re-evaluates relevant passages multiple times. As a result, it incurs redundant API costs, which are proportional to the number of inference tokens. The development of long-context LLMs enables the full ranking of all passages within a single inference, avoiding redundant API costs. In this paper, we conduct a comprehensive study of long-context LLMs for ranking tasks in terms of efficiency and effectiveness. Surprisingly, our experiments reveal that full ranking with long-context LLMs can deliver superior performance in the supervised fine-tuning setting with a huge efficiency improvement. Furthermore, we identify two limitations of fine-tuning the full ranking model based on existing methods: (1) sliding window strategy fails to produce a full ranking list as a training label, and (2) the language modeling loss cannot emphasize top-ranked passage IDs in the label. To alleviate these issues, we propose a new complete listwise label construction approach and a novel importance-aware learning objective for full ranking. Experiments show the superior performance of our method over baselines. Our codes are available at \\url{https://github.com/8421BCD/fullrank}.", "authors": ["Wenhan Liu", "Xinyu Ma", "Yutao Zhu", "Ziliang Zhao", "Shuaiqiang Wang", "Dawei Yin", "Zhicheng Dou"], "categories": ["cs.IR", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-19", "url": "https://arxiv.org/abs/2412.14574", "pdf_url": "https://arxiv.org/pdf/2412.14574v1", "arxiv_id": "2412.14574", "doi": "10.48550/arXiv.2412.14574", "citation_count": 13, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/8421BCD/fullrank}", "venue": "arXiv.org", "quality_score": 0.2865}
{"id": "60cc118b879e25f523669f616a1204bb9bcfd00701905e88edb8816a4f8c2f37", "sources": ["arxiv", "semantic_scholar"], "title": "LIFT: Improving Long Context Understanding Through Long Input Fine-Tuning", "abstract": "Long context understanding remains challenging for large language models due to their limited context windows. This paper introduces Long Input Fine-Tuning (LIFT) for long context modeling, a novel framework that enhances LLM performance on long-context tasks by adapting model parameters to the context at test time. LIFT enables efficient processing of lengthy inputs without the computational burden of offline long-context adaptation, and can improve the long-context capabilities of arbitrary short-context models. The framework is further enhanced by integrating in-context learning and pre-LIFT supervised fine-tuning. The combination of in-context learning and LIFT enables short-context models like Llama 3 to handle arbitrarily long contexts and consistently improves their performance on popular long-context benchmarks like LooGLE and LongBench. We also provide a comprehensive analysis of the strengths and limitations of LIFT on long context understanding, offering valuable directions for future research.", "authors": ["Yansheng Mao", "Jiaqi Li", "Fanxu Meng", "Jing Xiong", "Zilong Zheng", "Muhan Zhang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-18", "url": "https://arxiv.org/abs/2412.13626", "pdf_url": "https://arxiv.org/pdf/2412.13626v1", "arxiv_id": "2412.13626", "doi": "10.48550/arXiv.2412.13626", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2113}
{"id": "6420318819f6953fe0db30a4c2720e5e4986c66c423392ab3748d60734e334e7", "sources": ["arxiv", "semantic_scholar"], "title": "Core Context Aware Transformers for Long Context Language Modeling", "abstract": "Transformer-based Large Language Models (LLMs) have exhibited remarkable success in extensive tasks primarily attributed to self-attention mechanism, which requires a token to consider all preceding tokens as its context to compute attention. However, when the context length L becomes very large (e.g., 128K), the amount of potentially redundant information in the context tends to increase. The redundant context not only hampers the modeling representation performance but also incurs unnecessary computational and storage overhead. In this paper, we propose a plug-and-play Core Context Aware (CCA) Attention for efficient long-context modeling, comprising two complementary modules: 1) Globality-aware pooling module groups input tokens and dynamically compresses each group into one core token based on their significance. In this way, our method automatically focuses and strengthens core context while diminishing redundancy during the learning process, leading to effective long-term dependency modeling. 2) Locality-preserving module incorporates neighboring tokens to preserve local context for detailed representation. Notably, our CCA-Attention is able to replace the self-attention module in existing LLMs with minimal fine-tuning cost. Extensive experimental results show the superiority of our method in both long-context modeling and computational efficiency over state-of-the-art methods.", "authors": ["Yaofo Chen", "Zeng You", "Shuhai Zhang", "Haokun Li", "Yirui Li", "Yaowei Wang", "Mingkui Tan"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-17", "url": "https://arxiv.org/abs/2412.12465", "pdf_url": "https://arxiv.org/pdf/2412.12465v3", "arxiv_id": "2412.12465", "doi": null, "citation_count": 12, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2785}
{"id": "2f550f7b694ae9542ecca59c3f3de8368b19da765cc7efdda13907a75997e41d", "sources": ["arxiv", "semantic_scholar"], "title": "Emergence and Effectiveness of Task Vectors in In-Context Learning: An Encoder Decoder Perspective", "abstract": "Autoregressive transformers exhibit adaptive learning through in-context learning (ICL), which begs the question of how. Prior works have shown that transformers represent the ICL tasks as vectors in their representations. In this paper, we leverage the encoding-decoding framework to study how transformers form task vectors during pretraining and how their task encoding quality predicts ICL task performance. On synthetic ICL tasks, we analyze the training dynamics of a small transformer and report the coupled emergence of task encoding and decoding. As the model learns to encode different latent tasks (e.g., \"Finding the first noun in a sentence.\") into distinct, separable representations, it concurrently builds conditional decoding algorithms and improves its ICL performance. We validate this phenomenon across pretrained models of varying scales (Gemma-2 2B/9B/27B, Llama-3.1 8B/70B) and over the course of pretraining in OLMo-7B. Further, we demonstrate that the quality of task encoding inferred from representations predicts ICL performance, and that, surprisingly, finetuning the earlier layers can improve the task encoding and performance more than finetuning the latter layers. Our empirical insights shed light into better understanding the success and failure modes of large language models via their representations.", "authors": ["Seungwook Han", "Jinyeop Song", "Jeff Gore", "Pulkit Agrawal"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-16", "url": "https://arxiv.org/abs/2412.12276", "pdf_url": "https://arxiv.org/pdf/2412.12276v3", "arxiv_id": "2412.12276", "doi": null, "citation_count": 8, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.2386}
{"id": "358cb8377e1f262446701f17da18c3f907e595be8e0a73dcf116a9c07d0f6cd9", "sources": ["arxiv", "semantic_scholar"], "title": "V2PE: Improving Multimodal Long-Context Capability of Vision-Language Models with Variable Visual Position Encoding", "abstract": "Vision-Language Models (VLMs) have shown promising capabilities in handling various multimodal tasks, yet they struggle in long-context scenarios, particularly in tasks involving videos, high-resolution images, or lengthy image-text documents. In our work, we first conduct an empirical analysis of the long-context capabilities of VLMs using our augmented long-context multimodal datasets. Our findings reveal that directly applying the positional encoding mechanism used for textual tokens to visual tokens is suboptimal, and VLM performance degrades sharply when the position encoding exceeds the model's context window. To address this, we propose Variable Visual Position Encoding (V2PE), a novel positional encoding approach that employs variable and smaller increments for visual tokens, enabling more efficient management of long multimodal sequences. Our experiments demonstrate the effectiveness of V2PE to enhances VLMs' ability to effectively understand and reason over long multimodal contexts. We further integrate V2PE with our augmented long-context multimodal datasets to fine-tune the open-source VLM, InternVL2. The fine-tuned model achieves strong performance on both standard and long-context multimodal tasks. Notably, when the sequence length of the training dataset is increased to 256K tokens, the model is capable of processing multimodal sequences up to 1M tokens, highlighting its potential for real-world long-context applications.", "authors": ["Junqi Ge", "Ziyi Chen", "Jintao Lin", "Jinguo Zhu", "Xihui Liu", "Jifeng Dai", "Xizhou Zhu"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-12", "url": "https://arxiv.org/abs/2412.09616", "pdf_url": "https://arxiv.org/pdf/2412.09616v2", "arxiv_id": "2412.09616", "doi": "10.1109/ICCV51701.2025.01958", "citation_count": 31, "influential_citation_count": 3, "has_code": true, "code_url": "https://github.com/OpenGVLab/V2PE", "venue": "IEEE International Conference on Computer Vision", "quality_score": 0.3763}
{"id": "a90db617921852f97c3425450e608dab096c2f38749abcf07d119e4cd338120e", "sources": ["arxiv", "semantic_scholar"], "title": "Breaking the Stage Barrier: A Novel Single-Stage Approach to Long Context Extension for Large Language Models", "abstract": "Recently, Large language models (LLMs) have revolutionized Natural Language Processing (NLP). Pretrained LLMs, due to limited training context size, struggle with handling long token sequences, limiting their performance on various downstream tasks. Current solutions toward long context modeling often employ multi-stage continual pertaining, which progressively increases the effective context length through several continual pretraining stages. However, those approaches require extensive manual tuning and human expertise. In this paper, we introduce a novel single-stage continual pretraining method, Head-Adaptive Rotary Position Encoding (HARPE), to equip LLMs with long context modeling capabilities while simplifying the training process. Our HARPE leverages different Rotary Position Encoding (RoPE) base frequency values across different attention heads and directly trains LLMs on the target context length. Extensive experiments on 4 language modeling benchmarks, including the latest RULER benchmark, demonstrate that HARPE excels in understanding and integrating long-context tasks with single-stage training, matching and even outperforming existing multi-stage methods. Our results highlight that HARPE successfully breaks the stage barrier for training LLMs with long context modeling capabilities.", "authors": ["Haoran Lian", "Junmin Chen", "Wei Huang", "Yizhe Xiong", "Wenping Hu", "Guiguang Ding", "Hui Chen", "Jianwei Niu", "Zijia Lin", "Fuzheng Zhang", "Di Zhang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-10", "url": "https://arxiv.org/abs/2412.07171", "pdf_url": "https://arxiv.org/pdf/2412.07171v1", "arxiv_id": "2412.07171", "doi": "10.48550/arXiv.2412.07171", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Computational Linguistics", "quality_score": 0.1505}
{"id": "5eb483800722619e2acc847b0c69e2523e9a543e002252c2891bb2bbc12d1522", "sources": ["arxiv", "semantic_scholar"], "title": "Context Clues: Evaluating Long Context Models for Clinical Prediction Tasks on EHRs", "abstract": "Foundation Models (FMs) trained on Electronic Health Records (EHRs) have achieved state-of-the-art results on numerous clinical prediction tasks. However, most existing EHR FMs have context windows of <1k tokens. This prevents them from modeling full patient EHRs which can exceed 10k's of events. Recent advancements in subquadratic long-context architectures (e.g., Mamba) offer a promising solution. However, their application to EHR data has not been well-studied. We address this gap by presenting the first systematic evaluation of the effect of context length on modeling EHR data. We find that longer context models improve predictive performance -- our Mamba-based model surpasses the prior state-of-the-art on 9/14 tasks on the EHRSHOT prediction benchmark. For clinical applications, however, model performance alone is insufficient -- robustness to the unique properties of EHR is crucial. Thus, we also evaluate models across three previously underexplored properties of EHR data: (1) the prevalence of \"copy-forwarded\" diagnoses which creates artificial repetition of tokens within EHR sequences; (2) the irregular time intervals between EHR events which can lead to a wide range of timespans within a context window; and (3) the natural increase in disease complexity over time which makes later tokens in the EHR harder to predict than earlier ones. Stratifying our EHRSHOT results, we find that higher levels of each property correlate negatively with model performance, but that longer context models are more robust to more extreme levels of these properties. Our work highlights the potential for using long-context architectures to model EHR data, and offers a case study for identifying new challenges in modeling sequential data motivated by domains outside of natural language. We release our models and code at: https://github.com/som-shahlab/long_context_clues", "authors": ["Michael Wornow", "Suhana Bedi", "Miguel Angel Fuentes Hernandez", "Ethan Steinberg", "Jason Alan Fries", "Christopher Re", "Sanmi Koyejo", "Nigam H. Shah"], "categories": ["cs.LG", "cs.AI", "cs.CE"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-09", "url": "https://arxiv.org/abs/2412.16178", "pdf_url": "https://arxiv.org/pdf/2412.16178v2", "arxiv_id": "2412.16178", "doi": "10.48550/arXiv.2412.16178", "citation_count": 38, "influential_citation_count": 6, "has_code": true, "code_url": "https://github.com/som-shahlab/long_context_clues", "venue": "arXiv.org", "quality_score": 0.4225}
{"id": "fa2acfd7ad850de04095447062e92e48fc2ab94c5371d36b1e67d7e45cc94f40", "sources": ["arxiv", "semantic_scholar"], "title": "Ltri-LLM: Streaming Long Context Inference for LLMs with Training-Free Dynamic Triangular Attention Pattern", "abstract": "The quadratic computational complexity of the attention mechanism in current Large Language Models (LLMs) renders inference with long contexts prohibitively expensive. To address this challenge, various approaches aim to retain critical portions of the context to optimally approximate Full Attention (FA) through Key-Value (KV) compression or Sparse Attention (SA), enabling the processing of virtually unlimited text lengths in a streaming manner. However, these methods struggle to achieve performance levels comparable to FA, particularly in retrieval tasks. In this paper, our analysis of attention head patterns reveals that LLMs' attention distributions show strong local correlations, naturally reflecting a chunking mechanism for input context. We propose Ltri-LLM framework, which divides KVs into spans, stores them in an offline index, and retrieves the relevant KVs into memory for various queries. Experimental results on popular long text benchmarks show that Ltri-LLM can achieve performance close to FA while maintaining efficient, streaming-based inference.", "authors": ["Hongyin Tang", "Di Xiu", "Lanrui Wang", "Xiurui Geng", "Jingang Wang", "Xunliang Cai"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-12-06", "url": "https://arxiv.org/abs/2412.04757", "pdf_url": "https://arxiv.org/pdf/2412.04757v1", "arxiv_id": "2412.04757", "doi": "10.48550/arXiv.2412.04757", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1193}
{"id": "fffbd9a61733074f73b8427954bd62a8b17e8f4681c2d1e09739599d9a7f0e77", "sources": ["arxiv", "semantic_scholar"], "title": "When Precision Meets Position: BFloat16 Breaks Down RoPE in Long-Context Training", "abstract": "Extending context window sizes allows large language models (LLMs) to process longer sequences and handle more complex tasks. Rotary Positional Embedding (RoPE) has become the de facto standard due to its relative positional encoding properties that benefit long-context training. However, we observe that using RoPE with BFloat16 format results in numerical issues, causing it to deviate from its intended relative positional encoding, especially in long-context scenarios. This issue arises from BFloat16's limited precision and accumulates as context length increases, with the first token contributing significantly to this problem. To address this, we develop AnchorAttention, a plug-and-play attention method that alleviates numerical issues caused by BFloat16, improves long-context capabilities, and speeds up training. AnchorAttention reduces unnecessary attention computations, maintains semantic coherence, and boosts computational efficiency by treating the first token as a shared anchor with a consistent position ID, making it visible to all documents within the training context. Experiments on three types of LLMs demonstrate that AnchorAttention significantly improves long-context performance and reduces training time by over 50\\% compared to standard full attention mechanisms, while preserving the original LLM's capabilities on general tasks. Our code is available at https://github.com/haonan3/AnchorContext.", "authors": ["Haonan Wang", "Qian Liu", "Chao Du", "Tongyao Zhu", "Cunxiao Du", "Kenji Kawaguchi", "Tianyu Pang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-11-20", "url": "https://arxiv.org/abs/2411.13476", "pdf_url": "https://arxiv.org/pdf/2411.13476v2", "arxiv_id": "2411.13476", "doi": "10.48550/arXiv.2411.13476", "citation_count": 16, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/haonan3/AnchorContext", "venue": "arXiv.org", "quality_score": 0.3076}
{"id": "c7c11fd3c7dad94904065dc66d7aaf0d45a9a270233fe09124cdf30dd0eb2101", "sources": ["arxiv", "semantic_scholar"], "title": "Membership Inference Attack against Long-Context Large Language Models", "abstract": "Recent advances in Large Language Models (LLMs) have enabled them to overcome their context window limitations, and demonstrate exceptional retrieval and reasoning capacities on longer context. Quesion-answering systems augmented with Long-Context Language Models (LCLMs) can automatically search massive external data and incorporate it into their contexts, enabling faithful predictions and reducing issues such as hallucinations and knowledge staleness. Existing studies targeting LCLMs mainly concentrate on addressing the so-called lost-in-the-middle problem or improving the inference effiencicy, leaving their privacy risks largely unexplored. In this paper, we aim to bridge this gap and argue that integrating all information into the long context makes it a repository of sensitive information, which often contains private data such as medical records or personal identities. We further investigate the membership privacy within LCLMs external context, with the aim of determining whether a given document or sequence is included in the LCLMs context. Our basic idea is that if a document lies in the context, it will exhibit a low generation loss or a high degree of semantic similarity to the contents generated by LCLMs. We for the first time propose six membership inference attack (MIA) strategies tailored for LCLMs and conduct extensive experiments on various popular models. Empirical results demonstrate that our attacks can accurately infer membership status in most cases, e.g., 90.66% attack F1-score on Multi-document QA datasets with LongChat-7b-v1.5-32k, highlighting significant risks of membership leakage within LCLMs input contexts. Furthermore, we examine the underlying reasons why LCLMs are susceptible to revealing such membership information.", "authors": ["Zixiong Wang", "Gaoyang Liu", "Yang Yang", "Chen Wang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-11-18", "url": "https://arxiv.org/abs/2411.11424", "pdf_url": "https://arxiv.org/pdf/2411.11424v1", "arxiv_id": "2411.11424", "doi": "10.48550/arXiv.2411.11424", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1193}
{"id": "1203fc0d997f47603a5103a8be271bee5389fb9730d6febeee87d9c7ac51f48a", "sources": ["arxiv", "semantic_scholar"], "title": "Squeezed Attention: Accelerating Long Context Length LLM Inference", "abstract": "Emerging Large Language Model (LLM) applications require long input context in order to perform complex tasks like document analysis and code generation. For these long context length applications, the length of the input prompt poses a significant challenge in terms of inference efficiency since the inference costs increase linearly with sequence length. However, for many of these applications, much of the context in the prompt is fixed across different user inputs, thereby providing the opportunity to perform offline optimizations in order to process user inputs quickly, as they are received. We propose Squeezed Attention to accelerate LLM applications where a large portion of the input context is fixed. We first leverage K-means clustering offline to group the keys for the fixed context based on semantic similarity and represent each cluster with a single centroid value. During inference, we compare query tokens from the user input with the centroids to predict which keys from the fixed context are semantically relevant, and then compute exact attention using only the important keys, thereby reducing bandwidth and computational costs. We also present a hierarchical version of our algorithm which can reduce the complexity of attention from linear to logarithmic with respect to the fixed context length. We evaluate our method on long-context benchmarks including LongBench, where it achieves a 3.1$\\times$ reduction in KV budget with no noticeable accuracy loss and up to an 8$\\times$ reduction with only a 0.5 point accuracy gap for the LLaMA-2-7B-32K, LWM-Text-Chat-1M, and Longchat-7B-v1.5-32K models. Futhermore, we implement kernels for centroid comparison and sparse FlashAttention with important keys, achieving more than 4$\\times$ speedups during both the prefill and generation phases for long-context inference. Our code is available at https://github.com/SqueezeAILab/SqueezedAttention.", "authors": ["Coleman Hooper", "Sehoon Kim", "Hiva Mohammadzadeh", "Monishwaran Maheswaran", "Sebastian Zhao", "June Paik", "Michael W. Mahoney", "Kurt Keutzer", "Amir Gholami"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-11-14", "url": "https://arxiv.org/abs/2411.09688", "pdf_url": "https://arxiv.org/pdf/2411.09688v3", "arxiv_id": "2411.09688", "doi": "10.48550/arXiv.2411.09688", "citation_count": 51, "influential_citation_count": 3, "has_code": true, "code_url": "https://github.com/SqueezeAILab/SqueezedAttention", "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.429}
{"id": "fdb4a3dcc10ea13ef547390886d45f6f61b4e5888af5c425210ae810f6104809", "sources": ["arxiv", "semantic_scholar"], "title": "LongSafety: Enhance Safety for Long-Context LLMs", "abstract": "Recent advancements in model architectures and length extrapolation techniques have significantly extended the context length of large language models (LLMs), paving the way for their application in increasingly complex tasks. However, despite the growing capabilities of long-context LLMs, the safety issues in long-context scenarios remain underexplored. While safety alignment in short context has been widely studied, the safety concerns of long-context LLMs have not been adequately addressed. In this work, we introduce \\textbf{LongSafety}, a comprehensive safety alignment dataset for long-context LLMs, containing 10 tasks and 17k samples, with an average length of 40.9k tokens. Our experiments demonstrate that training with LongSafety can enhance long-context safety performance while enhancing short-context safety and preserving general capabilities. Furthermore, we demonstrate that long-context safety does not equal long-context alignment with short-context safety data and LongSafety has generalizing capabilities in context length and long-context safety scenarios.", "authors": ["Mianqiu Huang", "Xiaoran Liu", "Shaojun Zhou", "Mozhi Zhang", "Qipeng Guo", "Linyang Li", "Chenkun Tan", "Yang Gao", "Pengyu Wang", "Linlin Li", "Qun Liu", "Yaqian Zhou", "Xipeng Qiu", "Xuanjing Huang"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-11-11", "url": "https://arxiv.org/abs/2411.06899", "pdf_url": "https://arxiv.org/pdf/2411.06899v2", "arxiv_id": "2411.06899", "doi": null, "citation_count": 8, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2386}
{"id": "1f40fa50e3bc364a9e2cad7bbdeb5d59ba0baef04789595c273b12caf3703232", "sources": ["arxiv", "semantic_scholar"], "title": "FinDVer: Explainable Claim Verification over Long and Hybrid-Content Financial Documents", "abstract": "We introduce FinDVer, a comprehensive benchmark specifically designed to evaluate the explainable claim verification capabilities of LLMs in the context of understanding and analyzing long, hybrid-content financial documents. FinDVer contains 2,400 expert-annotated examples, divided into three subsets: information extraction, numerical reasoning, and knowledge-intensive reasoning, each addressing common scenarios encountered in real-world financial contexts. We assess a broad spectrum of LLMs under long-context and RAG settings. Our results show that even the current best-performing system, GPT-4o, still lags behind human experts. We further provide in-depth analysis on long-context and RAG setting, Chain-of-Thought reasoning, and model reasoning errors, offering insights to drive future advancements. We believe that FinDVer can serve as a valuable benchmark for evaluating LLMs in claim verification over complex, expert-domain documents.", "authors": ["Yilun Zhao", "Yitao Long", "Yuru Jiang", "Chengye Wang", "Weiyuan Chen", "Hongjun Liu", "Yiming Zhang", "Xiangru Tang", "Chen Zhao", "Arman Cohan"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-11-08", "url": "https://arxiv.org/abs/2411.05764", "pdf_url": "https://arxiv.org/pdf/2411.05764v1", "arxiv_id": "2411.05764", "doi": "10.18653/v1/2024.emnlp-main.818", "citation_count": 19, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.3253}
{"id": "c4fd9dbd2a28997e4aa10a957d0bd17befd9821c986043cf33caf34af24577c0", "sources": ["arxiv", "semantic_scholar"], "title": "Reducing Distraction in Long-Context Language Models by Focused Learning", "abstract": "Recent advancements in Large Language Models (LLMs) have significantly enhanced their capacity to process long contexts. However, effectively utilizing this long context remains a challenge due to the issue of distraction, where irrelevant information dominates lengthy contexts, causing LLMs to lose focus on the most relevant segments. To address this, we propose a novel training method that enhances LLMs' ability to discern relevant information through a unique combination of retrieval-based data augmentation and contrastive learning. Specifically, during fine-tuning with long contexts, we employ a retriever to extract the most relevant segments, serving as augmented inputs. We then introduce an auxiliary contrastive learning objective to explicitly ensure that outputs from the original context and the retrieved sub-context are closely aligned. Extensive experiments on long single-document and multi-document QA benchmarks demonstrate the effectiveness of our proposed method.", "authors": ["Zijun Wu", "Bingyuan Liu", "Ran Yan", "Lei Chen", "Thomas Delteil"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-11-08", "url": "https://arxiv.org/abs/2411.05928", "pdf_url": "https://arxiv.org/pdf/2411.05928v1", "arxiv_id": "2411.05928", "doi": "10.48550/arXiv.2411.05928", "citation_count": 9, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.25}
{"id": "3606aaa78e080937bed7b6f0973054106a088fef3d4a3ea988bd510e54324138", "sources": ["arxiv", "semantic_scholar"], "title": "Abstract2Appendix: Academic Reviews Enhance LLM Long-Context Capabilities", "abstract": "Large language models (LLMs) have shown remarkable performance across various tasks, yet their ability to handle long-context reading remains challenging. This study explores the effectiveness of leveraging high-quality academic peer review data for fine-tuning LLMs to enhance their long-context capabilities. We compare the Direct Preference Optimization (DPO) method with the Supervised Fine-Tuning (SFT) method, demonstrating DPO's superiority and data efficiency. Our experiments show that the fine-tuned model achieves a 4.04-point improvement over phi-3 and a 2.6\\% increase on the Qasper benchmark using only 2000 samples. Despite facing limitations in data scale and processing costs, this study underscores the potential of DPO and high-quality data in advancing LLM performance. Additionally, the zero-shot benchmark results indicate that aggregated high-quality human reviews are overwhelmingly preferred over LLM-generated responses, even for the most capable models like GPT-4o. This suggests that high-quality human reviews are extremely rich in information, reasoning, and long-context retrieval, capabilities that even the most advanced models have not fully captured. These findings highlight the high utility of leveraging human reviews to further advance the field.", "authors": ["Shengzhi Li", "Kittipat Kampa", "Rongyu Lin", "Bohang Li", "Shichao Pei"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-11-07", "url": "https://arxiv.org/abs/2411.05232", "pdf_url": "https://arxiv.org/pdf/2411.05232v1", "arxiv_id": "2411.05232", "doi": "10.48550/arXiv.2411.05232", "citation_count": 0, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/findalexli/Abstract2Appendix", "venue": "arXiv.org", "quality_score": 0.0}
{"id": "00445c1af0074671a4ae03041ec70468f0bc1ad79e399eb6c3181ffdcb1cbbb0", "sources": ["arxiv", "semantic_scholar"], "title": "Long Context RAG Performance of Large Language Models", "abstract": "Retrieval Augmented Generation (RAG) has emerged as a crucial technique for enhancing the accuracy of Large Language Models (LLMs) by incorporating external information. With the advent of LLMs that support increasingly longer context lengths, there is a growing interest in understanding how these models perform in RAG scenarios. Can these new long context models improve RAG performance? This paper presents a comprehensive study of the impact of increased context length on RAG performance across 20 popular open source and commercial LLMs. We ran RAG workflows while varying the total context length from 2,000 to 128,000 tokens (and 2 million tokens when possible) on three domain-specific datasets, and report key insights on the benefits and limitations of long context in RAG applications. Our findings reveal that while retrieving more documents can improve performance, only a handful of the most recent state of the art LLMs can maintain consistent accuracy at long context above 64k tokens. We also identify distinct failure modes in long context scenarios, suggesting areas for future research.", "authors": ["Quinn Leng", "Jacob Portes", "Sam Havens", "Matei Zaharia", "Michael Carbin"], "categories": ["cs.LG", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-11-05", "url": "https://arxiv.org/abs/2411.03538", "pdf_url": "https://arxiv.org/pdf/2411.03538v1", "arxiv_id": "2411.03538", "doi": "10.48550/arXiv.2411.03538", "citation_count": 34, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.386}
{"id": "f70b293cc7ac1d98d72b8ce671823e05b03dda0379af3b16cea31c21dd8a79cb", "sources": ["arxiv", "semantic_scholar"], "title": "Self-Consistency Falls Short! The Adverse Effects of Positional Bias on Long-Context Problems", "abstract": "Self-consistency (SC) improves the performance of large language models (LLMs) across various tasks and domains that involve short content. However, does this support its effectiveness for long-context problems? We challenge the assumption that SC's benefits generalize to long-context settings, where LLMs often struggle with position bias, the systematic over-reliance on specific context regions-which hinders their ability to utilize information effectively from all parts of their context. Through comprehensive experimentation with varying state-of-the-art models, tasks, and SC formulations, we find that SC not only fails to improve but actively degrades performance on long-context tasks. This degradation is driven by persistent position bias, which worsens with longer context lengths and smaller model sizes but remains invariant to prompt format or task type. Unlike short-context tasks, where SC diversifies reasoning paths, long-context SC amplifies positional errors. These comprehensive results provide valuable insight into the limitations of current LLMs in long-context understanding and highlight the need for more sophisticated approaches.", "authors": ["Adam Byerly", "Daniel Khashabi"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-11-02", "url": "https://arxiv.org/abs/2411.01101", "pdf_url": "https://arxiv.org/pdf/2411.01101v3", "arxiv_id": "2411.01101", "doi": "10.1162/tacl.a.625", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Transactions of the Association for Computational Linguistics", "quality_score": 0.1747}
{"id": "80f43077421a7ae41ad9cea1cf0cfef6a58b264402f8412b935dddd5ec187111", "sources": ["arxiv", "semantic_scholar"], "title": "In-Context LoRA for Diffusion Transformers", "abstract": "Recent research arXiv:2410.15027 has explored the use of diffusion transformers (DiTs) for task-agnostic image generation by simply concatenating attention tokens across images. However, despite substantial computational resources, the fidelity of the generated images remains suboptimal. In this study, we reevaluate and streamline this framework by hypothesizing that text-to-image DiTs inherently possess in-context generation capabilities, requiring only minimal tuning to activate them. Through diverse task experiments, we qualitatively demonstrate that existing text-to-image DiTs can effectively perform in-context generation without any tuning. Building on this insight, we propose a remarkably simple pipeline to leverage the in-context abilities of DiTs: (1) concatenate images instead of tokens, (2) perform joint captioning of multiple images, and (3) apply task-specific LoRA tuning using small datasets (e.g., 20~100 samples) instead of full-parameter tuning with large datasets. We name our models In-Context LoRA (IC-LoRA). This approach requires no modifications to the original DiT models, only changes to the training data. Remarkably, our pipeline generates high-fidelity image sets that better adhere to prompts. While task-specific in terms of tuning data, our framework remains task-agnostic in architecture and pipeline, offering a powerful tool for the community and providing valuable insights for further research on product-level task-agnostic generation systems. We release our code, data, and models at https://github.com/ali-vilab/In-Context-LoRA", "authors": ["Lianghua Huang", "Wei Wang", "Zhi-Fan Wu", "Yupeng Shi", "Huanzhang Dou", "Chen Liang", "Yutong Feng", "Yu Liu", "Jingren Zhou"], "categories": ["cs.CV", "cs.GR"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-31", "url": "https://arxiv.org/abs/2410.23775", "pdf_url": "https://arxiv.org/pdf/2410.23775v3", "arxiv_id": "2410.23775", "doi": "10.48550/arXiv.2410.23775", "citation_count": 153, "influential_citation_count": 21, "has_code": true, "code_url": "https://github.com/ali-vilab/In-Context-LoRA", "venue": "arXiv.org", "quality_score": 0.6712}
{"id": "d463372870534d757c53be8a8eee35beeb3e38afd23e8dff162928806aa1aa59", "sources": ["arxiv", "semantic_scholar"], "title": "Understanding Synthetic Context Extension via Retrieval Heads", "abstract": "Long-context LLMs are increasingly in demand for applications such as retrieval-augmented generation. To defray the cost of pretraining LLMs over long contexts, recent work takes an approach of synthetic context extension: fine-tuning LLMs with synthetically generated long-context data in a post-training stage. However, it remains unclear how and why this synthetic context extension imparts abilities for downstream long-context tasks. In this paper, we investigate fine-tuning on synthetic data for three long-context tasks that require retrieval and reasoning. We vary the realism of \"needle\" concepts to be retrieved and diversity of the surrounding \"haystack\" context, from using LLMs to construct synthetic documents to using templated relations and creating symbolic datasets. We find that models trained on synthetic data fall short of the real data, but surprisingly, the mismatch can be interpreted and even predicted in terms of a special set of attention heads that are responsible for retrieval over long context, retrieval heads (Wu et al., 2024). The retrieval heads learned on synthetic data have high overlap with retrieval heads learned on real data, and there is a strong correlation between the recall of heads learned and the downstream performance of a model. Furthermore, with attention knockout and activation patching, we mechanistically show that retrieval heads are necessary and explain model performance, although they are not totally sufficient. Our results shed light on how to interpret synthetic data fine-tuning performance and how to approach creating better data for learning real-world capabilities over long contexts.", "authors": ["Xinyu Zhao", "Fangcong Yin", "Greg Durrett"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-29", "url": "https://arxiv.org/abs/2410.22316", "pdf_url": "https://arxiv.org/pdf/2410.22316v4", "arxiv_id": "2410.22316", "doi": "10.48550/arXiv.2410.22316", "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.2603}
{"id": "1d776616a97c908753f4a9eec76a3bfaa4167349403b7631855adf6e1100f654", "sources": ["arxiv", "semantic_scholar"], "title": "HoPE: A Novel Positional Encoding Without Long-Term Decay for Enhanced Context Awareness and Extrapolation", "abstract": "Many positional encodings (PEs) are designed to exhibit long-term decay, based on an entrenched and long-standing inductive opinion: tokens farther away from the current position carry less relevant information. We argue that long-term decay is outdated in the era of LLMs, as LLMs are now applied to tasks demanding precise retrieval of in-context information from arbitrary positions. Firstly, we present empirical analyses on various PEs, demonstrating that models inherently learn attention with only a local-decay pattern while forming a U-shape pattern globally, contradicting the principle of long-term decay. Furthermore, we conduct a detailed analysis of rotary position encoding (RoPE, a prevalent relative positional encoding in LLMs), and found that the U-shape attention is caused by some learned components, which are also the key factor limiting RoPE's expressiveness and extrapolation.Inspired by these insights, we propose High-frequency rotary Position Encoding (HoPE). HoPE replaces the specific components in RoPE with position-independent ones, retaining only high-frequency signals, which also breaks the principle of long-term decay in theory. HoPE achieves two major advantages: (1) Without constraints imposed by long-term decay, contradictory factors that limit spontaneous attention optimization and model extrapolation performance are removed. (2) Components representing positions and semantics are are optimized. These enhances model's context awareness and extrapolation, as validated by extensive experiments.", "authors": ["Yuhan Chen", "Ang Lv", "Jian Luan", "Bin Wang", "Wei Liu"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-28", "url": "https://arxiv.org/abs/2410.21216", "pdf_url": "https://arxiv.org/pdf/2410.21216v2", "arxiv_id": "2410.21216", "doi": "10.48550/arXiv.2410.21216", "citation_count": 17, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.3138}
{"id": "896017e7aecc9bfb2433b7031471e70cfbb7e07106b757c86fc4c9a1b95c057c", "sources": ["arxiv", "semantic_scholar"], "title": "Two are better than one: Context window extension with multi-grained self-injection", "abstract": "The limited context window of contemporary large language models (LLMs) remains a huge barrier to their broader application across various domains. While continual pre-training on long-context data is a straightforward and effective solution, it incurs substantial costs in terms of data acquisition and computational resources. To alleviate this issue, we propose SharedLLM, a novel approach grounded in the design philosophy of multi-grained context compression and query-aware information retrieval. SharedLLM is composed of two short-context LLMs such as LLaMA-2, termed upper model and lower model. The lower model functions as a compressor while the upper model acts as a decoder. The upper model receives compressed, multi-grained context information from the lower model and performs context-aware modeling on the running text. Information transfer between the compressor and decoder occurs only at the lowest layers to refrain from long forward paths in the lower model and redundant cross-attention modules in the upper model. Based on this architecture, we introduce a specialized tree-style data structure to efficiently encode, store and retrieve multi-grained contextual information for text chunks. This structure, combined with a search algorithm, enables rapid encoding and retrieval of relevant information from various levels of the tree based on the input query. This entire process, wherein the sender and receiver are derived from the same LLM layer, is referred to as self-injection.", "authors": ["Wei Han", "Pan Zhou", "Soujanya Poria", "Shuicheng Yan"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-25", "url": "https://arxiv.org/abs/2410.19318", "pdf_url": "https://arxiv.org/pdf/2410.19318v1", "arxiv_id": "2410.19318", "doi": "10.48550/arXiv.2410.19318", "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Clement25/SharedLLM", "venue": "arXiv.org", "quality_score": 0.1193}
{"id": "934a21dc76a4823432e0f910a9a4de59f1967c1c44608d43ff33b1e2ce3029b4", "sources": ["arxiv", "semantic_scholar"], "title": "LOGO -- Long cOntext aliGnment via efficient preference Optimization", "abstract": "Long-context models(LCMs) have shown great potential in processing long input sequences(even more than 100M tokens) conveniently and effectively. With significant progress, recent research has pointed out that LCMs can accurately locate token-level salient information within the context. Yet, the generation performance of these LCMs is far from satisfactory and might result in misaligned responses, such as hallucinations. To enhance the generation capability of LCMs, existing works have investigated the effects of data size and quality for both pre-training and instruction tuning. Though achieving meaningful improvement, previous methods fall short in either effectiveness or efficiency. In this paper, we introduce LOGO(Long cOntext aliGnment via efficient preference Optimization), a training strategy that first introduces preference optimization for long-context alignment. To overcome the GPU memory-bound issue caused by the long sequence, LOGO employs a reference-free preference optimization strategy and adopts a position synthesis method to construct the training data. By training with only 0.3B data on a single 8$\\times$A800 GPU machine for 16 hours, LOGO allows the Llama-3-8B-Instruct-80K model to achieve comparable performance with GPT-4 in real-world long-context tasks while preserving the model's original capabilities on other tasks, e.g., language modeling and MMLU. Moreover, LOGO can extend the model's context window size while enhancing its generation performance.", "authors": ["Zecheng Tang", "Zechen Sun", "Juntao Li", "Qiaoming Zhu", "Min Zhang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-24", "url": "https://arxiv.org/abs/2410.18533", "pdf_url": "https://arxiv.org/pdf/2410.18533v1", "arxiv_id": "2410.18533", "doi": "10.48550/arXiv.2410.18533", "citation_count": 9, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.25}
{"id": "37bead1d2efe9b87d208bfe017b685de4d12a036e64b2f3ccb1b7d0db8aecadd", "sources": ["arxiv", "semantic_scholar"], "title": "Why Does the Effective Context Length of LLMs Fall Short?", "abstract": "Advancements in distributed training and efficient attention mechanisms have significantly expanded the context window sizes of large language models (LLMs). However, recent work reveals that the effective context lengths of open-source LLMs often fall short, typically not exceeding half of their training lengths. In this work, we attribute this limitation to the left-skewed frequency distribution of relative positions formed in LLMs pretraining and post-training stages, which impedes their ability to effectively gather distant information. To address this challenge, we introduce ShifTed Rotray position embeddING (STRING). STRING shifts well-trained positions to overwrite the original ineffective positions during inference, enhancing performance within their existing training lengths. Experimental results show that without additional training, STRING dramatically improves the performance of the latest large-scale models, such as Llama3.1 70B and Qwen2 72B, by over 10 points on popular long-context benchmarks RULER and InfiniteBench, establishing new state-of-the-art results for open-source LLMs. Compared to commercial models, Llama 3.1 70B with \\method even achieves better performance than GPT-4-128K and clearly surpasses Claude 2 and Kimi-chat.", "authors": ["Chenxin An", "Jun Zhang", "Ming Zhong", "Lei Li", "Shansan Gong", "Yao Luo", "Jingjing Xu", "Lingpeng Kong"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-24", "url": "https://arxiv.org/abs/2410.18745", "pdf_url": "https://arxiv.org/pdf/2410.18745v1", "arxiv_id": "2410.18745", "doi": "10.48550/arXiv.2410.18745", "citation_count": 57, "influential_citation_count": 3, "has_code": true, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.4409}
{"id": "c7b247b3614f1ed5153ae24a36226f2f4d1ad5ec5cbacbad71246296f4777da2", "sources": ["arxiv", "semantic_scholar"], "title": "Analyzing Context Contributions in LLM-based Machine Translation", "abstract": "Large language models (LLMs) have achieved state-of-the-art performance in machine translation (MT) and demonstrated the ability to leverage in-context learning through few-shot examples. However, the mechanisms by which LLMs use different parts of the input context remain largely unexplored. In this work, we provide a comprehensive analysis of context utilization in MT, studying how LLMs use various context parts, such as few-shot examples and the source text, when generating translations. We highlight several key findings: (1) the source part of few-shot examples appears to contribute more than its corresponding targets, irrespective of translation direction; (2) finetuning LLMs with parallel data alters the contribution patterns of different context parts; and (3) there is a positional bias where earlier few-shot examples have higher contributions to the translated sequence. Finally, we demonstrate that inspecting anomalous context contributions can potentially uncover pathological translations, such as hallucinations. Our findings shed light on the internal workings of LLM-based MT which go beyond those known for standard encoder-decoder MT models.", "authors": ["Emmanouil Zaranis", "Nuno M. Guerreiro", "André F. T. Martins"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-21", "url": "https://arxiv.org/abs/2410.16246", "pdf_url": "https://arxiv.org/pdf/2410.16246v1", "arxiv_id": "2410.16246", "doi": "10.48550/arXiv.2410.16246", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.0753}
{"id": "02f73b453233c3456dbd1594282a43a02202e7236b456dda518754d9a4be2f42", "sources": ["arxiv", "semantic_scholar"], "title": "Context-Aware or Context-Insensitive? Assessing LLMs' Performance in Document-Level Translation", "abstract": "Large language models (LLMs) are increasingly strong contenders in machine translation. In this work, we focus on document-level translation, where some words cannot be translated without context from outside the sentence. Specifically, we investigate the ability of prominent LLMs to utilize the document context during translation through a perturbation analysis (analyzing models' robustness to perturbed and randomized document context) and an attribution analysis (examining the contribution of relevant context to the translation). We conduct an extensive evaluation across nine LLMs from diverse model families and training paradigms, including translation-specialized LLMs, alongside two encoder-decoder transformer baselines. We find that LLMs' improved document-translation performance compared to encoder-decoder models is not reflected in pronoun translation performance. Our analysis highlight the need for context-aware finetuning of LLMs with a focus on relevant parts of the context to improve their reliability for document-level translation.", "authors": ["Wafaa Mohammed", "Vlad Niculae"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-18", "url": "https://arxiv.org/abs/2410.14391", "pdf_url": "https://arxiv.org/pdf/2410.14391v2", "arxiv_id": "2410.14391", "doi": null, "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Machine Translation Summit", "quality_score": 0.1193}
{"id": "ba4acb9cb3b2244bad5d19a0476a6e1f0f452ae5f3e7434fbccf14fb870ccf39", "sources": ["arxiv", "semantic_scholar"], "title": "Vital Insight: Assisting Experts' Context-Driven Sensemaking of Multi-modal Personal Tracking Data Using Visualization and Human-In-The-Loop LLM", "abstract": "Passive tracking methods, such as phone and wearable sensing, have become dominant in monitoring human behaviors in modern ubiquitous computing studies. While there have been significant advances in machine-learning approaches to translate periods of raw sensor data to model momentary behaviors, (e.g., physical activity recognition), there still remains a significant gap in the translation of these sensing streams into meaningful, high-level, context-aware insights that are required for various applications (e.g., summarizing an individual's daily routine). To bridge this gap, experts often need to employ a context-driven sensemaking process in real-world studies to derive insights. This process often requires manual effort and can be challenging even for experienced researchers due to the complexity of human behaviors. We conducted three rounds of user studies with 21 experts to explore solutions to address challenges with sensemaking. We follow a human-centered design process to identify needs and design, iterate, build, and evaluate Vital Insight (VI), a novel, LLM-assisted, prototype system to enable human-in-the-loop inference (sensemaking) and visualizations of multi-modal passive sensing data from smartphones and wearables. Using the prototype as a technology probe, we observe experts' interactions with it and develop an expert sensemaking model that explains how experts move between direct data representations and AI-supported inferences to explore, question, and validate insights. Through this iterative process, we also synthesize and discuss a list of design implications for the design of future AI-augmented visualization systems to better assist experts' sensemaking processes in multi-modal health sensing data.", "authors": ["Jiachen Li", "Xiwen Li", "Justin Steinberg", "Akshat Choube", "Bingsheng Yao", "Xuhai Xu", "Dakuo Wang", "Elizabeth Mynatt", "Varun Mishra"], "categories": ["cs.HC", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-18", "url": "https://arxiv.org/abs/2410.14879", "pdf_url": "https://arxiv.org/pdf/2410.14879v4", "arxiv_id": "2410.14879", "doi": "10.1145/3749508", "citation_count": 16, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies", "quality_score": 0.3076}
{"id": "73f447cee6a3bc92eeb6e55c1e689c325289dccbab0b5b1b0c6a35e8483b7cfb", "sources": ["arxiv", "semantic_scholar"], "title": "How much do contextualized representations encode long-range context?", "abstract": "We analyze contextual representations in neural autoregressive language models, emphasizing long-range contexts that span several thousand tokens. Our methodology employs a perturbation setup and the metric \\emph{Anisotropy-Calibrated Cosine Similarity}, to capture the degree of contextualization of long-range patterns from the perspective of representation geometry. We begin the analysis with a case study on standard decoder-only Transformers, demonstrating that similar perplexity can exhibit markedly different downstream task performance, which can be explained by the difference in contextualization of long-range content. Next, we extend the analysis to other models, covering recent novel architectural designs and various training configurations. The representation-level results illustrate a reduced capacity for high-complexity (i.e., less compressible) sequences across architectures, and that fully recurrent models rely heavily on local context, whereas hybrid models more effectively encode the entire sequence structure. Finally, preliminary analysis of model size and training configurations on the encoding of long-range context suggest potential directions for improving existing language models.", "authors": ["Simeng Sun", "Cheng-Ping Hsieh"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-16", "url": "https://arxiv.org/abs/2410.12292", "pdf_url": "https://arxiv.org/pdf/2410.12292v2", "arxiv_id": "2410.12292", "doi": "10.48550/arXiv.2410.12292", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "North American Chapter of the Association for Computational Linguistics", "quality_score": 0.0}
{"id": "98f93da40c3d07af21568b9737a64fbd3245f2a0f494694fb5b88a40d8519cef", "sources": ["arxiv", "semantic_scholar"], "title": "DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads", "abstract": "Deploying long-context large language models (LLMs) is essential but poses significant computational and memory challenges. Caching all Key and Value (KV) states across all attention heads consumes substantial memory. Existing KV cache pruning methods either damage the long-context capabilities of LLMs or offer only limited efficiency improvements. In this paper, we identify that only a fraction of attention heads, a.k.a, Retrieval Heads, are critical for processing long contexts and require full attention across all tokens. In contrast, all other heads, which primarily focus on recent tokens and attention sinks--referred to as Streaming Heads--do not require full attention. Based on this insight, we introduce DuoAttention, a framework that only applies a full KV cache to retrieval heads while using a light-weight, constant-length KV cache for streaming heads, which reduces both LLM's decoding and pre-filling memory and latency without compromising its long-context abilities. DuoAttention uses a lightweight, optimization-based algorithm with synthetic data to identify retrieval heads accurately. Our method significantly reduces long-context inference memory by up to 2.55x for MHA and 1.67x for GQA models while speeding up decoding by up to 2.18x and 1.50x and accelerating pre-filling by up to 1.73x and 1.63x for MHA and GQA models, respectively, with minimal accuracy loss compared to full attention. Notably, combined with quantization, DuoAttention enables Llama-3-8B decoding with 3.3 million context length on a single A100 GPU. Code is provided in https://github.com/mit-han-lab/duo-attention.", "authors": ["Guangxuan Xiao", "Jiaming Tang", "Jingwei Zuo", "Junxian Guo", "Shang Yang", "Haotian Tang", "Yao Fu", "Song Han"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-14", "url": "https://arxiv.org/abs/2410.10819", "pdf_url": "https://arxiv.org/pdf/2410.10819v1", "arxiv_id": "2410.10819", "doi": "10.48550/arXiv.2410.10819", "citation_count": 250, "influential_citation_count": 23, "has_code": true, "code_url": "https://github.com/mit-han-lab/duo-attention", "venue": "International Conference on Learning Representations", "quality_score": 0.6901}
{"id": "fd8c4baaa3c9aeb9e77eba18f786602b5b1ef3c8aa4540fb7c05641c73f3195c", "sources": ["arxiv", "semantic_scholar"], "title": "Enhancing Long Context Performance in LLMs Through Inner Loop Query Mechanism", "abstract": "Transformers have a quadratic scaling of computational complexity with input size, which limits the input context window size of large language models (LLMs) in both training and inference. Meanwhile, retrieval-augmented generation (RAG) besed models can better handle longer contexts by using a retrieval system to filter out unnecessary information. However, most RAG methods only perform retrieval based on the initial query, which may not work well with complex questions that require deeper reasoning. We introduce a novel approach, Inner Loop Memory Augmented Tree Retrieval (ILM-TR), involving inner-loop queries, based not only on the query question itself but also on intermediate findings. At inference time, our model retrieves information from the RAG system, integrating data from lengthy documents at various levels of abstraction. Based on the information retrieved, the LLM generates texts stored in an area named Short-Term Memory (STM) which is then used to formulate the next query. This retrieval process is repeated until the text in STM converged. Our experiments demonstrate that retrieval with STM offers improvements over traditional retrieval-augmented LLMs, particularly in long context tests such as Multi-Needle In A Haystack (M-NIAH) and BABILong.", "authors": ["Yimin Tang", "Yurong Xu", "Ning Yan", "Masood Mortazavi"], "categories": ["cs.CL", "cs.AI", "cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-11", "url": "https://arxiv.org/abs/2410.12859", "pdf_url": "https://arxiv.org/pdf/2410.12859v1", "arxiv_id": "2410.12859", "doi": "10.48550/arXiv.2410.12859", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1193}
{"id": "d6200f974bff103076f9972960728197bbf44cdf6d5796bf7027abcd8b8d0570", "sources": ["arxiv", "semantic_scholar"], "title": "Metalic: Meta-Learning In-Context with Protein Language Models", "abstract": "Predicting the biophysical and functional properties of proteins is essential for in silico protein design. Machine learning has emerged as a promising technique for such prediction tasks. However, the relative scarcity of in vitro annotations means that these models often have little, or no, specific data on the desired fitness prediction task. As a result of limited data, protein language models (PLMs) are typically trained on general protein sequence modeling tasks, and then fine-tuned, or applied zero-shot, to protein fitness prediction. When no task data is available, the models make strong assumptions about the correlation between the protein sequence likelihood and fitness scores. In contrast, we propose meta-learning over a distribution of standard fitness prediction tasks, and demonstrate positive transfer to unseen fitness prediction tasks. Our method, called Metalic (Meta-Learning In-Context), uses in-context learning and fine-tuning, when data is available, to adapt to new tasks. Crucially, fine-tuning enables considerable generalization, even though it is not accounted for during meta-training. Our fine-tuned models achieve strong results with 18 times fewer parameters than state-of-the-art models. Moreover, our method sets a new state-of-the-art in low-data settings on ProteinGym, an established fitness-prediction benchmark. Due to data scarcity, we believe meta-learning will play a pivotal role in advancing protein engineering.", "authors": ["Jacob Beck", "Shikha Surana", "Manus McAuliffe", "Oliver Bent", "Thomas D. Barrett", "Juan Jose Garau Luis", "Paul Duckworth"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-10", "url": "https://arxiv.org/abs/2410.08355", "pdf_url": "https://arxiv.org/pdf/2410.08355v3", "arxiv_id": "2410.08355", "doi": "10.48550/arXiv.2410.08355", "citation_count": 4, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/instadeepai/metalic", "venue": "International Conference on Learning Representations", "quality_score": 0.1747}
{"id": "363043c74f3b43e6481eb433e748c41da38c6238070fc4f58e9b55b4f05af82e", "sources": ["arxiv", "semantic_scholar"], "title": "FltLM: An Intergrated Long-Context Large Language Model for Effective Context Filtering and Understanding", "abstract": "The development of Long-Context Large Language Models (LLMs) has markedly advanced natural language processing by facilitating the process of textual data across long documents and multiple corpora. However, Long-Context LLMs still face two critical challenges: The lost in the middle phenomenon, where crucial middle-context information is likely to be missed, and the distraction issue that the models lose focus due to overly extended contexts. To address these challenges, we propose the Context Filtering Language Model (FltLM), a novel integrated Long-Context LLM which enhances the ability of the model on multi-document question-answering (QA) tasks. Specifically, FltLM innovatively incorporates a context filter with a soft mask mechanism, identifying and dynamically excluding irrelevant content to concentrate on pertinent information for better comprehension and reasoning. Our approach not only mitigates these two challenges, but also enables the model to operate conveniently in a single forward pass. Experimental results demonstrate that FltLM significantly outperforms supervised fine-tuning and retrieval-based methods in complex QA scenarios, suggesting a promising solution for more accurate and reliable long-context natural language understanding applications.", "authors": ["Jingyang Deng", "Zhengyang Shen", "Boyang Wang", "Lixin Su", "Suqi Cheng", "Ying Nie", "Junfeng Wang", "Dawei Yin", "Jinwen Ma"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-09", "url": "https://arxiv.org/abs/2410.06886", "pdf_url": "https://arxiv.org/pdf/2410.06886v1", "arxiv_id": "2410.06886", "doi": "10.3233/FAIA240965", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "European Conference on Artificial Intelligence", "quality_score": 0.1505}
{"id": "a4f323f760e35656ab49e81905e2e3409f6290a8f327dbfb301a65751e5cf3e7", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Context Linear System Identification", "abstract": "This paper addresses the problem of long-context linear system identification, where the state $x_t$ of a dynamical system at time $t$ depends linearly on previous states $x_s$ over a fixed context window of length $p$. We establish a sample complexity bound that matches the i.i.d. parametric rate up to logarithmic factors for a broad class of systems, extending previous works that considered only first-order dependencies. Our findings reveal a learning-without-mixing phenomenon, indicating that learning long-context linear autoregressive models is not hindered by slow mixing properties potentially associated with extended context windows. Additionally, we extend these results to (i) shared low-rank representations, where rank-regularized estimators improve the dependence of the rates on the dimensionality, and (ii) misspecified context lengths in strictly stable systems, where shorter contexts offer statistical advantages.", "authors": ["Oğuz Kaan Yüksel", "Mathieu Even", "Nicolas Flammarion"], "categories": ["stat.ML", "cs.LG", "eess.SY", "math.ST"], "fields_of_study": ["Computer Science", "Mathematics", "Engineering"], "published_date": "2024-10-08", "url": "https://arxiv.org/abs/2410.05690", "pdf_url": "https://arxiv.org/pdf/2410.05690v2", "arxiv_id": "2410.05690", "doi": "10.48550/arXiv.2410.05690", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.1193}
{"id": "e4f7c6350b531af51de151693b4e0f5c140d25ae4ac78a8497a0aa13c5331aba", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Context LLMs Meet RAG: Overcoming Challenges for Long Inputs in RAG", "abstract": "Retrieval-augmented generation (RAG) empowers large language models (LLMs) to utilize external knowledge sources. The increasing capacity of LLMs to process longer input sequences opens up avenues for providing more retrieved information, to potentially enhance the quality of generated outputs. It is plausible to assume that a larger retrieval set would contain more relevant information (higher recall), that might result in improved performance. However, our empirical findings demonstrate that for many long-context LLMs, the quality of generated output initially improves first, but then subsequently declines as the number of retrieved passages increases. This paper investigates this phenomenon, identifying the detrimental impact of retrieved \"hard negatives\" as a key contributor. To mitigate this and enhance the robustness of long-context LLM-based RAG, we propose both training-free and training-based approaches. We first showcase the effectiveness of retrieval reordering as a simple yet powerful training-free optimization. Furthermore, we explore training-based methods, specifically RAG-specific implicit LLM fine-tuning and RAG-oriented fine-tuning with intermediate reasoning, demonstrating their capacity for substantial performance gains. Finally, we conduct a systematic analysis of design choices for these training-based methods, including data distribution, retriever selection, and training context length.", "authors": ["Bowen Jin", "Jinsung Yoon", "Jiawei Han", "Sercan O. Arik"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-08", "url": "https://arxiv.org/abs/2410.05983", "pdf_url": "https://arxiv.org/pdf/2410.05983v1", "arxiv_id": "2410.05983", "doi": "10.48550/arXiv.2410.05983", "citation_count": 142, "influential_citation_count": 9, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.5388}
{"id": "67c950a3fbe14600ac1ce632cab34b268c12596eb836820cabf83e0b8a50c84c", "sources": ["arxiv", "semantic_scholar"], "title": "Timer-XL: Long-Context Transformers for Unified Time Series Forecasting", "abstract": "We present Timer-XL, a causal Transformer for unified time series forecasting. To uniformly predict multidimensional time series, we generalize next token prediction, predominantly adopted for 1D token sequences, to multivariate next token prediction. The paradigm formulates various forecasting tasks as a long-context prediction problem. We opt for decoder-only Transformers that capture causal dependencies from varying-length contexts for unified forecasting, making predictions on non-stationary univariate time series, multivariate series with complicated dynamics and correlations, as well as covariate-informed contexts that include exogenous variables. Technically, we propose a universal TimeAttention to capture fine-grained intra- and inter-series dependencies of flattened time series tokens (patches), which is further enhanced by deft position embedding for temporal causality and variable equivalence. Timer-XL achieves state-of-the-art performance across task-specific forecasting benchmarks through a unified approach. Based on large-scale pre-training, Timer-XL achieves state-of-the-art zero-shot performance, making it a promising architecture for pre-trained time series models. Code is available at this repository: https://github.com/thuml/Timer-XL.", "authors": ["Yong Liu", "Guo Qin", "Xiangdong Huang", "Jianmin Wang", "Mingsheng Long"], "categories": ["cs.LG", "stat.ML"], "fields_of_study": ["Computer Science", "Mathematics"], "published_date": "2024-10-07", "url": "https://arxiv.org/abs/2410.04803", "pdf_url": "https://arxiv.org/pdf/2410.04803v4", "arxiv_id": "2410.04803", "doi": "10.48550/arXiv.2410.04803", "citation_count": 85, "influential_citation_count": 16, "has_code": true, "code_url": "https://github.com/thuml/Timer-XL", "venue": "International Conference on Learning Representations", "quality_score": 0.6152}
{"id": "6ad572c7c2c808cc3b3a3f3ced8ce58cfedf7b46e3fed23faed89c1ab38ae755", "sources": ["arxiv", "semantic_scholar"], "title": "MathHay: An Automated Benchmark for Long-Context Mathematical Reasoning in LLMs", "abstract": "Recent large language models (LLMs) have demonstrated versatile capabilities in long-context scenarios. Although some recent benchmarks have been developed to evaluate the long-context capabilities of LLMs, there is a lack of benchmarks evaluating the mathematical reasoning abilities of LLMs over long contexts, which is crucial for LLMs' application in real-world scenarios. In this paper, we introduce MathHay, an automated benchmark designed to assess the long-context mathematical reasoning capabilities of LLMs. Unlike previous benchmarks like Needle in a Haystack, which focus primarily on information retrieval within long texts, MathHay demands models with both information-seeking and complex mathematical reasoning abilities. We conduct extensive experiments on MathHay to assess the long-context mathematical reasoning abilities of eight top-performing LLMs. Even the best-performing model, Gemini-1.5-Pro-002, still struggles with mathematical reasoning over long contexts, achieving only 51.26% accuracy at 128K tokens. This highlights the significant room for improvement on the MathHay benchmark.", "authors": ["Lei Wang", "Shan Dong", "Yuhui Xu", "Hanze Dong", "Yalu Wang", "Amrita Saha", "Ee-Peng Lim", "Caiming Xiong", "Doyen Sahoo"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-07", "url": "https://arxiv.org/abs/2410.04698", "pdf_url": "https://arxiv.org/pdf/2410.04698v1", "arxiv_id": "2410.04698", "doi": "10.48550/arXiv.2410.04698", "citation_count": 13, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2865}
{"id": "54ef1fb351afdb29fc4bbe2c9b65c8fd3aff6f54cf7ce4878d0a64b8745f13f9", "sources": ["arxiv", "semantic_scholar"], "title": "LongGenBench: Long-context Generation Benchmark", "abstract": "Current long-context benchmarks primarily focus on retrieval-based tests, requiring Large Language Models (LLMs) to locate specific information within extensive input contexts, such as the needle-in-a-haystack (NIAH) benchmark. Long-context generation refers to the ability of a language model to generate coherent and contextually accurate text that spans across lengthy passages or documents. While recent studies show strong performance on NIAH and other retrieval-based long-context benchmarks, there is a significant lack of benchmarks for evaluating long-context generation capabilities. To bridge this gap and offer a comprehensive assessment, we introduce a synthetic benchmark, LongGenBench, which allows for flexible configurations of customized generation context lengths. LongGenBench advances beyond traditional benchmarks by redesigning the format of questions and necessitating that LLMs respond with a single, cohesive long-context answer. Upon extensive evaluation using LongGenBench, we observe that: (1) both API accessed and open source models exhibit performance degradation in long-context generation scenarios, ranging from 1.2% to 47.1%; (2) different series of LLMs exhibit varying trends of performance degradation, with the Gemini-1.5-Flash model showing the least degradation among API accessed models, and the Qwen2 series exhibiting the least degradation in LongGenBench among open source models.", "authors": ["Xiang Liu", "Peijie Dong", "Xuming Hu", "Xiaowen Chu"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-05", "url": "https://arxiv.org/abs/2410.04199", "pdf_url": "https://arxiv.org/pdf/2410.04199v3", "arxiv_id": "2410.04199", "doi": "10.48550/arXiv.2410.04199", "citation_count": 29, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/Dominic789654/LongGenBench", "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.3693}
{"id": "110171013492053dd92b914078d72502c98b7aaeae58f4d4160b13a0f09f7469", "sources": ["arxiv", "semantic_scholar"], "title": "MELODI: Exploring Memory Compression for Long Contexts", "abstract": "We present MELODI, a novel memory architecture designed to efficiently process long documents using short context windows. The key principle behind MELODI is to represent short-term and long-term memory as a hierarchical compression scheme across both network layers and context windows. Specifically, the short-term memory is achieved through recurrent compression of context windows across multiple layers, ensuring smooth transitions between windows. In contrast, the long-term memory performs further compression within a single middle layer and aggregates information across context windows, effectively consolidating crucial information from the entire history. Compared to a strong baseline - the Memorizing Transformer employing dense attention over a large long-term memory (64K key-value pairs) - our method demonstrates superior performance on various long-context datasets while remarkably reducing the memory footprint by a factor of 8.", "authors": ["Yinpeng Chen", "DeLesley Hutchins", "Aren Jansen", "Andrey Zhmoginov", "David Racz", "Jesper Andersen"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-04", "url": "https://arxiv.org/abs/2410.03156", "pdf_url": "https://arxiv.org/pdf/2410.03156v1", "arxiv_id": "2410.03156", "doi": "10.48550/arXiv.2410.03156", "citation_count": 5, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.1945}
{"id": "007e1b82ae125ca43874c6e9e879d1da3db0d89c9e05ccd6268adf7605a05103", "sources": ["arxiv", "semantic_scholar"], "title": "ALR$^2$: A Retrieve-then-Reason Framework for Long-context Question Answering", "abstract": "The context window of large language models (LLMs) has been extended significantly in recent years. However, while the context length that the LLM can process has grown, the capability of the model to accurately reason over that context degrades noticeably. This occurs because modern LLMs often become overwhelmed by the vast amount of information in the context; when answering questions, the model must identify and reason over relevant evidence sparsely distributed throughout the text. To alleviate the challenge of long-context reasoning, we develop a retrieve-then-reason framework, enabling LLMs to reason over relevant evidence collected during an intermediate retrieval step. We find that modern LLMs struggle to accurately retrieve relevant facts and instead, often hallucinate \"retrieved facts\", resulting in flawed reasoning and the production of incorrect answers. To address these issues, we introduce ALR$^2$, a method that augments the long-context reasoning capability of LLMs via an explicit two-stage procedure, i.e., aligning LLMs with the objectives of both retrieval and reasoning. We demonstrate the efficacy of ALR$^2$ for mitigating performance degradation in long-context reasoning tasks. Through extensive experiments on long-context QA benchmarks, we find our method to outperform competitive baselines by large margins, achieving at least 8.4 and 7.9 EM gains on the long-context versions of HotpotQA and SQuAD datasets, respectively.", "authors": ["Huayang Li", "Pat Verga", "Priyanka Sen", "Bowen Yang", "Vijay Viswanathan", "Patrick Lewis", "Taro Watanabe", "Yixuan Su"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-04", "url": "https://arxiv.org/abs/2410.03227", "pdf_url": "https://arxiv.org/pdf/2410.03227v1", "arxiv_id": "2410.03227", "doi": "10.48550/arXiv.2410.03227", "citation_count": 24, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3495}
{"id": "23f1b01aa706e11155610207348b8e27dfedd0cecbd05fd421cca5bb5ff82f1a", "sources": ["arxiv", "semantic_scholar"], "title": "Extending Context Window of Large Language Models from a Distributional Perspective", "abstract": "Scaling the rotary position embedding (RoPE) has become a common method for extending the context window of RoPE-based large language models (LLMs). However, existing scaling methods often rely on empirical approaches and lack a profound understanding of the internal distribution within RoPE, resulting in suboptimal performance in extending the context window length. In this paper, we propose to optimize the context window extending task from the view of rotary angle distribution. Specifically, we first estimate the distribution of the rotary angles within the model and analyze the extent to which length extension perturbs this distribution. Then, we present a novel extension strategy that minimizes the disturbance between rotary angle distributions to maintain consistency with the pre-training phase, enhancing the model's capability to generalize to longer sequences. Experimental results compared to the strong baseline methods demonstrate that our approach reduces by up to 72% of the distributional disturbance when extending LLaMA2's context window to 8k, and reduces by up to 32% when extending to 16k. On the LongBench-E benchmark, our method achieves an average improvement of up to 4.33% over existing state-of-the-art methods. Furthermore, Our method maintains the model's performance on the Hugging Face Open LLM benchmark after context window extension, with only an average performance fluctuation ranging from -0.12 to +0.22.", "authors": ["Yingsheng Wu", "Yuxuan Gu", "Xiaocheng Feng", "Weihong Zhong", "Dongliang Xu", "Qing Yang", "Hongtao Liu", "Bing Qin"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-02", "url": "https://arxiv.org/abs/2410.01490", "pdf_url": "https://arxiv.org/pdf/2410.01490v2", "arxiv_id": "2410.01490", "doi": "10.48550/arXiv.2410.01490", "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.2603}
{"id": "4431714785968cab33a303d412e31da0e9424290e3edf0c0c438a31df80af63e", "sources": ["arxiv", "semantic_scholar"], "title": "A Little Goes a Long Way: Efficient Long Context Training and Inference with Partial Contexts", "abstract": "Training and serving long-context large language models (LLMs) incurs substantial overhead. To address this, two critical steps are often required: a pretrained LLM typically undergoes a separate stage for context length extension by training on long-context data, followed by architectural modifications to reduce the overhead of KV cache during serving. This paper argues that integrating length extension with a GPU-friendly KV cache reduction architecture not only reduces training overhead during length extension, but also achieves better long-context performance. This leads to our proposed LongGen, which finetunes a pretrained LLM into an efficient architecture during length extension. LongGen builds on three key insights: (1) Sparse attention patterns, such as window attention (attending to recent tokens), attention sink (initial ones), and blockwise sparse attention (strided token blocks) are well-suited for building efficient long-context models, primarily due to their GPU-friendly memory access patterns, enabling efficiency gains not just theoretically but in practice as well. (2) It is essential for the model to have direct access to all tokens. A hybrid architecture with 1/3 full attention layers and 2/3 efficient ones achieves a balanced trade-off between efficiency and long-context performance. (3) Lightweight training on 5B long-context data is sufficient to extend the hybrid model's context length from 4K to 128K. We evaluate LongGen on both Llama-2 7B and Llama-2 70B, demonstrating its effectiveness across different scales. During training with 128K-long contexts, LongGen achieves 1.55x training speedup and reduces wall-clock time by 36%, compared to a full-attention baseline. During inference, LongGen reduces KV cache memory by 62%, achieving 1.67x prefilling speedup and 1.41x decoding speedup.", "authors": ["Suyu Ge", "Xihui Lin", "Yunan Zhang", "Jiawei Han", "Hao Peng"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-02", "url": "https://arxiv.org/abs/2410.01485", "pdf_url": "https://arxiv.org/pdf/2410.01485v2", "arxiv_id": "2410.01485", "doi": "10.48550/arXiv.2410.01485", "citation_count": 10, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.2603}
{"id": "6cf5712a326049c92628349df71c5ac6981d4ed4d76fd062f912a5131e4af9ab", "sources": ["arxiv", "semantic_scholar"], "title": "Locret: Enhancing Eviction in Long-Context LLM Inference with Trained Retaining Heads on Consumer-Grade Devices", "abstract": "Scaling the input context length of a large language model (LLM) incurs a significant increase in computation cost and memory footprint to maintain the attention key-value (KV) cache. Existing KV cache compression methods suffer from inefficient compression strategies and limited memory reduction effects, making it difficult for LLMs to conduct long-context inference on consumer-grade devices, especially when inferring long-context stream input. Such obstacles prevent consumer-grade devices from supporting more complex applications, creating challenges for the democratization of LLMs. To overcome this, we propose Locret, the first framework to create an eviction policy compatible with chunked prefill. By evaluating the causal importance of KV cache units by learnable retaining heads, Locret enables precise eviction of cache units, facilitating efficient long-context inference. In our extensive empirical studies, Locret outperforms the recent popular and competitive approaches in terms of memory efficiency and generation quality -- Locret achieves up to 20x of KV cache compression ratio within less than 10% performance loss. Furthermore, Locret achieves 128K+ long-context inference on a single NVIDIA 4090 GPU without compromising generation quality and only costs <1 GPU hour of additional training.", "authors": ["Yuxiang Huang", "Binhang Yuan", "Xu Han", "Chaojun Xiao", "Zhiyuan Liu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-02", "url": "https://arxiv.org/abs/2410.01805", "pdf_url": "https://arxiv.org/pdf/2410.01805v2", "arxiv_id": "2410.01805", "doi": "10.48550/arXiv.2410.01805", "citation_count": 15, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.301}
{"id": "d265a24afa0c6181aa4f3d30a0dc4e2f3929b99462f9588fb9bfdf14b7949026", "sources": ["arxiv", "semantic_scholar"], "title": "InfiniPot: Infinite Context Processing on Memory-Constrained LLMs", "abstract": "Handling long input contexts remains a significant challenge for Large Language Models (LLMs), particularly in resource-constrained environments such as mobile devices. Our work aims to address this limitation by introducing InfiniPot, a novel KV cache control framework designed to enable pre-trained LLMs to manage extensive sequences within fixed memory constraints efficiently, without requiring additional training. InfiniPot leverages Continual Context Distillation (CCD), an iterative process that compresses and retains essential information through novel importance metrics, effectively maintaining critical data even without access to future context. Our comprehensive evaluations indicate that InfiniPot significantly outperforms models trained for long contexts in various NLP tasks, establishing its efficacy and versatility. This work represents a substantial advancement toward making LLMs applicable to a broader range of real-world scenarios.", "authors": ["Minsoo Kim", "Kyuhong Shim", "Jungwook Choi", "Simyung Chang"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-10-02", "url": "https://arxiv.org/abs/2410.01518", "pdf_url": "https://arxiv.org/pdf/2410.01518v1", "arxiv_id": "2410.01518", "doi": "10.48550/arXiv.2410.01518", "citation_count": 25, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.3537}
{"id": "ccc68c3b27acd74874141a471bc38b99a98a2d29996c5363e50f8c195c387018", "sources": ["arxiv", "semantic_scholar"], "title": "Visual Context Window Extension: A New Perspective for Long Video Understanding", "abstract": "Large Multimodal Models (LMMs) have demonstrated impressive performance in short video understanding tasks but face great challenges when applied to long video understanding. In contrast, Large Language Models (LLMs) exhibit outstanding capabilities in modeling long texts. Existing work attempts to address this issue by introducing long video-text pairs during training. However, these approaches require substantial computational and data resources. In this paper, we tackle the challenge of long video understanding from the perspective of context windows, aiming to apply LMMs to long video tasks without retraining on long video datasets. We first conduct an in-depth analysis of why pretrained LMMs struggle to understand lengthy video content, identifying that discrepancies between visual and language modalities lead to different context windows for visual and language tokens, making it difficult to directly extend the visual tokens to match the language context window. Based on this, we propose to adapt LMMs for long video understanding tasks by extending the visual context window, eliminating the need for retraining on large scalelong video datasets. To further mitigate the significant memory consumption caused by long sequences, we introduce a progressive pooling inference strategy that selectively adjusts the spatial resolution of frame embeddings, reducing the number of visual tokens while retaining important spatial information. Across multiple long video understanding benchmarks, our method consistently improves the performance as the number of video frames increases. On the MLVU benchmark, our method outperforms GPT-4o, even though our model size is only 7B. Additionally, in the 256-frame setting, our method reduces memory usage by approximately 45% compared to the baseline, without introducing any performance loss.", "authors": ["Hongchen Wei", "Zhenzhong Chen"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-30", "url": "https://arxiv.org/abs/2409.20018", "pdf_url": "https://arxiv.org/pdf/2409.20018v2", "arxiv_id": "2409.20018", "doi": null, "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.0753}
{"id": "9973ba0d2c666e44789cf53fdaca352c8947cd116339b90b667b8b75f42db5a2", "sources": ["arxiv", "semantic_scholar"], "title": "Discovering the Gems in Early Layers: Accelerating Long-Context LLMs with 1000x Input Token Reduction", "abstract": "Large Language Models (LLMs) have demonstrated remarkable capabilities in handling long context inputs, but this comes at the cost of increased computational resources and latency. Our research introduces a novel approach for the long context bottleneck to accelerate LLM inference and reduce GPU memory consumption. Our research demonstrates that LLMs can identify relevant tokens in the early layers before generating answers to a query. Leveraging this insight, we propose an algorithm that uses early layers of an LLM as filters to select and compress input tokens, significantly reducing the context length for subsequent processing. Our method, GemFilter, demonstrates substantial improvements in both speed and memory efficiency compared to existing techniques, such as standard attention and SnapKV/H2O. Notably, it achieves a 2.4$\\times$ speedup and 30\\% reduction in GPU memory usage compared to SOTA methods. Evaluation on the Needle in a Haystack task shows that GemFilter significantly outperforms standard attention, SnapKV and demonstrates comparable performance on the LongBench challenge. GemFilter is simple, training-free, and broadly applicable across different LLMs. Crucially, it provides interpretability by allowing humans to inspect the selected input sequence. These findings not only offer practical benefits for LLM deployment, but also enhance our understanding of LLM internal mechanisms, paving the way for further optimizations in LLM design and inference. Our code is available at \\url{https://github.com/SalesforceAIResearch/GemFilter}.", "authors": ["Zhenmei Shi", "Yifei Ming", "Xuan-Phi Nguyen", "Yingyu Liang", "Shafiq Joty"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-25", "url": "https://arxiv.org/abs/2409.17422", "pdf_url": "https://arxiv.org/pdf/2409.17422v1", "arxiv_id": "2409.17422", "doi": "10.48550/arXiv.2409.17422", "citation_count": 51, "influential_citation_count": 7, "has_code": true, "code_url": "https://github.com/SalesforceAIResearch/GemFilter}", "venue": "arXiv.org", "quality_score": 0.4515}
{"id": "3b1a532ac913c068b9f5625ceac9ca3667a6b2d6f4d3f130dfcdd9e2b1a87973", "sources": ["arxiv", "semantic_scholar"], "title": "A Controlled Study on Long Context Extension and Generalization in LLMs", "abstract": "Broad textual understanding and in-context learning require language models that utilize full document contexts. Due to the implementation challenges associated with directly training long-context models, many methods have been proposed for extending models to handle long contexts. However, owing to differences in data and model classes, it has been challenging to compare these approaches, leading to uncertainty as to how to evaluate long-context performance and whether it differs from standard evaluation. We implement a controlled protocol for extension methods with a standardized evaluation, utilizing consistent base models and extension data. Our study yields several insights into long-context behavior. First, we reaffirm the critical role of perplexity as a general-purpose performance indicator even in longer-context tasks. Second, we find that current approximate attention methods systematically underperform across long-context tasks. Finally, we confirm that exact fine-tuning based methods are generally effective within the range of their extension, whereas extrapolation remains challenging. All codebases, models, and checkpoints will be made available open-source, promoting transparency and facilitating further research in this critical area of AI development.", "authors": ["Yi Lu", "Jing Nathan Yan", "Songlin Yang", "Justin T. Chiu", "Siyu Ren", "Fei Yuan", "Wenting Zhao", "Zhiyong Wu", "Alexander M. Rush"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-18", "url": "https://arxiv.org/abs/2409.12181", "pdf_url": "https://arxiv.org/pdf/2409.12181v2", "arxiv_id": "2409.12181", "doi": "10.48550/arXiv.2409.12181", "citation_count": 25, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3537}
{"id": "923f310036846572c7d99458abaf7d29742e96c4812ae342a4a82d125ddd7e25", "sources": ["arxiv", "semantic_scholar"], "title": "E2LLM: Encoder Elongated Large Language Models for Long-Context Understanding and Reasoning", "abstract": "Processing long contexts is increasingly important for Large Language Models (LLMs) in tasks like multi-turn dialogues, code generation, and document summarization. This paper addresses the challenges of achieving high long-context performance, low computational complexity, and compatibility with pretrained models -- collectively termed the ``impossible triangle''. We introduce E2LLM (Encoder Elongated Large Language Models), a novel approach that effectively navigates this paradox. E2LLM divides long contexts into chunks, compresses each into soft prompts using a pretrained text encoder, and aligns these representations with a decoder-only LLM via an adapter. To enhance the LLM's reasoning with these soft prompts, we employ two training objectives: encoder output reconstruction and long-context instruction fine-tuning. Extensive experiments reveal that E2LLM not only outperforms 8 state-of-the-art (SOTA) methods in effectiveness and efficiency for document summarization and question answering, but also achieves the best performance on LongBench v2 among models of comparable size.", "authors": ["Zihan Liao", "Jun Wang", "Hang Yu", "Lingxiao Wei", "Jianguo Li", "Jun Wang", "Wei Zhang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-10", "url": "https://arxiv.org/abs/2409.06679", "pdf_url": "https://arxiv.org/pdf/2409.06679v3", "arxiv_id": "2409.06679", "doi": "10.48550/arXiv.2409.06679", "citation_count": 11, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.2698}
{"id": "ac623488f6780ca5ec73758363e559841cd42724c1d453447b743683ff87a743", "sources": ["arxiv", "semantic_scholar"], "title": "Untie the Knots: An Efficient Data Augmentation Strategy for Long-Context Pre-Training in Language Models", "abstract": "Large language models (LLM) have prioritized expanding the context window from which models can incorporate more information. However, training models to handle long contexts presents significant challenges. These include the scarcity of high-quality natural long-context data, the potential for performance degradation on short-context tasks, and the reduced training efficiency associated with attention mechanisms. In this paper, we introduce Untie the Knots (\\textbf{UtK}), a novel data augmentation strategy employed during the continue pre-training phase, designed to efficiently enable LLMs to gain long-context capabilities without the need to modify the existing data mixture. In particular, we chunk the documents, shuffle the chunks, and create a complex and knotted structure of long texts; LLMs are then trained to untie these knots and identify relevant segments within seemingly chaotic token sequences. This approach greatly improves the model's performance by accurately attending to relevant information in long context and the training efficiency is also largely increased. We conduct extensive experiments on models with 7B and 72B parameters, trained on 20 billion tokens, demonstrating that UtK achieves 75\\% and 84.5\\% accurracy on RULER at 128K context length, significantly outperforming other long context strategies. The trained models will open-source for further research.", "authors": ["Junfeng Tian", "Da Zheng", "Yang Cheng", "Rui Wang", "Colin Zhang", "Debing Zhang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-07", "url": "https://arxiv.org/abs/2409.04774", "pdf_url": "https://arxiv.org/pdf/2409.04774v1", "arxiv_id": "2409.04774", "doi": "10.48550/arXiv.2409.04774", "citation_count": 8, "influential_citation_count": 2, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2386}
{"id": "2923a24c01ab851bc1ff5e3e74217452a4b35f269eafecd09a7827d81d97529b", "sources": ["arxiv", "semantic_scholar"], "title": "Efficient LLM Context Distillation", "abstract": "Large Language Models (LLMs) demonstrate proficiency across diverse tasks but often require targeted adaptations for specific applications. Various methods have been proposed to facilitate this adaptation, including fewshot fine-tuning, in-context learning, and context distillation. This paper specifically investigates context distillation a method that extends the utility of task-specific examples by internalizing them, thus augmenting the example set accessible for model inference. We conduct a comparative analysis of context distillation with in-context learning (ICL) and few-shot fine-tuning (FT), aiming to ascertain the efficacy of context distillation in adapting models using minimal in-context examples. Employing matched datasets from Mobach, our experiments leverage OPT models of various sizes. The results indicate that context distillation effectively adapts models, with student models attaining comparable in-domain and out-of-domain accuracies to in-context learning. Although context distillation surpasses ICL in out-of-domain generalization, it does not achieve the performance levels of FT. However, the reduced dataset size and computational demands position context distillation as a viable alternative, especially for smaller datasets. Overall, this study presents context distillation as an efficient and potent method for customizing LLMs to specific tasks.", "authors": ["Rajesh Upadhayaya", "Manish Raj Osti", "Zachary Smith", "Chritopher Kottmyer"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-03", "url": "https://arxiv.org/abs/2409.01930", "pdf_url": "https://arxiv.org/pdf/2409.01930v2", "arxiv_id": "2409.01930", "doi": "10.48550/arXiv.2409.01930", "citation_count": 4, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1747}
{"id": "da17a8c72a2fd4405e0698fb9a265292575e7696499d60bed1d5a976d0230e1d", "sources": ["arxiv", "semantic_scholar"], "title": "In Defense of RAG in the Era of Long-Context Language Models", "abstract": "Overcoming the limited context limitations in early-generation LLMs, retrieval-augmented generation (RAG) has been a reliable solution for context-based answer generation in the past. Recently, the emergence of long-context LLMs allows the models to incorporate much longer text sequences, making RAG less attractive. Recent studies show that long-context LLMs significantly outperform RAG in long-context applications. Unlike the existing works favoring the long-context LLM over RAG, we argue that the extremely long context in LLMs suffers from a diminished focus on relevant information and leads to potential degradation in answer quality. This paper revisits the RAG in long-context answer generation. We propose an order-preserve retrieval-augmented generation (OP-RAG) mechanism, which significantly improves the performance of RAG for long-context question-answer applications. With OP-RAG, as the number of retrieved chunks increases, the answer quality initially rises, and then declines, forming an inverted U-shaped curve. There exist sweet points where OP-RAG could achieve higher answer quality with much less tokens than long-context LLM taking the whole context as input. Extensive experiments on public benchmark demonstrate the superiority of our OP-RAG.", "authors": ["Tan Yu", "Anbang Xu", "Rama Akkiraju"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-03", "url": "https://arxiv.org/abs/2409.01666", "pdf_url": "https://arxiv.org/pdf/2409.01666v1", "arxiv_id": "2409.01666", "doi": "10.48550/arXiv.2409.01666", "citation_count": 56, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.439}
{"id": "bfa094d51293499e513780fedd2fb25ec7b5ddbc490eb9c8a2760cdca7643664", "sources": ["arxiv", "semantic_scholar"], "title": "LongGenBench: Benchmarking Long-Form Generation in Long Context LLMs", "abstract": "Current benchmarks like Needle-in-a-Haystack (NIAH), Ruler, and Needlebench focus on models' ability to understand long-context input sequences but fail to capture a critical dimension: the generation of high-quality long-form text. Applications such as design proposals, technical documentation, and creative writing rely on coherent, instruction-following outputs over extended sequences - a challenge that existing benchmarks do not adequately address. To fill this gap, we introduce LongGenBench, a novel benchmark designed to rigorously evaluate large language models' (LLMs) ability to generate long text while adhering to complex instructions. Through tasks requiring specific events or constraints within generated text, LongGenBench evaluates model performance across four distinct scenarios, three instruction types, and two generation-lengths (16K and 32K tokens). Our evaluation of ten state-of-the-art LLMs reveals that, despite strong results on Ruler, all models struggled with long text generation on LongGenBench, particularly as text length increased. This suggests that current LLMs are not yet equipped to meet the demands of real-world, long-form text generation.", "authors": ["Yuhao Wu", "Ming Shan Hee", "Zhiqing Hu", "Roy Ka-Wei Lee"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-03", "url": "https://arxiv.org/abs/2409.02076", "pdf_url": "https://arxiv.org/pdf/2409.02076v7", "arxiv_id": "2409.02076", "doi": "10.48550/arXiv.2409.02076", "citation_count": 50, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/mozhu621/LongGenBench/", "venue": "International Conference on Learning Representations", "quality_score": 0.4269}
{"id": "4cec8ade66b9f306b9a0374945b52704453e0923ea85ed76992923ced6a77f9a", "sources": ["arxiv", "semantic_scholar"], "title": "You Only Use Reactive Attention Slice For Long Context Retrieval", "abstract": "Supporting longer context for Large Language Models (LLM) is a promising direction to advance LLMs. As training a model for a longer context window is computationally expensive, many alternative solutions, such as Retrieval Augmented Generation (RAG), have been used. However, most existing RAG methods adopt embedding-based retrieval that falls short on long contexts. To address such challenges, we propose an attention-based retrieval technique, You Only Use Reactive Attention slice (YOURA). YOURA leverages a novel retrieval heuristic called reaction score to rank the relevance of each sentence in the input context with the query sentence. Intuitively, we measure how the per-token attention score \"reacts\" to the query and greedily retrieves the most reactive sentences. Internally, YOURA generates a token-indexed vector (called reaction vector) for the whole input context. To map each sentence to the token-indexed vector, we propose an Embedding-Agnostic Sentence Yield (EASY), a best-effort token wiggling algorithm. We evaluate our retrieval technique on three open-source pre-trained LLM models across six LongBench QA datasets. Our technique achieves up to 30% vLLM inference throughput improvement for serving long-context queries with a nearly identical quality score to the simple yet effective truncate-middle approach.", "authors": ["Yun Joon Soh", "Hanxian Huang", "Yuandong Tian", "Jishen Zhao"], "categories": ["cs.CL", "cs.AI", "cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-03", "url": "https://arxiv.org/abs/2409.13695", "pdf_url": "https://arxiv.org/pdf/2409.13695v1", "arxiv_id": "2409.13695", "doi": "10.48550/arXiv.2409.13695", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "3093531b715aebea83e3c8cd9c695b2255d5e2ec3c7dd1c63e91f3dafaa86e29", "sources": ["arxiv", "semantic_scholar"], "title": "Prompt Compression with Context-Aware Sentence Encoding for Fast and Improved LLM Inference", "abstract": "Large language models (LLMs) have triggered a new stream of research focusing on compressing the context length to reduce the computational cost while ensuring the retention of helpful information for LLMs to answer the given question. Token-based removal methods are one of the most prominent approaches in this direction, but risk losing the semantics of the context caused by intermediate token removal, especially under high compression ratios, while also facing challenges in computational efficiency. In this work, we propose context-aware prompt compression (CPC), a sentence-level prompt compression technique where its key innovation is a novel context-aware sentence encoder that provides a relevance score for each sentence for a given question. To train this encoder, we generate a new dataset consisting of questions, positives, and negative pairs where positives are sentences relevant to the question, while negatives are irrelevant context sentences. We train the encoder in a contrastive setup to learn context-aware sentence representations. Our method considerably outperforms prior works on prompt compression on benchmark datasets and is up to 10.93x faster at inference compared to the best token-level compression method. We also find better improvement for shorter length constraints in most benchmarks, showing the effectiveness of our proposed solution in the compression of relevant information in a shorter context. Finally, we release the code and the dataset for quick reproducibility and further development: https://github.com/Workday/cpc.", "authors": ["Barys Liskavets", "Maxim Ushakov", "Shuvendu Roy", "Mark Klibanov", "Ali Etemad", "Shane Luke"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-02", "url": "https://arxiv.org/abs/2409.01227", "pdf_url": "https://arxiv.org/pdf/2409.01227v3", "arxiv_id": "2409.01227", "doi": "10.48550/arXiv.2409.01227", "citation_count": 47, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/Workday/cpc", "venue": "AAAI Conference on Artificial Intelligence", "quality_score": 0.4203}
{"id": "c9c9aa77c28721a0e415aad7b5c048bf9216d8ad65d5f5cdec3cdb23ea0c7c11", "sources": ["arxiv", "semantic_scholar"], "title": "DataSculpt: Crafting Data Landscapes for Long-Context LLMs through Multi-Objective Partitioning", "abstract": "In recent years, Large Language Models (LLMs) have demonstrated significant improvements across a variety of tasks, one of which is the long-context capability. The key to improving long-context performance lies in effective data organization and management strategies that integrate data from multiple domains and optimize the context window during training. Through extensive experimental analysis, we identified three key challenges in designing effective data management strategies that enable the model to achieve long-context capability without sacrificing performance in other tasks: (1) a shortage of long documents across multiple domains, (2) effective construction of context windows, and (3) efficient organization of large-scale datasets. To address these challenges, we introduce DataSculpt, a novel data management framework designed for long-context training. We first formulate the organization of training data as a multi-objective combinatorial optimization problem, focusing on attributes including relevance, homogeneity, integrity, and efficiency. Specifically, our approach utilizes a coarse-to-fine methodology to optimize training data organization both efficiently and effectively. We begin by clustering the data based on semantic similarity (coarse), followed by a multi-objective greedy search within each cluster to score and concatenate documents into various context windows (fine). Our comprehensive evaluations demonstrate that DataSculpt significantly enhances long-context training performance, resulting in improvements of 18.09% in retrieval augmentation, 21.23% in summarization, 21.27% in reading comprehension, and a 3.81% increase in code completion, while also maintaining overall model proficiency with a 4.88% improvement.", "authors": ["Keer Lu", "Xiaonan Nie", "Zheng Liang", "Da Pan", "Shusen Zhang", "Keshi Zhao", "Weipeng Chen", "Zenan Zhou", "Guosheng Dong", "Bin Cui", "Wentao Zhang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-09-02", "url": "https://arxiv.org/abs/2409.00997", "pdf_url": "https://arxiv.org/pdf/2409.00997v2", "arxiv_id": "2409.00997", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.0}
{"id": "dbe1ad89cc9b32fd327efaf8a6c09ee2a2641897188079a17b7949ff4c1a55d2", "sources": ["arxiv", "semantic_scholar"], "title": "LongRecipe: Recipe for Efficient Long Context Generalization in Large Language Models", "abstract": "Large language models (LLMs) face significant challenges in handling long-context tasks because of their limited effective context window size during pretraining, which restricts their ability to generalize over extended sequences. Meanwhile, extending the context window in LLMs through post-pretraining is highly resource-intensive. To address this, we introduce LongRecipe, an efficient training strategy for extending the context window of LLMs, including impactful token analysis, position index transformation, and training optimization strategies. It simulates long-sequence inputs while maintaining training efficiency and significantly improves the model's understanding of long-range dependencies. Experiments on three types of LLMs show that LongRecipe can utilize long sequences while requiring only 30% of the target context window size, and reduces computational training resource over 85% compared to full sequence training. Furthermore, LongRecipe also preserves the original LLM's capabilities in general tasks. Ultimately, we can extend the effective context window of open-source LLMs from 8k to 128k, achieving performance close to GPT-4 with just one day of dedicated training using a single GPU with 80G memory. Our code is released at https://github.com/zhiyuanhubj/LongRecipe.", "authors": ["Zhiyuan Hu", "Yuliang Liu", "Jinman Zhao", "Suyuchen Wang", "Yan Wang", "Wei Shen", "Qing Gu", "Anh Tuan Luu", "See-Kiong Ng", "Zhiwei Jiang", "Bryan Hooi"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-08-31", "url": "https://arxiv.org/abs/2409.00509", "pdf_url": "https://arxiv.org/pdf/2409.00509v2", "arxiv_id": "2409.00509", "doi": "10.48550/arXiv.2409.00509", "citation_count": 26, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/zhiyuanhubj/LongRecipe", "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.3578}
{"id": "281659834b1bdf4ef75131fea6c24f9cf115b0511739a8f41ffb04786f40b7ad", "sources": ["arxiv", "semantic_scholar"], "title": "Training Ultra Long Context Language Model with Fully Pipelined Distributed Transformer", "abstract": "Large Language Models (LLMs) with long context capabilities are integral to complex tasks in natural language processing and computational biology, such as text generation and protein sequence analysis. However, training LLMs directly on extremely long contexts demands considerable GPU resources and increased memory, leading to higher costs and greater complexity. Alternative approaches that introduce long context capabilities via downstream finetuning or adaptations impose significant design limitations. In this paper, we propose Fully Pipelined Distributed Transformer (FPDT) for efficiently training long-context LLMs with extreme hardware efficiency. For GPT and Llama models, we achieve a 16x increase in sequence length that can be trained on the same hardware compared to current state-of-the-art solutions. With our dedicated sequence chunk pipeline design, we can now train 8B LLM with 2 million sequence length on only 4 GPUs, while also maintaining over 55% of MFU. Our proposed FPDT is agnostic to existing training techniques and is proven to work efficiently across different LLM models.", "authors": ["Jinghan Yao", "Sam Ade Jacobs", "Masahiro Tanaka", "Olatunji Ruwase", "Hari Subramoni", "Dhabaleswar K. Panda"], "categories": ["cs.DC", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-08-30", "url": "https://arxiv.org/abs/2408.16978", "pdf_url": "https://arxiv.org/pdf/2408.16978v2", "arxiv_id": "2408.16978", "doi": "10.48550/arXiv.2408.16978", "citation_count": 10, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Conference on Machine Learning and Systems", "quality_score": 0.2603}
{"id": "07c7c547b873f2733c7252c93bd0be7c3142263ce0951f6a3e79ebec3acef273", "sources": ["arxiv", "semantic_scholar"], "title": "FocusLLM: Precise Understanding of Long Context by Dynamic Condensing", "abstract": "Empowering LLMs with the ability to precisely understand long contexts is crucial for many downstream applications. However, handling long contexts with conventional transformer architecture requires substantial training and inference resources. Existing context condensing methods cannot accurately understand the full context, as there is a considerable amount of information loss in the condensing process. To address these issues, we present FocusLLM, a framework designed to extend the fixed context length of any decoder-only LLM, allowing the model to focus on relevant information from very long sequences. FocusLLM first divides long text input into chunks based on the model's original context length. It then employs the dynamic condensing process to distill crucial information from each chunk. Ultimately, through the novel parallel decoding mechanism, FocusLLM can integrate the extracted information into its local context. FocusLLM stands out for great training efficiency and versatility: trained with an 8K input length and with much less training cost than previous methods, FocusLLM exhibits superior performance across downstream tasks and maintains strong language modeling ability when handling extensive long texts, even up to 400K tokens. Our code is available at https://github.com/leezythu/FocusLLM.", "authors": ["Zhenyu Li", "Yike Zhang", "Tengyu Pan", "Yutao Sun", "Zhichao Duan", "Junjie Fang", "Rong Han", "Zixuan Wang", "Jianyong Wang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-08-21", "url": "https://arxiv.org/abs/2408.11745", "pdf_url": "https://arxiv.org/pdf/2408.11745v2", "arxiv_id": "2408.11745", "doi": null, "citation_count": 14, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/leezythu/FocusLLM", "venue": null, "quality_score": 0.294}
{"id": "0cc4df70aaa00930cc44e9d1a294b362214fc58e2a36cd83da2b021817dfcebb", "sources": ["arxiv", "semantic_scholar"], "title": "Multilingual Needle in a Haystack: Investigating Long-Context Behavior of Multilingual Large Language Models", "abstract": "While recent large language models (LLMs) demonstrate remarkable abilities in responding to queries in diverse languages, their ability to handle long multilingual contexts is unexplored. As such, a systematic evaluation of the long-context capabilities of LLMs in multilingual settings is crucial, specifically in the context of information retrieval. To address this gap, we introduce the MultiLingual Needle-in-a-Haystack (MLNeedle) test, designed to assess a model's ability to retrieve relevant information (the needle) from a collection of multilingual distractor texts (the haystack). This test serves as an extension of the multilingual question-answering task, encompassing both monolingual and cross-lingual retrieval. We evaluate four state-of-the-art LLMs on MLNeedle. Our findings reveal that model performance can vary significantly with language and needle position. Specifically, we observe that model performance is the lowest when the needle is (i) in a language outside the English language family and (ii) located in the middle of the input context. Furthermore, although some models claim a context size of $8k$ tokens or greater, none demonstrate satisfactory cross-lingual retrieval performance as the context length increases. Our analysis provides key insights into the long-context behavior of LLMs in multilingual settings to guide future evaluation protocols. To our knowledge, this is the first study to investigate the multilingual long-context behavior of LLMs.", "authors": ["Amey Hengle", "Prasoon Bajpai", "Soham Dan", "Tanmoy Chakraborty"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-08-19", "url": "https://arxiv.org/abs/2408.10151", "pdf_url": "https://arxiv.org/pdf/2408.10151v1", "arxiv_id": "2408.10151", "doi": "10.48550/arXiv.2408.10151", "citation_count": 10, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "North American Chapter of the Association for Computational Linguistics", "quality_score": 0.2603}
{"id": "c82679d50aea4dc28fcf9e5e2c6013b6fbc293964f005ce29df1fca7ad5ac56f", "sources": ["arxiv", "semantic_scholar"], "title": "LongVILA: Scaling Long-Context Visual Language Models for Long Videos", "abstract": "Long-context capability is critical for multi-modal foundation models, especially for long video understanding. We introduce LongVILA, a full-stack solution for long-context visual-language models by co-designing the algorithm and system. For model training, we upgrade existing VLMs to support long video understanding by incorporating two additional stages, i.e., long context extension and long video supervised fine-tuning. However, training on long video is computationally and memory intensive. We introduce the long-context Multi-Modal Sequence Parallelism (MM-SP) system that efficiently parallelizes long video training and inference, enabling 2M context length training on 256 GPUs without any gradient checkpointing. LongVILA efficiently extends the number of video frames of VILA from 8 to 2048, achieving 99.8% accuracy in 6,000-frame (more than 1 million tokens) video needle-in-a-haystack. LongVILA-7B demonstrates strong accuracy on 9 popular video benchmarks, e.g. 65.1% VideoMME with subtitle. Besides, MM-SP is 2.1x - 5.7x faster than ring style sequence parallelism and 1.1x - 1.4x faster than Megatron with a hybrid context and tensor parallelism. Moreover, it seamlessly integrates with Hugging Face Transformers.", "authors": ["Yukang Chen", "Fuzhao Xue", "Dacheng Li", "Qinghao Hu", "Ligeng Zhu", "Xiuyu Li", "Yunhao Fang", "Haotian Tang", "Shang Yang", "Zhijian Liu", "Ethan He", "Hongxu Yin", "Pavlo Molchanov", "Jan Kautz", "Linxi Fan", "Yuke Zhu", "Yao Lu", "Song Han"], "categories": ["cs.CV", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-08-19", "url": "https://arxiv.org/abs/2408.10188", "pdf_url": "https://arxiv.org/pdf/2408.10188v6", "arxiv_id": "2408.10188", "doi": "10.48550/arXiv.2408.10188", "citation_count": 287, "influential_citation_count": 33, "has_code": true, "code_url": "https://github.com/NVlabs/VILA/tree/main/longvila", "venue": "International Conference on Learning Representations", "quality_score": 0.7657}
{"id": "c8e7cd0ec4b886272c01973f5b1df2418332d137b1c3249239108f3f85799b6c", "sources": ["arxiv", "semantic_scholar"], "title": "LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs", "abstract": "Current long context large language models (LLMs) can process inputs up to 100,000 tokens, yet struggle to generate outputs exceeding even a modest length of 2,000 words. Through controlled experiments, we find that the model's effective generation length is inherently bounded by the sample it has seen during supervised fine-tuning (SFT). In other words, their output limitation is due to the scarcity of long-output examples in existing SFT datasets. To address this, we introduce AgentWrite, an agent-based pipeline that decomposes ultra-long generation tasks into subtasks, enabling off-the-shelf LLMs to generate coherent outputs exceeding 20,000 words. Leveraging AgentWrite, we construct LongWriter-6k, a dataset containing 6,000 SFT data with output lengths ranging from 2k to 32k words. By incorporating this dataset into model training, we successfully scale the output length of existing models to over 10,000 words while maintaining output quality. We also develop LongBench-Write, a comprehensive benchmark for evaluating ultra-long generation capabilities. Our 9B parameter model, further improved through DPO, achieves state-of-the-art performance on this benchmark, surpassing even much larger proprietary models. In general, our work demonstrates that existing long context LLM already possesses the potential for a larger output window--all you need is data with extended output during model alignment to unlock this capability. Our code & models are at: https://github.com/THUDM/LongWriter.", "authors": ["Yushi Bai", "Jiajie Zhang", "Xin Lv", "Linzhi Zheng", "Siqi Zhu", "Lei Hou", "Yuxiao Dong", "Jie Tang", "Juanzi Li"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-08-13", "url": "https://arxiv.org/abs/2408.07055", "pdf_url": "https://arxiv.org/pdf/2408.07055v1", "arxiv_id": "2408.07055", "doi": "10.48550/arXiv.2408.07055", "citation_count": 140, "influential_citation_count": 32, "has_code": true, "code_url": "https://github.com/THUDM/LongWriter", "venue": "International Conference on Learning Representations", "quality_score": 0.7593}
{"id": "53f3b7abc0f22ed4f2fa960c9999ada8ca7d24a6855fc99e60a50b5a4a865509", "sources": ["arxiv", "semantic_scholar"], "title": "Efficient Solutions For An Intriguing Failure of LLMs: Long Context Window Does Not Mean LLMs Can Analyze Long Sequences Flawlessly", "abstract": "Large Language Models (LLMs) have demonstrated remarkable capabilities in comprehending and analyzing lengthy sequential inputs, owing to their extensive context windows that allow processing millions of tokens in a single forward pass. However, this paper uncovers a surprising limitation: LLMs fall short when handling long input sequences. We investigate this issue using three datasets and two tasks (sentiment analysis and news categorization) across various LLMs, including Claude 3, Gemini Pro, GPT 3.5 Turbo, Llama 3 Instruct, and Mistral Instruct models. To address this limitation, we propose and evaluate ad-hoc solutions that substantially enhance LLMs' performance on long input sequences by up to 50%, while reducing API cost and latency by up to 93% and 50%, respectively.", "authors": ["Peyman Hosseini", "Ignacio Castro", "Iacopo Ghinassi", "Matthew Purver"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-08-03", "url": "https://arxiv.org/abs/2408.01866", "pdf_url": "https://arxiv.org/pdf/2408.01866v3", "arxiv_id": "2408.01866", "doi": "10.48550/arXiv.2408.01866", "citation_count": 28, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Computational Linguistics", "quality_score": 0.3656}
{"id": "812ce25da387c152461894dbb94306ce42a9b405e8e04a45ec1989e312ff3464", "sources": ["arxiv", "semantic_scholar"], "title": "Large Language Model (LLM)-enabled In-context Learning for Wireless Network Optimization: A Case Study of Power Control", "abstract": "Large language model (LLM) has recently been considered a promising technique for many fields. This work explores LLM-based wireless network optimization via in-context learning. To showcase the potential of LLM technologies, we consider the base station (BS) power control as a case study, a fundamental but crucial technique that is widely investigated in wireless networks. Different from existing machine learning (ML) methods, our proposed in-context learning algorithm relies on LLM's inference capabilities. It avoids the complexity of tedious model training and hyper-parameter fine-tuning, which is a well-known bottleneck of many ML algorithms. Specifically, the proposed algorithm first describes the target task via formatted natural language, and then designs the in-context learning framework and demonstration examples. After that, it considers two cases, namely discrete-state and continuous-state problems, and proposes state-based and ranking-based methods to select appropriate examples for these two cases, respectively. Finally, the simulations demonstrate that the proposed algorithm can achieve comparable performance as conventional deep reinforcement learning (DRL) techniques without dedicated model training or fine-tuning. Such an efficient and low-complexity approach has great potential for future wireless network optimization.", "authors": ["Hao Zhou", "Chengming Hu", "Dun Yuan", "Ye Yuan", "Di Wu", "Xue Liu", "Charlie Zhang"], "categories": ["eess.SY"], "fields_of_study": ["Engineering", "Computer Science"], "published_date": "2024-08-01", "url": "https://arxiv.org/abs/2408.00214", "pdf_url": "https://arxiv.org/pdf/2408.00214v2", "arxiv_id": "2408.00214", "doi": "10.48550/arXiv.2408.00214", "citation_count": 41, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4058}
{"id": "3d5805b04a4d030ae15cd9b195119e5e4c4267ec8a5ce688f3dda466a52b0d3f", "sources": ["arxiv", "semantic_scholar"], "title": "mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval", "abstract": "We present systematic efforts in building long-context multilingual text representation model (TRM) and reranker from scratch for text retrieval. We first introduce a text encoder (base size) enhanced with RoPE and unpadding, pre-trained in a native 8192-token context (longer than 512 of previous multilingual encoders). Then we construct a hybrid TRM and a cross-encoder reranker by contrastive learning. Evaluations show that our text encoder outperforms the same-sized previous state-of-the-art XLM-R. Meanwhile, our TRM and reranker match the performance of large-sized state-of-the-art BGE-M3 models and achieve better results on long-context retrieval benchmarks. Further analysis demonstrate that our proposed models exhibit higher efficiency during both training and inference. We believe their efficiency and effectiveness could benefit various researches and industrial applications.", "authors": ["Xin Zhang", "Yanzhao Zhang", "Dingkun Long", "Wen Xie", "Ziqi Dai", "Jialong Tang", "Huan Lin", "Baosong Yang", "Pengjun Xie", "Fei Huang", "Meishan Zhang", "Wenjie Li", "Min Zhang"], "categories": ["cs.CL", "cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-29", "url": "https://arxiv.org/abs/2407.19669", "pdf_url": "https://arxiv.org/pdf/2407.19669v2", "arxiv_id": "2407.19669", "doi": "10.48550/arXiv.2407.19669", "citation_count": 326, "influential_citation_count": 23, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.6901}
{"id": "5222d84c8e65923ca542bbfee52186ba76b5d9f0e4a9bde9341ca0919ddb998c", "sources": ["arxiv", "semantic_scholar"], "title": "Polynomial Regression as a Task for Understanding In-context Learning Through Finetuning and Alignment", "abstract": "Simple function classes have emerged as toy problems to better understand in-context-learning in transformer-based architectures used for large language models. But previously proposed simple function classes like linear regression or multi-layer-perceptrons lack the structure required to explore things like prompting and alignment within models capable of in-context-learning. We propose univariate polynomial regression as a function class that is just rich enough to study prompting and alignment, while allowing us to visualize and understand what is going on clearly.", "authors": ["Max Wilcoxson", "Morten Svendgård", "Ria Doshi", "Dylan Davis", "Reya Vir", "Anant Sahai"], "categories": ["cs.LG", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-27", "url": "https://arxiv.org/abs/2407.19346", "pdf_url": "https://arxiv.org/pdf/2407.19346v1", "arxiv_id": "2407.19346", "doi": "10.48550/arXiv.2407.19346", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "bd214f34b4cd7be4932cd4706036f5d6aaf73cc18911bd5ad052b1a4517ea34f", "sources": ["arxiv", "semantic_scholar"], "title": "Retrieval Augmented Generation or Long-Context LLMs? A Comprehensive Study and Hybrid Approach", "abstract": "Retrieval Augmented Generation (RAG) has been a powerful tool for Large Language Models (LLMs) to efficiently process overly lengthy contexts. However, recent LLMs like Gemini-1.5 and GPT-4 show exceptional capabilities to understand long contexts directly. We conduct a comprehensive comparison between RAG and long-context (LC) LLMs, aiming to leverage the strengths of both. We benchmark RAG and LC across various public datasets using three latest LLMs. Results reveal that when resourced sufficiently, LC consistently outperforms RAG in terms of average performance. However, RAG's significantly lower cost remains a distinct advantage. Based on this observation, we propose Self-Route, a simple yet effective method that routes queries to RAG or LC based on model self-reflection. Self-Route significantly reduces the computation cost while maintaining a comparable performance to LC. Our findings provide a guideline for long-context applications of LLMs using RAG and LC.", "authors": ["Zhuowan Li", "Cheng Li", "Mingyang Zhang", "Qiaozhu Mei", "Michael Bendersky"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-23", "url": "https://arxiv.org/abs/2407.16833", "pdf_url": "https://arxiv.org/pdf/2407.16833v2", "arxiv_id": "2407.16833", "doi": "10.48550/arXiv.2407.16833", "citation_count": 150, "influential_citation_count": 9, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.5447}
{"id": "b64b977922a18f14ac21a9e873532af2ac90abf7faa37432313c222d609ab577", "sources": ["arxiv", "semantic_scholar"], "title": "ChatQA 2: Bridging the Gap to Proprietary LLMs in Long Context and RAG Capabilities", "abstract": "In this work, we introduce ChatQA 2, an Llama 3.0-based model with a 128K context window, designed to bridge the gap between open-source LLMs and leading proprietary models (e.g., GPT-4-Turbo-2024-04-09) in long context understanding and retrieval-augmented generation (RAG) capabilities. These two capabilities are complementary to each other and essential for LLMs to process large volumes of information that cannot fit into a single prompt. We present a detailed continued training recipe to extend the context window of Llama3-70B-base from 8K to 128K tokens, along with a three-stage instruction tuning process to enhance the model's instruction-following, RAG performance, and long-context understanding capabilities. Our results demonstrate that the Llama3-ChatQA-2-70B model outperforms most existing state-of-the-art models, including GPT-4-Turbo-2024-04-09, Qwen2-72B-Instruct, and Llama3.1-70B-Instruct, on ultra-long tasks beyond 100K tokens, as well as on the RAG benchmark using only a 4K context window, showing the strong long context capability across varying sequence lengths. We further provide extensive comparisons between direct long-context and RAG solutions using the same state-of-the-art long-context LLMs. Interestingly, we find that the performance of strong long-context LLMs using RAG improves when retrieving a larger number of chunks. With a large set of top-k chunks, RAG consistently outperforms direct long-context solution using the same state-of-the-art long-context models (e.g., Llama3-ChatQA-2-70B and Qwen2-72B-Instruct) on both 32K and 128K benchmarks. We open-source the model weights, training data, and the evaluation setup for the for the community: https://chatqa2-project.github.io/", "authors": ["Peng Xu", "Wei Ping", "Xianchao Wu", "Chejian Xu", "Zihan Liu", "Mohammad Shoeybi", "Bryan Catanzaro"], "categories": ["cs.CL", "cs.AI", "cs.IR", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-19", "url": "https://arxiv.org/abs/2407.14482", "pdf_url": "https://arxiv.org/pdf/2407.14482v3", "arxiv_id": "2407.14482", "doi": "10.48550/arXiv.2407.14482", "citation_count": 52, "influential_citation_count": 3, "has_code": true, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.4311}
{"id": "f9139cf88612d4b5b99aa3cebc8e34de32a35abb91690a1c8987d9bed9151464", "sources": ["arxiv", "semantic_scholar"], "title": "MEMO: Fine-grained Tensor Management For Ultra-long Context LLM Training", "abstract": "Nowadays, Large Language Models (LLMs) have been trained using extended context lengths to foster more creative applications. However, long context training poses great challenges considering the constraint of GPU memory. It not only leads to substantial activation memory consumption during training, but also incurs considerable memory fragmentation. To facilitate long context training, existing frameworks have adopted strategies such as recomputation and various forms of parallelisms. Nevertheless, these techniques rely on redundant computation or extensive communication, resulting in low Model FLOPS Utilization (MFU). In this paper, we propose MEMO, a novel LLM training framework designed for fine-grained activation memory management. Given the quadratic scaling of computation and linear scaling of memory with sequence lengths when using FlashAttention, we offload memory-consuming activations to CPU memory after each layer's forward pass and fetch them during the backward pass. To maximize the swapping of activations without hindering computation, and to avoid exhausting limited CPU memory, we implement a token-wise activation recomputation and swapping mechanism. Furthermore, we tackle the memory fragmentation issue by employing a bi-level Mixed Integer Programming (MIP) approach, optimizing memory reuse across transformer layers. Empirical results demonstrate that MEMO achieves an average of 1.97x and 1.80x MFU compared to Megatron-LM and DeepSpeed, respectively. This improvement is attributed to MEMO's ability to minimize memory fragmentation, reduce recomputation and intensive communication, and circumvent the delays associated with the memory reorganization process due to fragmentation. By leveraging fine-grained activation memory management, MEMO facilitates efficient training of 7B LLM with 1 million sequence length on just 8 A800 GPUs, achieving an MFU of 52.30%.", "authors": ["Pinxue Zhao", "Hailin Zhang", "Fangcheng Fu", "Xiaonan Nie", "Qibin Liu", "Fang Yang", "Yuanbo Peng", "Dian Jiao", "Shuaipeng Li", "Jinbao Xue", "Yangyu Tao", "Bin Cui"], "categories": ["cs.LG", "cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-16", "url": "https://arxiv.org/abs/2407.12117", "pdf_url": "https://arxiv.org/pdf/2407.12117v3", "arxiv_id": "2407.12117", "doi": "10.1145/3709703", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2386}
{"id": "6ff003afc1ad06a382987a59df7897a2d0931ed54c5a286a2f503cfbfac07062", "sources": ["arxiv", "semantic_scholar"], "title": "Fine-Tuning Medical Language Models for Enhanced Long-Contextual Understanding and Domain Expertise", "abstract": "Large Language Models (LLMs) have been widely applied in various professional fields. By fine-tuning the models using domain specific question and answer datasets, the professional domain knowledge and Q\\&A abilities of these models have significantly improved, for example, medical professional LLMs that use fine-tuning of doctor-patient Q\\&A data exhibit extraordinary disease diagnostic abilities. However, we observed that despite improvements in specific domain knowledge, the performance of medical LLM in long-context understanding has significantly declined, especially compared to general language models with similar parameters. The purpose of this study is to investigate the phenomenon of reduced performance in understanding long-context in medical LLM. We designed a series of experiments to conduct open-book professional knowledge exams on all models to evaluate their ability to read long-context. By adjusting the proportion and quantity of general data and medical data in the process of fine-tuning, we can determine the best data composition to optimize the professional model and achieve a balance between long-context performance and specific domain knowledge.", "authors": ["Qimin Yang", "Rongsheng Wang", "Jiexin Chen", "Runqi Su", "Tao Tan"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-16", "url": "https://arxiv.org/abs/2407.11536", "pdf_url": "https://arxiv.org/pdf/2407.11536v1", "arxiv_id": "2407.11536", "doi": "10.48550/arXiv.2407.11536", "citation_count": 12, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2785}
{"id": "68ec441ec000d069e39bb22d957ba1f83172d8506df1776db1f6862db852809b", "sources": ["arxiv", "semantic_scholar"], "title": "ST-RetNet: A Long-term Spatial-Temporal Traffic Flow Prediction Method", "abstract": "Traffic flow forecasting is considered a critical task in the field of intelligent transportation systems. In this paper, to address the issue of low accuracy in long-term forecasting of spatial-temporal big data on traffic flow, we propose an innovative model called Spatial-Temporal Retentive Network (ST-RetNet). We extend the Retentive Network to address the task of traffic flow forecasting. At the spatial scale, we integrate a topological graph structure into Spatial Retentive Network(S-RetNet), utilizing an adaptive adjacency matrix to extract dynamic spatial features of the road network. We also employ Graph Convolutional Networks to extract static spatial features of the road network. These two components are then fused to capture dynamic and static spatial correlations. At the temporal scale, we propose the Temporal Retentive Network(T-RetNet), which has been demonstrated to excel in capturing long-term dependencies in traffic flow patterns compared to other time series models, including Recurrent Neural Networks based and transformer models. We achieve the spatial-temporal traffic flow forecasting task by integrating S-RetNet and T-RetNet to form ST-RetNet. Through experimental comparisons conducted on four real-world datasets, we demonstrate that ST-RetNet outperforms the state-of-the-art approaches in traffic flow forecasting.", "authors": ["Baichao Long", "Wang Zhu", "Jianli Xiao"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-13", "url": "https://arxiv.org/abs/2407.11074", "pdf_url": "https://arxiv.org/pdf/2407.11074v1", "arxiv_id": "2407.11074", "doi": "10.48550/arXiv.2407.11074", "citation_count": 9, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Chinese Conference on Pattern Recognition and Computer Vision", "quality_score": 0.25}
{"id": "cb9ff49fb86842a39abfe7a4eb641c43287f7f2168cff378b5cd72a85c11862b", "sources": ["arxiv", "semantic_scholar"], "title": "Evaluating Language Model Context Windows: A \"Working Memory\" Test and Inference-time Correction", "abstract": "Large language models are prominently used in real-world applications, often tasked with reasoning over large volumes of documents. An exciting development in this space is models boasting extended context capabilities, with some accommodating over 2 million tokens. Such long context model capabilities remain uncertain in production systems, motivating the need to benchmark their performance on real world use cases. We address this challenge by proposing SWiM, an evaluation framework that addresses the limitations of standard tests. Testing the framework on eight long context models, we find that even strong models such as GPT-4 and Claude 3 Opus degrade in performance when information is present in the middle of the context window (lost-in-the-middle effect). Next, in addition to our benchmark, we propose medoid voting, a simple, but effective training-free approach that helps alleviate this effect, by generating responses a few times, each time randomly permuting documents in the context, and selecting the medoid answer. We evaluate medoid voting on single document QA tasks, achieving up to a 24% lift in accuracy. Our code is available at https://github.com/snorkel-ai/long-context-eval.", "authors": ["Amanda Dsouza", "Christopher Glaze", "Changho Shin", "Frederic Sala"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-07-04", "url": "https://arxiv.org/abs/2407.03651", "pdf_url": "https://arxiv.org/pdf/2407.03651v2", "arxiv_id": "2407.03651", "doi": "10.48550/arXiv.2407.03651", "citation_count": 7, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/snorkel-ai/long-context-eval", "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "466ce27ee840ebcc963c30830831473fe234c46d11df699dc4385fefcda30a09", "sources": ["arxiv", "semantic_scholar"], "title": "Is It Really Long Context if All You Need Is Retrieval? Towards Genuinely Difficult Long Context NLP", "abstract": "Improvements in language models' capabilities have pushed their applications towards longer contexts, making long-context evaluation and development an active research area. However, many disparate use-cases are grouped together under the umbrella term of \"long-context\", defined simply by the total length of the model's input, including - for example - Needle-in-a-Haystack tasks, book summarization, and information aggregation. Given their varied difficulty, in this position paper we argue that conflating different tasks by their context length is unproductive. As a community, we require a more precise vocabulary to understand what makes long-context tasks similar or different. We propose to unpack the taxonomy of long-context based on the properties that make them more difficult with longer contexts. We propose two orthogonal axes of difficulty: (I) Diffusion: How hard is it to find the necessary information in the context? (II) Scope: How much necessary information is there to find? We survey the literature on long-context, provide justification for this taxonomy as an informative descriptor, and situate the literature with respect to it. We conclude that the most difficult and interesting settings, whose necessary information is very long and highly diffused within the input, is severely under-explored. By using a descriptive vocabulary and discussing the relevant properties of difficulty in long-context, we can implement more informed research in this area. We call for a careful design of tasks and benchmarks with distinctly long context, taking into account the characteristics that make it qualitatively different from shorter context.", "authors": ["Omer Goldman", "Alon Jacovi", "Aviv Slobodkin", "Aviya Maimon", "Ido Dagan", "Reut Tsarfaty"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-29", "url": "https://arxiv.org/abs/2407.00402", "pdf_url": "https://arxiv.org/pdf/2407.00402v4", "arxiv_id": "2407.00402", "doi": "10.48550/arXiv.2407.00402", "citation_count": 25, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.3537}
{"id": "30c8d904e2f40e80763dc30a22497c1590e0b68b1367d3a49ed58a2b14a0e155", "sources": ["arxiv", "semantic_scholar"], "title": "Mixture of In-Context Experts Enhance LLMs' Long Context Awareness", "abstract": "Many studies have revealed that large language models (LLMs) exhibit uneven awareness of different contextual positions. Their limited context awareness can lead to overlooking critical information and subsequent task failures. While several approaches have been proposed to enhance LLMs' context awareness, achieving both effectiveness and efficiency remains challenging. In this paper, for LLMs utilizing RoPE as position embeddings, we introduce a novel method called \"Mixture of In-Context Experts\" (MoICE) to address this challenge. MoICE comprises two key components: a router integrated into each attention head within LLMs and a lightweight router-only training optimization strategy: (1) MoICE views each RoPE angle as an `in-context' expert, demonstrated to be capable of directing the attention of a head to specific contextual positions. Consequently, each attention head flexibly processes tokens using multiple RoPE angles dynamically selected by the router to attend to the needed positions. This approach mitigates the risk of overlooking essential contextual information. (2) The router-only training strategy entails freezing LLM parameters and exclusively updating routers for only a few steps. When applied to open-source LLMs including Llama and Mistral, MoICE surpasses prior methods across multiple tasks on long context understanding and generation, all while maintaining commendable inference efficiency.", "authors": ["Hongzhan Lin", "Ang Lv", "Yuhan Chen", "Chen Zhu", "Yang Song", "Hengshu Zhu", "Rui Yan"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-28", "url": "https://arxiv.org/abs/2406.19598", "pdf_url": "https://arxiv.org/pdf/2406.19598v2", "arxiv_id": "2406.19598", "doi": "10.48550/arXiv.2406.19598", "citation_count": 23, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.3451}
{"id": "5725f979eb1d76b8cf364881842ada4eab37fb60d2a93a88cebc10a580b41e4c", "sources": ["arxiv", "semantic_scholar"], "title": "UIO-LLMs: Unbiased Incremental Optimization for Long-Context LLMs", "abstract": "Managing long texts is challenging for large language models (LLMs) due to limited context window sizes. This study introduces UIO-LLMs, an unbiased incremental optimization approach for memory-enhanced transformers under long-context settings. We initially conceptualize the process as a streamlined encoder-decoder framework where the weights-shared encoder and decoder respectively encapsulate a context segment into memories and leverage these memories to predict outputs of the subsequent segment. Subsequently, by treating our memory-enhanced transformers as fully-connected recurrent neural networks (RNNs), we refine the training process using the Truncated Backpropagation Through Time (TBPTT) algorithm, which incorporates innovative incremental optimization techniques. These techniques not only diminish time complexity but also address the bias in gradient computation through an unbiased optimization process. UIO-LLMs successfully handle long context, such as extending the context window of Llama2-7b-chat from 4K to 100K tokens with minimal 2% additional parameters, while keeping the inference cost nearly linear as context length increases.", "authors": ["Wenhao Li", "Mingbao Lin", "Yunshan Zhong", "Shuicheng Yan", "Rongrong Ji"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-26", "url": "https://arxiv.org/abs/2406.18173", "pdf_url": "https://arxiv.org/pdf/2406.18173v3", "arxiv_id": "2406.18173", "doi": "10.48550/arXiv.2406.18173", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0}
{"id": "150f253c5512d983bdbe481ccfcdec5f211a577978285f9d050b35d171b73890", "sources": ["arxiv", "semantic_scholar"], "title": "LoongTrain: Efficient Training of Long-Sequence LLMs with Head-Context Parallelism", "abstract": "Efficiently training LLMs with long sequences is important yet challenged by the massive computation and memory requirements. Sequence parallelism has been proposed to tackle these problems, but existing methods suffer from scalability or efficiency issues. We propose LoongTrain, a novel system to efficiently train LLMs with long sequences at scale. The core of LoongTrain is the 2D-Attention mechanism, which combines both head-parallel and context-parallel techniques to break the scalability constraints while maintaining efficiency. We introduce Double-Ring-Attention and analyze the performance of device placement strategies to further speed up training. We implement LoongTrain with the hybrid ZeRO and Selective Checkpoint++ techniques. Experiment results show that LoongTrain outperforms state-of-the-art baselines, i.e., DeepSpeed-Ulysses and Megatron Context Parallelism, in both end-to-end training speed and scalability, and improves Model FLOPs Utilization (MFU) by up to 2.88x.", "authors": ["Diandian Gu", "Peng Sun", "Qinghao Hu", "Ting Huang", "Xun Chen", "Yingtong Xiong", "Guoteng Wang", "Qiaoling Chen", "Shangchun Zhao", "Jiarui Fang", "Yonggang Wen", "Tianwei Zhang", "Xin Jin", "Xuanzhe Liu"], "categories": ["cs.DC"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-26", "url": "https://arxiv.org/abs/2406.18485", "pdf_url": "https://arxiv.org/pdf/2406.18485v1", "arxiv_id": "2406.18485", "doi": "10.48550/arXiv.2406.18485", "citation_count": 25, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3537}
{"id": "0a713c947e76df9f79d8bafea18d094dd1d9fbcc38181d99cbb0688c208ee876", "sources": ["arxiv", "semantic_scholar"], "title": "LongIns: A Challenging Long-context Instruction-based Exam for LLMs", "abstract": "The long-context capabilities of large language models (LLMs) have been a hot topic in recent years. To evaluate the performance of LLMs in different scenarios, various assessment benchmarks have emerged. However, as most of these benchmarks focus on identifying key information to answer questions, which mainly requires the retrieval ability of LLMs, these benchmarks can partially represent the reasoning performance of LLMs from large amounts of information. Meanwhile, although LLMs often claim to have context windows of 32k, 128k, 200k, or even longer, these benchmarks fail to reveal the actual supported length of these LLMs. To address these issues, we propose the LongIns benchmark dataset, a challenging long-context instruction-based exam for LLMs, which is built based on the existing instruction datasets. Specifically, in our LongIns, we introduce three evaluation settings: Global Instruction & Single Task (GIST), Local Instruction & Single Task (LIST), and Local Instruction & Multiple Tasks (LIMT). Based on LongIns, we perform comprehensive evaluations on existing LLMs and have the following important findings: (1). The top-performing GPT-4 with 128k context length performs poorly on the evaluation context window of 16k in our LongIns. (2). For the multi-hop reasoning ability of many existing LLMs, significant efforts are still needed under short context windows (less than 4k).", "authors": ["Shawn Gavin", "Tuney Zheng", "Jiaheng Liu", "Quehry Que", "Noah Wang", "Jian Yang", "Chenchen Zhang", "Wenhao Huang", "Ge Zhang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-25", "url": "https://arxiv.org/abs/2406.17588", "pdf_url": "https://arxiv.org/pdf/2406.17588v3", "arxiv_id": "2406.17588", "doi": "10.48550/arXiv.2406.17588", "citation_count": 8, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2386}
{"id": "f893477394f8cacc2e8d33005429cf8cc1c203ab0ea5b35905fca901c753883c", "sources": ["arxiv", "semantic_scholar"], "title": "Leave No Document Behind: Benchmarking Long-Context LLMs with Extended Multi-Doc QA", "abstract": "Long-context modeling capabilities have garnered widespread attention, leading to the emergence of Large Language Models (LLMs) with ultra-context windows. Meanwhile, benchmarks for evaluating long-context LLMs are gradually catching up. However, existing benchmarks employ irrelevant noise texts to artificially extend the length of test cases, diverging from the real-world scenarios of long-context applications. To bridge this gap, we propose a novel long-context benchmark, Loong, aligning with realistic scenarios through extended multi-document question answering (QA). Unlike typical document QA, in Loong's test cases, each document is relevant to the final answer, ignoring any document will lead to the failure of the answer. Furthermore, Loong introduces four types of tasks with a range of context lengths: Spotlight Locating, Comparison, Clustering, and Chain of Reasoning, to facilitate a more realistic and comprehensive evaluation of long-context understanding. Extensive experiments indicate that existing long-context language models still exhibit considerable potential for enhancement. Retrieval augmented generation (RAG) achieves poor performance, demonstrating that Loong can reliably assess the model's long-context modeling capabilities.", "authors": ["Minzheng Wang", "Longze Chen", "Cheng Fu", "Shengyi Liao", "Xinghua Zhang", "Bingli Wu", "Haiyang Yu", "Nan Xu", "Lei Zhang", "Run Luo", "Yunshui Li", "Min Yang", "Fei Huang", "Yongbin Li"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-25", "url": "https://arxiv.org/abs/2406.17419", "pdf_url": "https://arxiv.org/pdf/2406.17419v2", "arxiv_id": "2406.17419", "doi": "10.48550/arXiv.2406.17419", "citation_count": 138, "influential_citation_count": 9, "has_code": true, "code_url": "https://github.com/MozerWang/Loong", "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.5358}
{"id": "aed722fe0ea1f969753db56f4db7da4654fe9f407901bc1537e989fc882e0c18", "sources": ["arxiv", "semantic_scholar"], "title": "USDC: A Dataset of $\\underline{U}$ser $\\underline{S}$tance and $\\underline{D}$ogmatism in Long $\\underline{C}$onversations", "abstract": "Analyzing user opinion changes in long conversation threads is extremely critical for applications like enhanced personalization, market research, political campaigns, customer service, targeted advertising, and content moderation. Unfortunately, previous studies on stance and dogmatism in user conversations have focused on training models using datasets annotated at the post level, treating each post as independent and randomly sampling posts from conversation threads. Hence, first, we build a dataset for studying user opinion fluctuations in 764 long multi-user Reddit conversation threads, called USDC. USDC contains annotations for 2 tasks: i) User Stance classification, which involves labeling a user's stance in a post within a conversation on a five-point scale; ii) User Dogmatism classification, which involves labeling a user's overall opinion in the conversation on a four-point scale. Besides being time-consuming and costly, manual annotations for USDC are challenging because: 1) Conversation threads could be very long, increasing the chances of noisy annotations; and 2) Interpreting instances where a user changes their opinion within a conversation is difficult because often such transitions are subtle and not expressed explicitly. Hence, we leverage majority voting on zero-shot, one-shot, and few-shot annotations from Mistral Large and GPT-4 to automate the annotation process. Human annotations on 200 test conversations achieved inter-annotator agreement scores of 0.49 for stance and 0.50 for dogmatism with these LLM annotations, indicating a reasonable level of consistency between human and LLM annotations. USDC is then used to finetune and instruction-tune multiple deployable small language models like LLaMA, Falcon and Vicuna for the stance and dogmatism classification tasks. We make the code and dataset publicly available [https://github.com/mounikamarreddy/USDC].", "authors": ["Mounika Marreddy", "Subba Reddy Oota", "Venkata Charan Chinni", "Manish Gupta", "Lucie Flek"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-24", "url": "https://arxiv.org/abs/2406.16833", "pdf_url": "https://arxiv.org/pdf/2406.16833v2", "arxiv_id": "2406.16833", "doi": "10.48550/arXiv.2406.16833", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/mounikamarreddy/USDC", "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.0753}
{"id": "b241f863a0abef293cf00769c0a7b8640fbac7a16524aa5b6d547c42f165408d", "sources": ["arxiv", "semantic_scholar"], "title": "Found in the Middle: Calibrating Positional Attention Bias Improves Long Context Utilization", "abstract": "Large language models (LLMs), even when specifically trained to process long input contexts, struggle to capture relevant information located in the middle of their input. This phenomenon has been known as the lost-in-the-middle problem. In this work, we make three contributions. First, we set out to understand the factors that cause this phenomenon. In doing so, we establish a connection between lost-in-the-middle to LLMs' intrinsic attention bias: LLMs exhibit a U-shaped attention bias where the tokens at the beginning and at the end of its input receive higher attention, regardless of their relevance. Second, we mitigate this positional bias through a calibration mechanism, found-in-the-middle, that allows the model to attend to contexts faithfully according to their relevance, even though when they are in the middle. Third, we show found-in-the-middle not only achieves better performance in locating relevant information within a long context, but also eventually leads to improved retrieval-augmented generation (RAG) performance across various tasks, outperforming existing methods by up to 15 percentage points. These findings open up future directions in understanding LLM attention bias and its potential consequences.", "authors": ["Cheng-Yu Hsieh", "Yung-Sung Chuang", "Chun-Liang Li", "Zifeng Wang", "Long T. Le", "Abhishek Kumar", "James Glass", "Alexander Ratner", "Chen-Yu Lee", "Ranjay Krishna", "Tomas Pfister"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-23", "url": "https://arxiv.org/abs/2406.16008", "pdf_url": "https://arxiv.org/pdf/2406.16008v2", "arxiv_id": "2406.16008", "doi": "10.48550/arXiv.2406.16008", "citation_count": 107, "influential_citation_count": 9, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.5084}
{"id": "8afc4a171839dfecb412aac34fb61d5f0779d62c7ddd84feffad19a6ba8e1f1d", "sources": ["arxiv", "semantic_scholar"], "title": "Insights into LLM Long-Context Failures: When Transformers Know but Don't Tell", "abstract": "Large Language Models (LLMs) exhibit positional bias, struggling to utilize information from the middle or end of long contexts. Our study explores LLMs' long-context reasoning by probing their hidden representations. We find that while LLMs encode the position of target information, they often fail to leverage this in generating accurate responses. This reveals a disconnect between information retrieval and utilization, a \"know but don't tell\" phenomenon. We further analyze the relationship between extraction time and final accuracy, offering insights into the underlying mechanics of transformer models.", "authors": ["Taiming Lu", "Muhan Gao", "Kuai Yu", "Adam Byerly", "Daniel Khashabi"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-20", "url": "https://arxiv.org/abs/2406.14673", "pdf_url": "https://arxiv.org/pdf/2406.14673v2", "arxiv_id": "2406.14673", "doi": "10.48550/arXiv.2406.14673", "citation_count": 41, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.4058}
{"id": "b61a94cb10b903d80064b0daa4c568ff2f56c632f80e892500006bdce92b06ac", "sources": ["arxiv", "semantic_scholar"], "title": "DoubleDipper: Improving Long-Context LLMs via Context Recycling", "abstract": "Despite recent advancements in Large Language Models (LLMs), their performance on tasks involving long contexts remains sub-optimal. In this work, we propose DoubleDipper, a novel In-Context-Learning method that automatically generates few-shot examples for long context QA tasks by recycling contexts. Specifically, given a long input context (1-3k tokens) and a query, we generate additional query-output pairs from the given context as few-shot examples, while introducing the context only once. This ensures that the demonstrations are leveraging the same context as the target query while only adding a small number of tokens to the prompt. We further enhance each demonstration by instructing the model to explicitly identify the relevant paragraphs before the answer, which improves performance while providing fine-grained attribution to the answer source. We apply our method on multiple LLMs and obtain substantial improvements (+16 absolute points on average across models) on various QA datasets with long context. Surprisingly, despite introducing only single-hop ICL examples, LLMs successfully generalize to multi-hop long-context QA using our approach.", "authors": ["Arie Cattan", "Alon Jacovi", "Alex Fabrikant", "Jonathan Herzig", "Roee Aharoni", "Hannah Rashkin", "Dror Marcus", "Avinatan Hassidim", "Yossi Matias", "Idan Szpektor", "Avi Caciularu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-19", "url": "https://arxiv.org/abs/2406.13632", "pdf_url": "https://arxiv.org/pdf/2406.13632v4", "arxiv_id": "2406.13632", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.0}
{"id": "24583a039b57fcb7d075d67329ef459d6634a9a31d6638a9fd8bb4b218cf9b95", "sources": ["arxiv", "semantic_scholar"], "title": "Understanding the RoPE Extensions of Long-Context LLMs: An Attention Perspective", "abstract": "Enabling LLMs to handle lengthy context is currently a research hotspot. Most LLMs are built upon rotary position embedding (RoPE), a popular position encoding method. Therefore, a prominent path is to extrapolate the RoPE trained on comparably short texts to far longer texts. A heavy bunch of efforts have been dedicated to boosting the extrapolation via extending the formulations of the RoPE, however, few of them have attempted to showcase their inner workings comprehensively. In this paper, we are driven to offer a straightforward yet in-depth understanding of RoPE extensions from an attention perspective and on two benchmarking tasks. A broad array of experiments reveals several valuable findings: 1) Maintaining attention patterns to those at the pretrained length improves extrapolation; 2) Large attention uncertainty leads to retrieval errors; 3) Using longer continual pretraining lengths for RoPE extensions could reduce attention uncertainty and significantly enhance extrapolation.", "authors": ["Meizhi Zhong", "Chen Zhang", "Yikun Lei", "Xikai Liu", "Yan Gao", "Yao Hu", "Kehai Chen", "Min Zhang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-19", "url": "https://arxiv.org/abs/2406.13282", "pdf_url": "https://arxiv.org/pdf/2406.13282v3", "arxiv_id": "2406.13282", "doi": "10.48550/arXiv.2406.13282", "citation_count": 13, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "International Conference on Computational Linguistics", "quality_score": 0.2865}
{"id": "2b9c8ea845249708088910563938fef3faece30a15fdd6457df55ec7bf2ea21e", "sources": ["arxiv", "semantic_scholar"], "title": "In-Context Learning of Energy Functions", "abstract": "In-context learning is a powerful capability of certain machine learning models that arguably underpins the success of today's frontier AI models. However, in-context learning is critically limited to settings where the in-context distribution of interest $p_θ^{ICL}( x|\\mathcal{D})$ can be straightforwardly expressed and/or parameterized by the model; for instance, language modeling relies on expressing the next-token distribution as a categorical distribution parameterized by the network's output logits. In this work, we present a more general form of in-context learning without such a limitation that we call \\textit{in-context learning of energy functions}. The idea is to instead learn the unconstrained and arbitrary in-context energy function $E_θ^{ICL}(x|\\mathcal{D})$ corresponding to the in-context distribution $p_θ^{ICL}(x|\\mathcal{D})$. To do this, we use classic ideas from energy-based modeling. We provide preliminary evidence that our method empirically works on synthetic data. Interestingly, our work contributes (to the best of our knowledge) the first example of in-context learning where the input space and output space differ from one another, suggesting that in-context learning is a more-general capability than previously realized.", "authors": ["Rylan Schaeffer", "Mikail Khona", "Sanmi Koyejo"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-18", "url": "https://arxiv.org/abs/2406.12785", "pdf_url": "https://arxiv.org/pdf/2406.12785v1", "arxiv_id": "2406.12785", "doi": "10.48550/arXiv.2406.12785", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "437f5aee6ed0907e12b0130d344a5027d3d217643ef25dccfb2f661175fbe7ad", "sources": ["arxiv", "semantic_scholar"], "title": "SampleAttention: Near-Lossless Acceleration of Long Context LLM Inference with Adaptive Structured Sparse Attention", "abstract": "Large language models (LLMs) now support extremely long context windows, but the quadratic complexity of vanilla attention results in significantly long Time-to-First-Token (TTFT) latency. Existing approaches to address this complexity require additional pretraining or finetuning, and often sacrifice model accuracy. In this paper, we first provide both theoretical and empirical foundations for near-lossless sparse attention. We find dynamically capturing head-specific sparse patterns at runtime with low overhead is crucial. To address this, we propose SampleAttention, an adaptive structured and near-lossless sparse attention. Leveraging observed significant sparse patterns, SampleAttention attends to a fixed percentage of adjacent tokens to capture local window patterns, and employs a two-stage query-guided key-value filtering approach, which adaptively select a minimum set of key-values with low overhead, to capture column stripe patterns. Comprehensive evaluations show that SampleAttention can seamlessly replace vanilla attention in off-the-shelf LLMs with nearly no accuracy loss, and reduces TTFT by up to $2.42\\times$ compared with FlashAttention.", "authors": ["Qianchao Zhu", "Jiangfei Duan", "Chang Chen", "Siran Liu", "Guanyu Feng", "Xin Lv", "Xiao Chuanfu", "Dahua Lin", "Chao Yang"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-17", "url": "https://arxiv.org/abs/2406.15486", "pdf_url": "https://arxiv.org/pdf/2406.15486v3", "arxiv_id": "2406.15486", "doi": "10.48550/arXiv.2406.15486", "citation_count": 46, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Conference on Machine Learning and Systems", "quality_score": 0.418}
{"id": "5016474a8dbbae3dab4490b0ce7fbf4c52fc9b6ca72d41435e02458ad93b303f", "sources": ["arxiv", "semantic_scholar"], "title": "Long Code Arena: a Set of Benchmarks for Long-Context Code Models", "abstract": "Nowadays, the fields of code and natural language processing are evolving rapidly. In particular, models become better at processing long context windows - supported context sizes have increased by orders of magnitude over the last few years. However, there is a shortage of benchmarks for code processing that go beyond a single file of context, while the most popular ones are limited to a single method. With this work, we aim to close this gap by introducing Long Code Arena, a suite of six benchmarks for code processing tasks that require project-wide context. These tasks cover different aspects of code processing: library-based code generation, CI builds repair, project-level code completion, commit message generation, bug localization, and module summarization. For each task, we provide a manually verified dataset for testing, an evaluation suite, and open-source baseline solutions based on popular LLMs to showcase the usage of the dataset and to simplify adoption by other researchers. We publish the benchmark page on HuggingFace Spaces with the leaderboard, links to HuggingFace Hub for all the datasets, and link to the GitHub repository with baselines: https://huggingface.co/spaces/JetBrains-Research/long-code-arena.", "authors": ["Egor Bogomolov", "Aleksandra Eliseeva", "Timur Galimzyanov", "Evgeniy Glukhov", "Anton Shapkin", "Maria Tigina", "Yaroslav Golubev", "Alexander Kovrigin", "Arie van Deursen", "Maliheh Izadi", "Timofey Bryksin"], "categories": ["cs.LG", "cs.AI", "cs.IR", "cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-17", "url": "https://arxiv.org/abs/2406.11612", "pdf_url": "https://arxiv.org/pdf/2406.11612v1", "arxiv_id": "2406.11612", "doi": "10.48550/arXiv.2406.11612", "citation_count": 59, "influential_citation_count": 10, "has_code": true, "code_url": null, "venue": "arXiv.org", "quality_score": 0.5207}
{"id": "e7aa596e505630f54de1d94a7d59c6868d4167425fdcd7ff125de3c690fdbb25", "sources": ["arxiv", "semantic_scholar"], "title": "Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference", "abstract": "As the demand for long-context large language models (LLMs) increases, models with context windows of up to 128K or 1M tokens are becoming increasingly prevalent. However, long-context LLM inference is challenging since the inference speed decreases significantly as the sequence length grows. This slowdown is primarily caused by loading a large KV cache during self-attention. Previous works have shown that a small portion of critical tokens will dominate the attention outcomes. However, we observe the criticality of a token highly depends on the query. To this end, we propose Quest, a query-aware KV cache selection algorithm. Quest keeps track of the minimal and maximal Key values in KV cache pages and estimates the criticality of a given page using Query vectors. By only loading the Top-K critical KV cache pages for attention, Quest significantly speeds up self-attention without sacrificing accuracy. We show that Quest can achieve up to 2.23x self-attention speedup, which reduces inference latency by 7.03x while performing well on tasks with long dependencies with negligible accuracy loss. Code is available at http://github.com/mit-han-lab/Quest .", "authors": ["Jiaming Tang", "Yilong Zhao", "Kan Zhu", "Guangxuan Xiao", "Baris Kasikci", "Song Han"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-16", "url": "https://arxiv.org/abs/2406.10774", "pdf_url": "https://arxiv.org/pdf/2406.10774v2", "arxiv_id": "2406.10774", "doi": "10.48550/arXiv.2406.10774", "citation_count": 387, "influential_citation_count": 92, "has_code": true, "code_url": "http://github.com/mit-han-lab/Quest", "venue": "International Conference on Machine Learning", "quality_score": 0.9842}
{"id": "826705ad983090f79f0f88dbbe5847c852d7ef0793ccdce68cbd65588180dc4b", "sources": ["arxiv", "semantic_scholar"], "title": "3D-RPE: Enhancing Long-Context Modeling Through 3D Rotary Position Encoding", "abstract": "Inspired by the Bloch Sphere representation, we propose a novel rotary position encoding on a three-dimensional sphere, named 3D Rotary Position Encoding (3D-RPE). 3D-RPE is an advanced version of the widely used 2D Rotary Position Encoding (RoPE), with two major advantages for modeling long contexts: controllable long-term decay and improved position resolution. For controllable long-term decay, 3D-RPE allows for the regulation of long-term decay within the chunk size, ensuring the modeling of relative positional information between tokens at a distant relative position. For enhanced position resolution, 3D-RPE can mitigate the degradation of position resolution caused by position interpolation on RoPE. We have conducted experiments on long-context Natural Language Understanding (NLU) and long-sequence Language Modeling (LM) tasks. From the experimental results, 3D-RPE achieved performance improvements over RoPE, especially in long-context NLU tasks.", "authors": ["Xindian Ma", "Wenyuan Liu", "Peng Zhang", "Nan Xu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-14", "url": "https://arxiv.org/abs/2406.09897", "pdf_url": "https://arxiv.org/pdf/2406.09897v1", "arxiv_id": "2406.09897", "doi": "10.48550/arXiv.2406.09897", "citation_count": 16, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "AAAI Conference on Artificial Intelligence", "quality_score": 0.3076}
{"id": "78a0653f9c4ff284a9fa878e939d7c7d8965c1ccc57542d9da4b13f12c499f8e", "sources": ["arxiv", "semantic_scholar"], "title": "BABILong: Testing the Limits of LLMs with Long Context Reasoning-in-a-Haystack", "abstract": "In recent years, the input context sizes of large language models (LLMs) have increased dramatically. However, existing evaluation methods have not kept pace, failing to comprehensively assess the efficiency of models in handling long contexts. To bridge this gap, we introduce the BABILong benchmark, designed to test language models' ability to reason across facts distributed in extremely long documents. BABILong includes a diverse set of 20 reasoning tasks, including fact chaining, simple induction, deduction, counting, and handling lists/sets. These tasks are challenging on their own, and even more demanding when the required facts are scattered across long natural text. Our evaluations show that popular LLMs effectively utilize only 10-20\\% of the context and their performance declines sharply with increased reasoning complexity. Among alternatives to in-context reasoning, Retrieval-Augmented Generation methods achieve a modest 60\\% accuracy on single-fact question answering, independent of context length. Among context extension methods, the highest performance is demonstrated by recurrent memory transformers after fine-tuning, enabling the processing of lengths up to 50 million tokens. The BABILong benchmark is extendable to any length to support the evaluation of new upcoming models with increased capabilities, and we provide splits up to 10 million token lengths.", "authors": ["Yuri Kuratov", "Aydar Bulatov", "Petr Anokhin", "Ivan Rodkin", "Dmitry Sorokin", "Artyom Sorokin", "Mikhail Burtsev"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-14", "url": "https://arxiv.org/abs/2406.10149", "pdf_url": "https://arxiv.org/pdf/2406.10149v2", "arxiv_id": "2406.10149", "doi": "10.48550/arXiv.2406.10149", "citation_count": 224, "influential_citation_count": 35, "has_code": false, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.7782}
{"id": "3446a9f75e519a9eff5dc0aeba0eb345e0f8bb2c921d6564879b4f70bf05ff9a", "sources": ["arxiv", "semantic_scholar"], "title": "An Efficient Recipe for Long Context Extension via Middle-Focused Positional Encoding", "abstract": "Recently, many methods have been developed to extend the context length of pre-trained large language models (LLMs), but they often require fine-tuning at the target length ($\\gg4K$) and struggle to effectively utilize information from the middle part of the context. To address these issues, we propose $\\textbf{C}$ontinuity-$\\textbf{R}$elativity ind$\\textbf{E}$xing with g$\\textbf{A}$ussian $\\textbf{M}$iddle ($\\texttt{CREAM}$), which interpolates positional encodings by manipulating position indices. Apart from being simple, $\\texttt{CREAM}$ is training-efficient: it only requires fine-tuning at the pre-trained context window (e.g., Llama 2-4K) and can extend LLMs to a much longer target context length (e.g., 256K). To ensure that the model focuses more on the information in the middle, we introduce a truncated Gaussian to encourage sampling from the middle part of the context during fine-tuning, thus alleviating the \"Lost-in-the-Middle\" problem faced by long-context LLMs. Experimental results show that $\\texttt{CREAM}$ successfully extends LLMs to the target length for both Base and Chat versions of $\\texttt{Llama2-7B}$ with \"Never Miss A Beat\". Our code is publicly available at https://github.com/bigai-nlco/cream.", "authors": ["Tong Wu", "Yanpeng Zhao", "Zilong Zheng"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-11", "url": "https://arxiv.org/abs/2406.07138", "pdf_url": "https://arxiv.org/pdf/2406.07138v2", "arxiv_id": "2406.07138", "doi": "10.52202/079017-1794", "citation_count": 13, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/bigai-nlco/cream", "venue": "Neural Information Processing Systems", "quality_score": 0.2865}
{"id": "50949668e4187377222dfcc34ec04d6678b3eccbc31025818e57396765a813b3", "sources": ["arxiv", "semantic_scholar"], "title": "Recurrent Context Compression: Efficiently Expanding the Context Window of LLM", "abstract": "To extend the context length of Transformer-based large language models (LLMs) and improve comprehension capabilities, we often face limitations due to computational resources and bounded memory storage capacity. This work introduces a method called Recurrent Context Compression (RCC), designed to efficiently expand the context window length of LLMs within constrained storage space. We also investigate the issue of poor model responses when both instructions and context are compressed in downstream tasks, and propose an instruction reconstruction method to mitigate this problem. We validated the effectiveness of our approach on multiple tasks, achieving a compression rate of up to 32x on text reconstruction tasks with a BLEU4 score close to 0.95, and nearly 100\\% accuracy on a passkey retrieval task with a sequence length of 1M. Finally, our method demonstrated competitive performance in long-text question-answering tasks compared to non-compressed methods, while significantly saving storage resources in long-text inference tasks. Our code, models, and demo are available at https://github.com/WUHU-G/RCC_Transformer", "authors": ["Chensen Huang", "Guibo Zhu", "Xuepeng Wang", "Yifei Luo", "Guojing Ge", "Haoran Chen", "Dong Yi", "Jinqiao Wang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-10", "url": "https://arxiv.org/abs/2406.06110", "pdf_url": "https://arxiv.org/pdf/2406.06110v1", "arxiv_id": "2406.06110", "doi": "10.48550/arXiv.2406.06110", "citation_count": 7, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/WUHU-G/RCC_Transformer", "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "a7f27fa57783edf815415ba663ef65b8d38ca9b3138430b8b31b0205e11717c8", "sources": ["arxiv", "semantic_scholar"], "title": "Evaluating Zero-Shot Long-Context LLM Compression", "abstract": "This study evaluates the effectiveness of zero-shot compression techniques on large language models (LLMs) under long-context. We identify the tendency for computational errors to increase under long-context when employing certain compression methods. We propose a hypothesis to explain the varied behavior of different LLM compression techniques and explore remedies to mitigate the performance decline observed in some techniques under long-context. This is a course report for COS 598D Machine Learning and Systems by Prof. Kai Li at Princeton University. Due to limited computational resources, our experiments were conducted only on LLaMA-2-7B-32K.", "authors": ["Chenyu Wang", "Yihan Wang", "Kai Li"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-10", "url": "https://arxiv.org/abs/2406.06773", "pdf_url": "https://arxiv.org/pdf/2406.06773v2", "arxiv_id": "2406.06773", "doi": "10.48550/arXiv.2406.06773", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0}
{"id": "2aa0c7bbf1a2ba0269536e8133fd60c6862891dc049e6a704471051f67706e2c", "sources": ["arxiv", "semantic_scholar"], "title": "RepoQA: Evaluating Long Context Code Understanding", "abstract": "Recent advances have been improving the context windows of Large Language Models (LLMs). To quantify the real long-context capabilities of LLMs, evaluators such as the popular Needle in a Haystack have been developed to test LLMs over a large chunk of raw texts. While effective, current evaluations overlook the insight of how LLMs work with long-context code, i.e., repositories. To this end, we initiate the RepoQA benchmark to evaluate LLMs on long-context code understanding. Traditional needle testers ask LLMs to directly retrieve the answer from the context without necessary deep understanding. In RepoQA, we built our initial task, namely Searching Needle Function (SNF), which exercises LLMs to search functions given their natural-language description, i.e., LLMs cannot find the desired function if they cannot understand the description and code. RepoQA is multilingual and comprehensive: it includes 500 code search tasks gathered from 50 popular repositories across 5 modern programming languages. By evaluating 26 general and code-specific LLMs on RepoQA, we show (i) there is still a small gap between the best open and proprietary models; (ii) different models are good at different languages; and (iii) models may understand code better without comments.", "authors": ["Jiawei Liu", "Jia Le Tian", "Vijay Daita", "Yuxiang Wei", "Yifeng Ding", "Yuhan Katherine Wang", "Jun Yang", "Lingming Zhang"], "categories": ["cs.SE", "cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-10", "url": "https://arxiv.org/abs/2406.06025", "pdf_url": "https://arxiv.org/pdf/2406.06025v1", "arxiv_id": "2406.06025", "doi": "10.48550/arXiv.2406.06025", "citation_count": 39, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4005}
{"id": "f892474e98d61c6590a1378189f18388297b8a0e6b7d7f3b54148711069df4c8", "sources": ["arxiv", "semantic_scholar"], "title": "LoCoCo: Dropping In Convolutions for Long Context Compression", "abstract": "This paper tackles the memory hurdle of processing long context sequences in Large Language Models (LLMs), by presenting a novel approach, Dropping In Convolutions for Long Context Compression (LoCoCo). LoCoCo employs only a fixed-size Key-Value (KV) cache, and can enhance efficiency in both inference and fine-tuning stages. Diverging from prior methods that selectively drop KV pairs based on heuristics, LoCoCo leverages a data-driven adaptive fusion technique, blending previous KV pairs with incoming tokens to minimize the loss of contextual information and ensure accurate attention modeling. This token integration is achieved through injecting one-dimensional convolutional kernels that dynamically calculate mixing weights for each KV cache slot. Designed for broad compatibility with existing LLM frameworks, LoCoCo allows for straightforward \"drop-in\" integration without needing architectural modifications, while incurring minimal tuning overhead. Experiments demonstrate that LoCoCo maintains consistently outstanding performance across various context lengths and can achieve a high context compression rate during both inference and fine-tuning phases. During inference, we successfully compressed up to 3482 tokens into a 128-size KV cache, while retaining comparable performance to the full sequence - an accuracy improvement of up to 0.2791 compared to baselines at the same cache size. During post-training tuning, we also effectively extended the context length from 4K to 32K using a KV cache of fixed size 512, achieving performance similar to fine-tuning with entire sequences.", "authors": ["Ruisi Cai", "Yuandong Tian", "Zhangyang Wang", "Beidi Chen"], "categories": ["cs.LG", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-08", "url": "https://arxiv.org/abs/2406.05317", "pdf_url": "https://arxiv.org/pdf/2406.05317v2", "arxiv_id": "2406.05317", "doi": "10.48550/arXiv.2406.05317", "citation_count": 19, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.3495}
{"id": "d200589fd3b29438d6857da528f233790c7cfac0fdaac3cbb12fe9ccd071e8a0", "sources": ["arxiv", "semantic_scholar"], "title": "Chain of Agents: Large Language Models Collaborating on Long-Context Tasks", "abstract": "Addressing the challenge of effectively processing long contexts has become a critical issue for Large Language Models (LLMs). Two common strategies have emerged: 1) reducing the input length, such as retrieving relevant chunks by Retrieval-Augmented Generation (RAG), and 2) expanding the context window limit of LLMs. However, both strategies have drawbacks: input reduction has no guarantee of covering the part with needed information, while window extension struggles with focusing on the pertinent information for solving the task. To mitigate these limitations, we propose Chain-of-Agents (CoA), a novel framework that harnesses multi-agent collaboration through natural language to enable information aggregation and context reasoning across various LLMs over long-context tasks. CoA consists of multiple worker agents who sequentially communicate to handle different segmented portions of the text, followed by a manager agent who synthesizes these contributions into a coherent final output. CoA processes the entire input by interleaving reading and reasoning, and it mitigates long context focus issues by assigning each agent a short context. We perform comprehensive evaluation of CoA on a wide range of long-context tasks in question answering, summarization, and code completion, demonstrating significant improvements by up to 10% over strong baselines of RAG, Full-Context, and multi-agent LLMs.", "authors": ["Yusen Zhang", "Ruoxi Sun", "Yanfei Chen", "Tomas Pfister", "Rui Zhang", "Sercan Ö. Arik"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-04", "url": "https://arxiv.org/abs/2406.02818", "pdf_url": "https://arxiv.org/pdf/2406.02818v1", "arxiv_id": "2406.02818", "doi": "10.48550/arXiv.2406.02818", "citation_count": 243, "influential_citation_count": 16, "has_code": false, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.6152}
{"id": "19bae119a30be01b60adfc91e684978f36a1fe83584f93b5c2aae371a2a00007", "sources": ["arxiv", "semantic_scholar"], "title": "TabMDA: Tabular Manifold Data Augmentation for Any Classifier using Transformers with In-context Subsetting", "abstract": "Tabular data is prevalent in many critical domains, yet it is often challenging to acquire in large quantities. This scarcity usually results in poor performance of machine learning models on such data. Data augmentation, a common strategy for performance improvement in vision and language tasks, typically underperforms for tabular data due to the lack of explicit symmetries in the input space. To overcome this challenge, we introduce TabMDA, a novel method for manifold data augmentation on tabular data. This method utilises a pre-trained in-context model, such as TabPFN, to map the data into an embedding space. TabMDA performs label-invariant transformations by encoding the data multiple times with varied contexts. This process explores the learned embedding space of the underlying in-context models, thereby enlarging the training dataset. TabMDA is a training-free method, making it applicable to any classifier. We evaluate TabMDA on five standard classifiers and observe significant performance improvements across various tabular datasets. Our results demonstrate that TabMDA provides an effective way to leverage information from pre-trained in-context models to enhance the performance of downstream classifiers. Code is available at https://github.com/AdrianBZG/TabMDA.", "authors": ["Andrei Margeloiu", "Adrián Bazaga", "Nikola Simidjievski", "Pietro Liò", "Mateja Jamnik"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-06-03", "url": "https://arxiv.org/abs/2406.01805", "pdf_url": "https://arxiv.org/pdf/2406.01805v2", "arxiv_id": "2406.01805", "doi": "10.48550/arXiv.2406.01805", "citation_count": 8, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/AdrianBZG/TabMDA", "venue": "arXiv.org", "quality_score": 0.2386}
{"id": "d5c989903aac32704fa10e0cd827475bebab5860a35e526256aaef1b649de43b", "sources": ["arxiv", "semantic_scholar"], "title": "Is In-Context Learning Sufficient for Instruction Following in LLMs?", "abstract": "In-context learning (ICL) allows LLMs to learn from examples without changing their weights: this is a particularly promising capability for long-context LLMs that can potentially learn from many examples. Recently, Lin et al. (2024) proposed URIAL, a method using only three in-context examples to align base LLMs, achieving non-trivial instruction following performance. In this work, we show that, while effective, ICL alignment with URIAL still underperforms compared to instruction fine-tuning on the established benchmark MT-Bench, especially with more capable base LLMs. We then uncover the most relevant elements for successful in-context alignment, finding the crucial role of the decoding parameters. Based on these insights, we show that the approach of URIAL can indeed be improved by adding high-quality, potentially carefully selected via greedy search, demonstrations in context, getting closer to the performance of instruct models. Finally, we provide the first, to our knowledge, systematic comparison of ICL and instruction fine-tuning (IFT) for instruction following in the low data regime, where ICL can be a viable alternative to IFT. Overall, our work advances the understanding of ICL as an alignment technique and its relationship to IFT. We provide our code at https://github.com/tml-epfl/icl-alignment.", "authors": ["Hao Zhao", "Maksym Andriushchenko", "Francesco Croce", "Nicolas Flammarion"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-30", "url": "https://arxiv.org/abs/2405.19874", "pdf_url": "https://arxiv.org/pdf/2405.19874v3", "arxiv_id": "2405.19874", "doi": "10.48550/arXiv.2405.19874", "citation_count": 23, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/tml-epfl/icl-alignment", "venue": "International Conference on Learning Representations", "quality_score": 0.3451}
{"id": "02e54322416f41c377347af72cb513b1e77cf3cb0f4beddccf810a13d6917b5e", "sources": ["arxiv", "semantic_scholar"], "title": "Exploring Context Window of Large Language Models via Decomposed Positional Vectors", "abstract": "Transformer-based large language models (LLMs) typically have a limited context window, resulting in significant performance degradation when processing text beyond the length of the context window. Extensive studies have been proposed to extend the context window and achieve length extrapolation of LLMs, but there is still a lack of in-depth interpretation of these approaches. In this study, we explore the positional information within and beyond the context window for deciphering the underlying mechanism of LLMs. By using a mean-based decomposition method, we disentangle positional vectors from hidden states of LLMs and analyze their formation and effect on attention. Furthermore, when texts exceed the context window, we analyze the change of positional vectors in two settings, i.e., direct extrapolation and context window extension. Based on our findings, we design two training-free context window extension methods, positional vector replacement and attention window extension. Experimental results show that our methods can effectively extend the context window length.", "authors": ["Zican Dong", "Junyi Li", "Xin Men", "Wayne Xin Zhao", "Bingbing Wang", "Zhen Tian", "Weipeng Chen", "Ji-Rong Wen"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-28", "url": "https://arxiv.org/abs/2405.18009", "pdf_url": "https://arxiv.org/pdf/2405.18009v2", "arxiv_id": "2405.18009", "doi": "10.48550/arXiv.2405.18009", "citation_count": 31, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.3763}
{"id": "b4cc7d63cb871c8ac426bb26b827967fdae13cf18286b343f80f2906e5539b07", "sources": ["arxiv", "semantic_scholar"], "title": "Long Context is Not Long at All: A Prospector of Long-Dependency Data for Large Language Models", "abstract": "Long-context modeling capabilities are important for large language models (LLMs) in various applications. However, directly training LLMs with long context windows is insufficient to enhance this capability since some training samples do not exhibit strong semantic dependencies across long contexts. In this study, we propose a data mining framework \\textbf{ProLong} that can assign each training sample with a long dependency score, which can be used to rank and filter samples that are more advantageous for enhancing long-context modeling abilities in LLM training. Specifically, we first use delta perplexity scores to measure the \\textit{Dependency Strength} between text segments in a given document. Then we refine this metric based on the \\textit{Dependency Distance} of these segments to incorporate spatial relationships across long-contexts. Final results are calibrated with a \\textit{Dependency Specificity} metric to prevent trivial dependencies introduced by repetitive patterns. Moreover, a random sampling approach is proposed to optimize the computational efficiency of ProLong. Comprehensive experiments on multiple benchmarks indicate that ProLong effectively identifies documents that carry long dependencies and LLMs trained on these documents exhibit significantly enhanced long-context modeling capabilities.", "authors": ["Longze Chen", "Ziqiang Liu", "Wanwei He", "Yunshui Li", "Run Luo", "Min Yang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-28", "url": "https://arxiv.org/abs/2405.17915", "pdf_url": "https://arxiv.org/pdf/2405.17915v1", "arxiv_id": "2405.17915", "doi": "10.48550/arXiv.2405.17915", "citation_count": 19, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.3253}
{"id": "57abae3761ea6336534a2ad217ee0b73e5710766825dc2d8fcfe0f7275f0bbec", "sources": ["arxiv", "semantic_scholar"], "title": "Are Long-LLMs A Necessity For Long-Context Tasks?", "abstract": "The learning and deployment of long-LLMs remains a challenging problem despite recent progresses. In this work, we argue that the long-LLMs are not a necessity to solve long-context tasks, as common long-context tasks are short-context solvable, i.e. they can be solved by purely working with oracle short-contexts within the long-context tasks' inputs. On top of this argument, we propose a framework called LC-Boost (Long-Context Bootstrapper), which enables a short-LLM to address the long-context tasks in a bootstrapping manner. In our framework, the short-LLM prompts itself to reason for two critical decisions: 1) how to access to the appropriate part of context within the input, 2) how to make effective use of the accessed context. By adaptively accessing and utilizing the context based on the presented tasks, LC-Boost can serve as a general framework to handle diversified long-context processing problems. We comprehensively evaluate different types of tasks from popular long-context benchmarks, where LC-Boost is able to achieve a substantially improved performance with a much smaller consumption of resource.", "authors": ["Hongjin Qian", "Zheng Liu", "Peitian Zhang", "Kelong Mao", "Yujia Zhou", "Xu Chen", "Zhicheng Dou"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-24", "url": "https://arxiv.org/abs/2405.15318", "pdf_url": "https://arxiv.org/pdf/2405.15318v1", "arxiv_id": "2405.15318", "doi": "10.48550/arXiv.2405.15318", "citation_count": 28, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3656}
{"id": "c2bda5cc7a7872bfbd51f188ea29165bfe336bbd3fa4405a0e521633ccfe7958", "sources": ["arxiv", "semantic_scholar"], "title": "Does context matter in digital pathology?", "abstract": "The development of Artificial Intelligence for healthcare is of great importance. Models can sometimes achieve even superior performance to human experts, however, they can reason based on spurious features. This is not acceptable to the experts as it is expected that the models catch the valid patterns in the data following domain expertise. In the work, we analyse whether Deep Learning (DL) models for vision follow the histopathologists' practice so that when diagnosing a part of a lesion, they take into account also the surrounding tissues which serve as context. It turns out that the performance of DL models significantly decreases when the amount of contextual information is limited, therefore contextual information is valuable at prediction time. Moreover, we show that the models sometimes behave in an unstable way as for some images, they change the predictions many times depending on the size of the context. It may suggest that partial contextual information can be misleading.", "authors": ["Paulina Tomaszewska", "Mateusz Sperkowski", "Przemysław Biecek"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-23", "url": "https://arxiv.org/abs/2405.14301", "pdf_url": "https://arxiv.org/pdf/2405.14301v1", "arxiv_id": "2405.14301", "doi": "10.48550/arXiv.2405.14301", "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0}
{"id": "ba894529bdb589b7db1cd21913e7a957411bd51e8bd74036b3e6cac6ed284a78", "sources": ["arxiv", "semantic_scholar"], "title": "Long Context Alignment with Short Instructions and Synthesized Positions", "abstract": "Effectively handling instructions with extremely long context remains a challenge for Large Language Models (LLMs), typically necessitating high-quality long data and substantial computational resources. This paper introduces Step-Skipping Alignment (SkipAlign), a new technique designed to enhance the long-context capabilities of LLMs in the phase of alignment without the need for additional efforts beyond training with original data length. SkipAlign is developed on the premise that long-range dependencies are fundamental to enhancing an LLM's capacity of long context. Departing from merely expanding the length of input samples, SkipAlign synthesizes long-range dependencies from the aspect of positions indices. This is achieved by the strategic insertion of skipped positions within instruction-following samples, which utilizes the semantic structure of the data to effectively expand the context. Through extensive experiments on base models with a variety of context window sizes, SkipAlign demonstrates its effectiveness across a spectrum of long-context tasks. Particularly noteworthy is that with a careful selection of the base model and alignment datasets, SkipAlign with only 6B parameters achieves it's best performance and comparable with strong baselines like GPT-3.5-Turbo-16K on LongBench.", "authors": ["Wenhao Wu", "Yizhong Wang", "Yao Fu", "Xiang Yue", "Dawei Zhu", "Sujian Li"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-05-07", "url": "https://arxiv.org/abs/2405.03939", "pdf_url": "https://arxiv.org/pdf/2405.03939v1", "arxiv_id": "2405.03939", "doi": "10.48550/arXiv.2405.03939", "citation_count": 25, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3537}
{"id": "5f5236dfd142a93a1e7d32577b28a9d8e98825bc0dd54470b52e9269925fee46", "sources": ["arxiv", "semantic_scholar"], "title": "In-Context Learning with Long-Context Models: An In-Depth Exploration", "abstract": "As model context lengths continue to increase, the number of demonstrations that can be provided in-context approaches the size of entire training datasets. We study the behavior of in-context learning (ICL) at this extreme scale on multiple datasets and models. We show that, for many datasets with large label spaces, performance continues to increase with thousands of demonstrations. We contrast this with example retrieval and finetuning: example retrieval shows excellent performance at low context lengths but has diminished gains with more demonstrations; finetuning is more data hungry than ICL but can exceed long-context ICL performance with additional data. We use the ICL setting to study several properties of both in-context learning and long-context models. We show that long-context ICL is less sensitive to random input shuffling than short-context ICL, that grouping of same-label examples negatively impacts performance, and that the performance boosts do not arise from cumulative gain from encoding many examples together. We conclude that long-context ICL can be an effective tool, and may not require long-context for encoding the demonstration set at all.", "authors": ["Amanda Bertsch", "Maor Ivgi", "Emily Xiao", "Uri Alon", "Jonathan Berant", "Matthew R. Gormley", "Graham Neubig"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-30", "url": "https://arxiv.org/abs/2405.00200", "pdf_url": "https://arxiv.org/pdf/2405.00200v2", "arxiv_id": "2405.00200", "doi": "10.48550/arXiv.2405.00200", "citation_count": 147, "influential_citation_count": 15, "has_code": false, "code_url": null, "venue": "North American Chapter of the Association for Computational Linguistics", "quality_score": 0.6021}
{"id": "ef8a2df9aa9c80ee4554ae4deba7dbf3993646741a4d8ad5df9b35841053467f", "sources": ["arxiv", "semantic_scholar"], "title": "Make Your LLM Fully Utilize the Context", "abstract": "While many contemporary large language models (LLMs) can process lengthy input, they still struggle to fully utilize information within the long context, known as the lost-in-the-middle challenge. We hypothesize that it stems from insufficient explicit supervision during the long-context training, which fails to emphasize that any position in a long context can hold crucial information. Based on this intuition, our study presents information-intensive (IN2) training, a purely data-driven solution to overcome lost-in-the-middle. Specifically, IN2 training leverages a synthesized long-context question-answer dataset, where the answer requires (1) fine-grained information awareness on a short segment (~128 tokens) within a synthesized long context (4K-32K tokens), and (2) the integration and reasoning of information from two or more short segments. Through applying this information-intensive training on Mistral-7B, we present FILM-7B (FILl-in-the-Middle). To thoroughly assess the ability of FILM-7B for utilizing long contexts, we design three probing tasks that encompass various context styles (document, code, and structured-data context) and information retrieval patterns (forward, backward, and bi-directional retrieval). The probing results demonstrate that FILM-7B can robustly retrieve information from different positions in its 32K context window. Beyond these probing tasks, FILM-7B significantly improves the performance on real-world long-context tasks (e.g., 23.5->26.9 F1 score on NarrativeQA), while maintaining a comparable performance on short-context tasks (e.g., 59.3->59.2 accuracy on MMLU). Github Link: https://github.com/microsoft/FILM.", "authors": ["Shengnan An", "Zexiong Ma", "Zeqi Lin", "Nanning Zheng", "Jian-Guang Lou"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-25", "url": "https://arxiv.org/abs/2404.16811", "pdf_url": "https://arxiv.org/pdf/2404.16811v2", "arxiv_id": "2404.16811", "doi": "10.48550/arXiv.2404.16811", "citation_count": 186, "influential_citation_count": 9, "has_code": true, "code_url": "https://github.com/microsoft/FILM", "venue": "Neural Information Processing Systems", "quality_score": 0.568}
{"id": "dd0f8f0f09af097e8ab266fd69955e08f091ded26de69dda4c717a44f2b5d655", "sources": ["arxiv", "semantic_scholar"], "title": "LongEmbed: Extending Embedding Models for Long Context Retrieval", "abstract": "Embedding models play a pivot role in modern NLP applications such as IR and RAG. While the context limit of LLMs has been pushed beyond 1 million tokens, embedding models are still confined to a narrow context window not exceeding 8k tokens, refrained from application scenarios requiring long inputs such as legal contracts. This paper explores context window extension of existing embedding models, pushing the limit to 32k without requiring additional training. First, we examine the performance of current embedding models for long context retrieval on our newly constructed LongEmbed benchmark. LongEmbed comprises two synthetic tasks and four carefully chosen real-world tasks, featuring documents of varying length and dispersed target information. Benchmarking results underscore huge room for improvement in these models. Based on this, comprehensive experiments show that training-free context window extension strategies like position interpolation can effectively extend the context window of existing embedding models by several folds, regardless of their original context being 512 or beyond 4k. Furthermore, for models employing absolute position encoding (APE), we show the possibility of further fine-tuning to harvest notable performance gains while strictly preserving original behavior for short inputs. For models using rotary position embedding (RoPE), significant enhancements are observed when employing RoPE-specific methods, such as NTK and SelfExtend, indicating RoPE's superiority over APE for context window extension. To facilitate future research, we release E5-Base-4k and E5-RoPE-Base, along with the LongEmbed benchmark.", "authors": ["Dawei Zhu", "Liang Wang", "Nan Yang", "Yifan Song", "Wenhao Wu", "Furu Wei", "Sujian Li"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-18", "url": "https://arxiv.org/abs/2404.12096", "pdf_url": "https://arxiv.org/pdf/2404.12096v3", "arxiv_id": "2404.12096", "doi": "10.48550/arXiv.2404.12096", "citation_count": 57, "influential_citation_count": 7, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.4515}
{"id": "454c4b913ae07cda97e3405b00c59ccf17e55c92ba3903d0cac241a4e0e86ef9", "sources": ["arxiv", "semantic_scholar"], "title": "Characterizing LLM Abstention Behavior in Science QA with Context Perturbations", "abstract": "The correct model response in the face of uncertainty is to abstain from answering a question so as not to mislead the user. In this work, we study the ability of LLMs to abstain from answering context-dependent science questions when provided insufficient or incorrect context. We probe model sensitivity in several settings: removing gold context, replacing gold context with irrelevant context, and providing additional context beyond what is given. In experiments on four QA datasets with six LLMs, we show that performance varies greatly across models, across the type of context provided, and also by question type; in particular, many LLMs seem unable to abstain from answering boolean questions using standard QA prompts. Our analysis also highlights the unexpected impact of abstention performance on QA task accuracy. Counter-intuitively, in some settings, replacing gold context with irrelevant context or adding irrelevant context to gold context can improve abstention performance in a way that results in improvements in task performance. Our results imply that changes are needed in QA dataset design and evaluation to more effectively assess the correctness and downstream impacts of model abstention.", "authors": ["Bingbing Wen", "Bill Howe", "Lucy Lu Wang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-18", "url": "https://arxiv.org/abs/2404.12452", "pdf_url": "https://arxiv.org/pdf/2404.12452v2", "arxiv_id": "2404.12452", "doi": "10.48550/arXiv.2404.12452", "citation_count": 31, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.3763}
{"id": "3c65c0dbffe5291bf12696f8ee147d21b2a7a195518e4c1f681e10441783dcd6", "sources": ["arxiv", "semantic_scholar"], "title": "Hierarchical Context Merging: Better Long Context Understanding for Pre-trained LLMs", "abstract": "Large language models (LLMs) have shown remarkable performance in various natural language processing tasks. However, a primary constraint they face is the context limit, i.e., the maximum number of tokens they can process. Previous works have explored architectural changes and modifications in positional encoding to relax the constraint, but they often require expensive training or do not address the computational demands of self-attention. In this paper, we present Hierarchical cOntext MERging (HOMER), a new training-free scheme designed to overcome the limitations. HOMER uses a divide-and-conquer algorithm, dividing long inputs into manageable chunks. Each chunk is then processed collectively, employing a hierarchical strategy that merges adjacent chunks at progressive transformer layers. A token reduction technique precedes each merging, ensuring memory usage efficiency. We also propose an optimized computational order reducing the memory requirement to logarithmically scale with respect to input length, making it especially favorable for environments with tight memory restrictions. Our experiments demonstrate the proposed method's superior performance and memory efficiency, enabling the broader use of LLMs in contexts requiring extended context. Code is available at https://github.com/alinlab/HOMER.", "authors": ["Woomin Song", "Seunghyuk Oh", "Sangwoo Mo", "Jaehyung Kim", "Sukmin Yun", "Jung-Woo Ha", "Jinwoo Shin"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-16", "url": "https://arxiv.org/abs/2404.10308", "pdf_url": "https://arxiv.org/pdf/2404.10308v1", "arxiv_id": "2404.10308", "doi": "10.48550/arXiv.2404.10308", "citation_count": 40, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/alinlab/HOMER", "venue": "International Conference on Learning Representations", "quality_score": 0.4032}
{"id": "b7ccf8dbcc73b7241661400ba37c66b4ccab54223fdf9380593ec5f13c922b98", "sources": ["arxiv", "semantic_scholar"], "title": "Deferred NAM: Low-latency Top-K Context Injection via Deferred Context Encoding for Non-Streaming ASR", "abstract": "Contextual biasing enables speech recognizers to transcribe important phrases in the speaker's context, such as contact names, even if they are rare in, or absent from, the training data. Attention-based biasing is a leading approach which allows for full end-to-end cotraining of the recognizer and biasing system and requires no separate inference-time components. Such biasers typically consist of a context encoder; followed by a context filter which narrows down the context to apply, improving per-step inference time; and, finally, context application via cross attention. Though much work has gone into optimizing per-frame performance, the context encoder is at least as important: recognition cannot begin before context encoding ends. Here, we show the lightweight phrase selection pass can be moved before context encoding, resulting in a speedup of up to 16.1 times and enabling biasing to scale to 20K phrases with a maximum pre-decoding delay under 33ms. With the addition of phrase- and wordpiece-level cross-entropy losses, our technique also achieves up to a 37.5% relative WER reduction over the baseline without the losses and lightweight phrase selection pass.", "authors": ["Zelin Wu", "Gan Song", "Christopher Li", "Pat Rondon", "Zhong Meng", "Xavier Velez", "Weiran Wang", "Diamantino Caseiro", "Golan Pundak", "Tsendsuren Munkhdalai", "Angad Chandorkar", "Rohit Prabhavalkar"], "categories": ["cs.CL", "cs.AI", "cs.LG", "cs.NE", "eess.AS"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2024-04-15", "url": "https://arxiv.org/abs/2404.10180", "pdf_url": "https://arxiv.org/pdf/2404.10180v2", "arxiv_id": "2404.10180", "doi": "10.48550/arXiv.2404.10180", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "North American Chapter of the Association for Computational Linguistics", "quality_score": 0.2113}
{"id": "393a24fd57ed96ee60d5a1569c830ec0b52847128e8a3680cb2318690c5aaa93", "sources": ["arxiv", "semantic_scholar"], "title": "LLoCO: Learning Long Contexts Offline", "abstract": "Processing long contexts remains a challenge for large language models (LLMs) due to the quadratic computational and memory overhead of the self-attention mechanism and the substantial KV cache sizes during generation. We propose LLoCO, a novel approach to address this problem by learning contexts offline through context compression and in-domain parameter-efficient finetuning with LoRA. Our method enables an LLM to create a concise representation of the original context and efficiently retrieve relevant information to answer questions accurately. Our approach extends the effective context window of a 4k token LLaMA2-7B model to handle up to 128k tokens. We evaluate our approach on several long-context question-answering datasets, demonstrating that LLoCO significantly outperforms in-context learning while using $30\\times$ fewer tokens during inference. LLoCO achieves up to $7.62\\times$ speed-up during inference and $11.52\\times$ higher throughput during finetuning, substantially reduces the cost of long document question answering. This makes it a promising solution for efficient long context processing. Our code is publicly available on https://github.com/jeffreysijuntan/lloco.", "authors": ["Sijun Tan", "Xiuyu Li", "Shishir Patil", "Ziyang Wu", "Tianjun Zhang", "Kurt Keutzer", "Joseph E. Gonzalez", "Raluca Ada Popa"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-11", "url": "https://arxiv.org/abs/2404.07979", "pdf_url": "https://arxiv.org/pdf/2404.07979v2", "arxiv_id": "2404.07979", "doi": "10.48550/arXiv.2404.07979", "citation_count": 17, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/jeffreysijuntan/lloco", "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.3138}
{"id": "d5d2c44cbaed4559247c17d12f6cbf4a11a6438171821e104cea4959502f602b", "sources": ["arxiv", "semantic_scholar"], "title": "Ada-LEval: Evaluating long-context LLMs with length-adaptable benchmarks", "abstract": "Recently, the large language model (LLM) community has shown increasing interest in enhancing LLMs' capability to handle extremely long documents. As various long-text techniques and model architectures emerge, the precise and detailed evaluation of models' long-text capabilities has become increasingly important. Existing long-text evaluation benchmarks, such as L-Eval and LongBench, construct long-text test sets based on open-source datasets, focusing mainly on QA and summarization tasks. These datasets include test samples of varying lengths (from 2k to 32k+) entangled together, making it challenging to assess model capabilities across different length ranges. Moreover, they do not cover the ultralong settings (100k+ tokens) that the latest LLMs claim to achieve. In this paper, we introduce Ada-LEval, a length-adaptable benchmark for evaluating the long-context understanding of LLMs. Ada-LEval includes two challenging subsets, TSort and BestAnswer, which enable a more reliable evaluation of LLMs' long context capabilities. These benchmarks support intricate manipulation of the length of test cases, and can easily produce text samples up to 128k tokens. We evaluate 4 state-of-the-art closed-source API models and 6 open-source models with Ada-LEval. The evaluation results demonstrate the limitations of current LLMs, especially in ultra-long-context settings. Our code is available at https://github.com/open-compass/Ada-LEval.", "authors": ["Chonghua Wang", "Haodong Duan", "Songyang Zhang", "Dahua Lin", "Kai Chen"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-09", "url": "https://arxiv.org/abs/2404.06480", "pdf_url": "https://arxiv.org/pdf/2404.06480v2", "arxiv_id": "2404.06480", "doi": "10.48550/arXiv.2404.06480", "citation_count": 45, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/open-compass/Ada-LEval", "venue": "North American Chapter of the Association for Computational Linguistics", "quality_score": 0.4157}
{"id": "4f05240a05eafc144c56514060ddee1685c016b89afefac6d2605dfab3b6f223", "sources": ["arxiv", "semantic_scholar"], "title": "RULER: What's the Real Context Size of Your Long-Context Language Models?", "abstract": "The needle-in-a-haystack (NIAH) test, which examines the ability to retrieve a piece of information (the \"needle\") from long distractor texts (the \"haystack\"), has been widely adopted to evaluate long-context language models (LMs). However, this simple retrieval-based test is indicative of only a superficial form of long-context understanding. To provide a more comprehensive evaluation of long-context LMs, we create a new synthetic benchmark RULER with flexible configurations for customized sequence length and task complexity. RULER expands upon the vanilla NIAH test to encompass variations with diverse types and quantities of needles. Moreover, RULER introduces new task categories multi-hop tracing and aggregation to test behaviors beyond searching from context. We evaluate 17 long-context LMs with 13 representative tasks in RULER. Despite achieving nearly perfect accuracy in the vanilla NIAH test, almost all models exhibit large performance drops as the context length increases. While these models all claim context sizes of 32K tokens or greater, only half of them can maintain satisfactory performance at the length of 32K. Our analysis of Yi-34B, which supports context length of 200K, reveals large room for improvement as we increase input length and task complexity. We open source RULER to spur comprehensive evaluation of long-context LMs.", "authors": ["Cheng-Ping Hsieh", "Simeng Sun", "Samuel Kriman", "Shantanu Acharya", "Dima Rekesh", "Fei Jia", "Yang Zhang", "Boris Ginsburg"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-09", "url": "https://arxiv.org/abs/2404.06654", "pdf_url": "https://arxiv.org/pdf/2404.06654v3", "arxiv_id": "2404.06654", "doi": "10.48550/arXiv.2404.06654", "citation_count": 996, "influential_citation_count": 244, "has_code": true, "code_url": "https://github.com/hsiehjackson/RULER", "venue": "arXiv.org", "quality_score": 1.0}
{"id": "1b0dc399f94621be56087f7be2f6ab19661108e3945e2c2d7e8fc6302ed3088e", "sources": ["arxiv", "semantic_scholar"], "title": "XL$^2$Bench: A Benchmark for Extremely Long Context Understanding with Long-range Dependencies", "abstract": "Large Language Models (LLMs) have demonstrated remarkable performance across diverse tasks but are constrained by their small context window sizes. Various efforts have been proposed to expand the context window to accommodate even up to 200K input tokens. Meanwhile, building high-quality benchmarks with much longer text lengths and more demanding tasks to provide comprehensive evaluations is of immense practical interest to facilitate long context understanding research of LLMs. However, prior benchmarks create datasets that ostensibly cater to long-text comprehension by expanding the input of traditional tasks, which falls short to exhibit the unique characteristics of long-text understanding, including long dependency tasks and longer text length compatible with modern LLMs' context window size. In this paper, we introduce a benchmark for extremely long context understanding with long-range dependencies, XL$^2$Bench, which includes three scenarios: Fiction Reading, Paper Reading, and Law Reading, and four tasks of increasing complexity: Memory Retrieval, Detailed Understanding, Overall Understanding, and Open-ended Generation, covering 27 subtasks in English and Chinese. It has an average length of 100K+ words (English) and 200K+ characters (Chinese). Evaluating six leading LLMs on XL$^2$Bench, we find that their performance significantly lags behind human levels. Moreover, the observed decline in performance across both the original and enhanced datasets underscores the efficacy of our approach to mitigating data contamination.", "authors": ["Xuanfan Ni", "Hengyi Cai", "Xiaochi Wei", "Shuaiqiang Wang", "Dawei Yin", "Piji Li"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-08", "url": "https://arxiv.org/abs/2404.05446", "pdf_url": "https://arxiv.org/pdf/2404.05446v1", "arxiv_id": "2404.05446", "doi": null, "citation_count": 14, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.294}
{"id": "3b65d1cc5a06506048aa61f6269c9a2a3264ba222f50f72fb06173d4b1727204", "sources": ["arxiv", "semantic_scholar"], "title": "Long-context LLMs Struggle with Long In-context Learning", "abstract": "Large Language Models (LLMs) have made significant strides in handling long sequences. Some models like Gemini could even to be capable of dealing with millions of tokens. However, their performance evaluation has largely been confined to metrics like perplexity and synthetic tasks, which may not fully capture their true abilities in more challenging, real-world scenarios. We introduce a benchmark (LongICLBench) for long in-context learning in extreme-label classification using six datasets with 28 to 174 classes and input lengths from 2K to 50K tokens. Our benchmark requires LLMs to comprehend the entire input to recognize the massive label spaces to make correct predictions. We evaluate on 15 long-context LLMs and find that they perform well on less challenging classification tasks with smaller label space and shorter demonstrations. However, they struggle with more challenging task like Discovery with 174 labels, suggesting a gap in their ability to process long, context-rich sequences. Further analysis reveals a bias towards labels presented later in the sequence and a need for improved reasoning over multiple pieces of information. Our study reveals that long context understanding and reasoning is still a challenging task for the existing LLMs. We believe LongICLBench could serve as a more realistic evaluation for the future long-context LLMs.", "authors": ["Tianle Li", "Ge Zhang", "Quy Duc Do", "Xiang Yue", "Wenhu Chen"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-02", "url": "https://arxiv.org/abs/2404.02060", "pdf_url": "https://arxiv.org/pdf/2404.02060v3", "arxiv_id": "2404.02060", "doi": "10.48550/arXiv.2404.02060", "citation_count": 367, "influential_citation_count": 20, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.6611}
{"id": "b7c3afff9b8b6c493245a3681dc30fc436836c5d6f11139a5f680c30ffad08e3", "sources": ["arxiv", "semantic_scholar"], "title": "Position-Aware Parameter Efficient Fine-Tuning Approach for Reducing Positional Bias in LLMs", "abstract": "Recent advances in large language models (LLMs) have enhanced their ability to process long input contexts. This development is particularly crucial for tasks that involve retrieving knowledge from an external datastore, which can result in long inputs. However, recent studies show a positional bias in LLMs, demonstrating varying performance depending on the location of useful information within the input sequence. In this study, we conduct extensive experiments to investigate the root causes of positional bias. Our findings indicate that the primary contributor to LLM positional bias stems from the inherent positional preferences of different models. We demonstrate that merely employing prompt-based solutions is inadequate for overcoming the positional preferences. To address this positional bias issue of a pre-trained LLM, we developed a Position-Aware Parameter Efficient Fine-Tuning (PAPEFT) approach which is composed of a data augmentation technique and a parameter efficient adapter, enhancing a uniform attention distribution across the input context. Our experiments demonstrate that the proposed approach effectively reduces positional bias, improving LLMs' effectiveness in handling long context sequences for various tasks that require externally retrieved knowledge.", "authors": ["Zheng Zhang", "Fan Yang", "Ziyan Jiang", "Zheng Chen", "Zhengyang Zhao", "Chengyuan Ma", "Liang Zhao", "Yang Liu"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-04-01", "url": "https://arxiv.org/abs/2404.01430", "pdf_url": "https://arxiv.org/pdf/2404.01430v1", "arxiv_id": "2404.01430", "doi": "10.48550/arXiv.2404.01430", "citation_count": 13, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2865}
{"id": "9b51d31955dbaa8802be04c85f7ed12174d261051b2434589c311bf5c5b59787", "sources": ["arxiv", "semantic_scholar"], "title": "Tabular Learning: Encoding for Entity and Context Embeddings", "abstract": "Examining the effect of different encoding techniques on entity and context embeddings, the goal of this work is to challenge commonly used Ordinal encoding for tabular learning. Applying different preprocessing methods and network architectures over several datasets resulted in a benchmark on how the encoders influence the learning outcome of the networks. By keeping the test, validation and training data consistent, results have shown that ordinal encoding is not the most suited encoder for categorical data in terms of preprocessing the data and thereafter, classifying the target variable correctly. A better outcome was achieved, encoding the features based on string similarities by computing a similarity matrix as input for the network. This is the case for both, entity and context embeddings, where the transformer architecture showed improved performance for Ordinal and Similarity encoding with regard to multi-label classification tasks.", "authors": ["Fredy Reusser"], "categories": ["cs.LG", "cs.AI", "cs.CE"], "fields_of_study": ["Computer Science"], "published_date": "2024-03-28", "url": "https://arxiv.org/abs/2403.19405", "pdf_url": "https://arxiv.org/pdf/2403.19405v1", "arxiv_id": "2403.19405", "doi": "10.48550/arXiv.2403.19405", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2113}
{"id": "43b18cf9f431ae1c0f902b1168fef926d19519bd5b302d215955d35aa07663ce", "sources": ["arxiv", "semantic_scholar"], "title": "Naive Bayes-based Context Extension for Large Language Models", "abstract": "Large Language Models (LLMs) have shown promising in-context learning abilities. However, conventional In-Context Learning (ICL) approaches are often impeded by length limitations of transformer architecture, which pose challenges when attempting to effectively integrate supervision from a substantial number of demonstration examples. In this paper, we introduce a novel framework, called Naive Bayes-based Context Extension (NBCE), to enable existing LLMs to perform ICL with an increased number of demonstrations by significantly expanding their context size. Importantly, this expansion does not require fine-tuning or dependence on particular model architectures, all the while preserving linear efficiency. NBCE initially splits the context into equal-sized windows fitting the target LLM's maximum length. Then, it introduces a voting mechanism to select the most relevant window, regarded as the posterior context. Finally, it employs Bayes' theorem to generate the test task. Our experimental results demonstrate that NBCE substantially enhances performance, particularly as the number of demonstration examples increases, consistently outperforming alternative methods. The NBCE code will be made publicly accessible. The code NBCE is available at: https://github.com/amurtadha/NBCE-master", "authors": ["Jianlin Su", "Murtadha Ahmed", " Wenbo", "Luo Ao", "Mingren Zhu", "Yunfeng Liu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-03-26", "url": "https://arxiv.org/abs/2403.17552", "pdf_url": "https://arxiv.org/pdf/2403.17552v1", "arxiv_id": "2403.17552", "doi": "10.48550/arXiv.2403.17552", "citation_count": 8, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/amurtadha/NBCE-master", "venue": "North American Chapter of the Association for Computational Linguistics", "quality_score": 0.2386}
{"id": "cbde06cd370d1f57e7c2674eb2585cb95e85809dbf7d96144adc680a3332385c", "sources": ["arxiv", "semantic_scholar"], "title": "In-Context Matting", "abstract": "We introduce in-context matting, a novel task setting of image matting. Given a reference image of a certain foreground and guided priors such as points, scribbles, and masks, in-context matting enables automatic alpha estimation on a batch of target images of the same foreground category, without additional auxiliary input. This setting marries good performance in auxiliary input-based matting and ease of use in automatic matting, which finds a good trade-off between customization and automation. To overcome the key challenge of accurate foreground matching, we introduce IconMatting, an in-context matting model built upon a pre-trained text-to-image diffusion model. Conditioned on inter- and intra-similarity matching, IconMatting can make full use of reference context to generate accurate target alpha mattes. To benchmark the task, we also introduce a novel testing dataset ICM-$57$, covering 57 groups of real-world images. Quantitative and qualitative results on the ICM-57 testing set show that IconMatting rivals the accuracy of trimap-based matting while retaining the automation level akin to automatic matting. Code is available at https://github.com/tiny-smart/in-context-matting", "authors": ["He Guo", "Zixuan Ye", "Zhiguo Cao", "Hao Lu"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-03-23", "url": "https://arxiv.org/abs/2403.15789", "pdf_url": "https://arxiv.org/pdf/2403.15789v1", "arxiv_id": "2403.15789", "doi": "10.1109/CVPR52733.2024.00356", "citation_count": 6, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/tiny-smart/in-context-matting", "venue": "Computer Vision and Pattern Recognition", "quality_score": 0.2113}
{"id": "02f6dccfb46f56d44e2e0419beed0143cdce6338efe73e2b8d4fb4d90f097dee", "sources": ["arxiv", "semantic_scholar"], "title": "Found in the Middle: How Language Models Use Long Contexts Better via Plug-and-Play Positional Encoding", "abstract": "This paper aims to overcome the \"lost-in-the-middle\" challenge of large language models (LLMs). While recent advancements have successfully enabled LLMs to perform stable language modeling with up to 4 million tokens, the persistent difficulty faced by most LLMs in identifying relevant information situated in the middle of the context has not been adequately tackled. To address this problem, this paper introduces Multi-scale Positional Encoding (Ms-PoE) which is a simple yet effective plug-and-play approach to enhance the capacity of LLMs to handle the relevant information located in the middle of the context, without fine-tuning or introducing any additional overhead. Ms-PoE leverages the position indice rescaling to relieve the long-term decay effect introduced by RoPE, while meticulously assigning distinct scaling ratios to different attention heads to preserve essential knowledge learned during the pre-training step, forming a multi-scale context fusion from short to long distance. Extensive experiments with a wide range of LLMs demonstrate the efficacy of our approach. Notably, Ms-PoE achieves an average accuracy gain of up to 3.8 on the Zero-SCROLLS benchmark over the original LLMs. Code are available at https://github.com/VITA-Group/Ms-PoE.", "authors": ["Zhenyu Zhang", "Runjin Chen", "Shiwei Liu", "Zhewei Yao", "Olatunji Ruwase", "Beidi Chen", "Xiaoxia Wu", "Zhangyang Wang"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-03-05", "url": "https://arxiv.org/abs/2403.04797", "pdf_url": "https://arxiv.org/pdf/2403.04797v1", "arxiv_id": "2403.04797", "doi": "10.48550/arXiv.2403.04797", "citation_count": 83, "influential_citation_count": 5, "has_code": true, "code_url": "https://github.com/VITA-Group/Ms-PoE", "venue": "Neural Information Processing Systems", "quality_score": 0.4811}
{"id": "39f73b28b3e251d772850e410d855feb07cb8fd53ecd320774f63397b296deb5", "sources": ["arxiv", "semantic_scholar"], "title": "Long-Context Language Modeling with Parallel Context Encoding", "abstract": "Extending large language models (LLMs) to process longer inputs is crucial for a wide range of applications. However, the substantial computational cost of transformers and limited generalization of positional encoding restrict the size of their context window. We introduce Context Expansion with Parallel Encoding (CEPE), a framework that can be applied to any existing decoder-only LLMs to extend their context window. CEPE employs a small encoder to process long inputs chunk by chunk, enabling the frozen decoder to utilize additional contexts via cross-attention. CEPE is efficient, generalizable, and versatile: trained with 8K-token documents, it extends the context window of LLAMA-2 to 128K tokens, offering 10x the throughput with only 1/6 of the memory. CEPE yields strong performance on language modeling and in-context learning. CEPE also excels in retrieval-augmented applications, while existing long-context models degenerate with retrieved contexts. We further introduce a CEPE variant that can extend the context window of instruction-tuned models using only unlabeled data, and showcase its effectiveness on LLAMA-2-CHAT, leading to a strong instruction-following model that can leverage very long contexts on downstream tasks.", "authors": ["Howard Yen", "Tianyu Gao", "Danqi Chen"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-26", "url": "https://arxiv.org/abs/2402.16617", "pdf_url": "https://arxiv.org/pdf/2402.16617v2", "arxiv_id": "2402.16617", "doi": "10.48550/arXiv.2402.16617", "citation_count": 92, "influential_citation_count": 11, "has_code": true, "code_url": "https://github.com/princeton-nlp/CEPE", "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.5396}
{"id": "46f185859a4f1d2853a85c82d008685681fdf49628d7d237b250152467f4c9b6", "sources": ["arxiv", "semantic_scholar"], "title": "How Large Language Models Encode Context Knowledge? A Layer-Wise Probing Study", "abstract": "Previous work has showcased the intriguing capability of large language models (LLMs) in retrieving facts and processing context knowledge. However, only limited research exists on the layer-wise capability of LLMs to encode knowledge, which challenges our understanding of their internal mechanisms. In this paper, we devote the first attempt to investigate the layer-wise capability of LLMs through probing tasks. We leverage the powerful generative capability of ChatGPT to construct probing datasets, providing diverse and coherent evidence corresponding to various facts. We employ $\\mathcal V$-usable information as the validation metric to better reflect the capability in encoding context knowledge across different layers. Our experiments on conflicting and newly acquired knowledge show that LLMs: (1) prefer to encode more context knowledge in the upper layers; (2) primarily encode context knowledge within knowledge-related entity tokens at lower layers while progressively expanding more knowledge within other tokens at upper layers; and (3) gradually forget the earlier context knowledge retained within the intermediate layers when provided with irrelevant evidence. Code is publicly available at https://github.com/Jometeorie/probing_llama.", "authors": ["Tianjie Ju", "Weiwei Sun", "Wei Du", "Xinwei Yuan", "Zhaochun Ren", "Gongshen Liu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-25", "url": "https://arxiv.org/abs/2402.16061", "pdf_url": "https://arxiv.org/pdf/2402.16061v2", "arxiv_id": "2402.16061", "doi": "10.48550/arXiv.2402.16061", "citation_count": 79, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/Jometeorie/probing_llama", "venue": "International Conference on Language Resources and Evaluation", "quality_score": 0.4758}
{"id": "ded7e5d4f135de53d68a9a013ba9d466369b68d9e153e6812a7dfa1369e28d8d", "sources": ["arxiv", "semantic_scholar"], "title": "CFIR: Fast and Effective Long-Text To Image Retrieval for Large Corpora", "abstract": "Text-to-image retrieval aims to find the relevant images based on a text query, which is important in various use-cases, such as digital libraries, e-commerce, and multimedia databases. Although Multimodal Large Language Models (MLLMs) demonstrate state-of-the-art performance, they exhibit limitations in handling large-scale, diverse, and ambiguous real-world needs of retrieval, due to the computation cost and the injective embeddings they produce. This paper presents a two-stage Coarse-to-Fine Index-shared Retrieval (CFIR) framework, designed for fast and effective large-scale long-text to image retrieval. The first stage, Entity-based Ranking (ER), adapts to long-text query ambiguity by employing a multiple-queries-to-multiple-targets paradigm, facilitating candidate filtering for the next stage. The second stage, Summary-based Re-ranking (SR), refines these rankings using summarized queries. We also propose a specialized Decoupling-BEiT-3 encoder, optimized for handling ambiguous user needs and both stages, which also enhances computational efficiency through vector-based similarity inference. Evaluation on the AToMiC dataset reveals that CFIR surpasses existing MLLMs by up to 11.06% in Recall@1000, while reducing training and retrieval times by 68.75% and 99.79%, respectively. We will release our code to facilitate future research at https://github.com/longkukuhi/CFIR.", "authors": ["Zijun Long", "Xuri Ge", "Richard Mccreadie", "Joemon Jose"], "categories": ["cs.IR", "cs.AI", "cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-23", "url": "https://arxiv.org/abs/2402.15276", "pdf_url": "https://arxiv.org/pdf/2402.15276v3", "arxiv_id": "2402.15276", "doi": "10.1145/3626772.3657741", "citation_count": 14, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/longkukuhi/CFIR", "venue": "Annual International ACM SIGIR Conference on Research and Development in Information Retrieval", "quality_score": 0.294}
{"id": "37f0750c3cae9fc9c1be28b2d2adef60d473cdbb0f50310d6de6e4e54aeda5a8", "sources": ["arxiv", "semantic_scholar"], "title": "LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens", "abstract": "Large context window is a desirable feature in large language models (LLMs). However, due to high fine-tuning costs, scarcity of long texts, and catastrophic values introduced by new token positions, current extended context windows are limited to around 128k tokens. This paper introduces LongRoPE that, for the first time, extends the context window of pre-trained LLMs to an impressive 2048k tokens, with up to only 1k fine-tuning steps at within 256k training lengths, while maintaining performance at the original short context window. This is achieved by three key innovations: (i) we identify and exploit two forms of non-uniformities in positional interpolation through an efficient search, providing a better initialization for fine-tuning and enabling an 8x extension in non-fine-tuning scenarios; (ii) we introduce a progressive extension strategy that first fine-tunes a 256k length LLM and then conducts a second positional interpolation on the fine-tuned extended LLM to achieve a 2048k context window; (iii) we readjust LongRoPE on 8k length to recover the short context window performance. Extensive experiments on LLaMA2 and Mistral across various tasks demonstrate the effectiveness of our method. Models extended via LongRoPE retain the original architecture with minor modifications to the positional embedding, and can reuse most pre-existing optimizations.", "authors": ["Yiran Ding", "Li Lyna Zhang", "Chengruidong Zhang", "Yuanyuan Xu", "Ning Shang", "Jiahang Xu", "Fan Yang", "Mao Yang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-21", "url": "https://arxiv.org/abs/2402.13753", "pdf_url": "https://arxiv.org/pdf/2402.13753v1", "arxiv_id": "2402.13753", "doi": "10.48550/arXiv.2402.13753", "citation_count": 343, "influential_citation_count": 22, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.6809}
{"id": "751a5e76447439defb00cd49d12cc3bfd8409742a0cbf4e4aaaded7c69ee9650", "sources": ["arxiv", "semantic_scholar"], "title": "$\\infty$Bench: Extending Long Context Evaluation Beyond 100K Tokens", "abstract": "Processing and reasoning over long contexts is crucial for many practical applications of Large Language Models (LLMs), such as document comprehension and agent construction. Despite recent strides in making LLMs process contexts with more than 100K tokens, there is currently a lack of a standardized benchmark to evaluate this long-context capability. Existing public benchmarks typically focus on contexts around 10K tokens, limiting the assessment and comparison of LLMs in processing longer contexts. In this paper, we propose $\\infty$Bench, the first LLM benchmark featuring an average data length surpassing 100K tokens. $\\infty$Bench comprises synthetic and realistic tasks spanning diverse domains, presented in both English and Chinese. The tasks in $\\infty$Bench are designed to require well understanding of long dependencies in contexts, and make simply retrieving a limited number of passages from contexts not sufficient for these tasks. In our experiments, based on $\\infty$Bench, we evaluate the state-of-the-art proprietary and open-source LLMs tailored for processing long contexts. The results indicate that existing long context LLMs still require significant advancements to effectively process 100K+ context. We further present three intriguing analyses regarding the behavior of LLMs processing long context.", "authors": ["Xinrong Zhang", "Yingfa Chen", "Shengding Hu", "Zihang Xu", "Junhao Chen", "Moo Khai Hao", "Xu Han", "Zhen Leng Thai", "Shuo Wang", "Zhiyuan Liu", "Maosong Sun"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-21", "url": "https://arxiv.org/abs/2402.13718", "pdf_url": "https://arxiv.org/pdf/2402.13718v3", "arxiv_id": "2402.13718", "doi": "10.48550/arXiv.2402.13718", "citation_count": 356, "influential_citation_count": 34, "has_code": true, "code_url": null, "venue": "Volume 1", "quality_score": 0.772}
{"id": "5b41c37224cdd4bd13c9fbe2c8d83012cd5834f5ae7a9db359cec755f30cf1c3", "sources": ["arxiv", "semantic_scholar"], "title": "Extensible Embedding: A Flexible Multipler For LLM's Context Length", "abstract": "Large language models (LLMs) call for extension of context to handle many critical applications. However, the existing approaches are prone to expensive costs and inferior quality of context extension. In this work, we propose Extensible Embedding, which realizes high-quality extension of LLM's context with strong flexibility and cost-effectiveness. Extensible embedding stand as an enhancement of typical token embedding, which represents the information for an extensible scope of context instead of a single token. By leveraging such compact input units of higher information density, the LLM can access to a vast scope of context even with a small context window. Extensible embedding is systematically optimized in architecture and training method, which leads to multiple advantages. 1) High flexibility of context extension, which flexibly supports ad-hoc extension of diverse context lengths. 2) Strong sample efficiency of training, which enables the embedding model to be learned in a cost-effective way. 3) Superior compatibility with the existing LLMs, where the extensible embedding can be seamlessly introduced as a plug-in component. Comprehensive evaluations on long-context language modeling and understanding tasks verify extensible embedding as an effective, efficient, flexible, and compatible method to extend the LLM's context.", "authors": ["Ninglu Shao", "Shitao Xiao", "Zheng Liu", "Peitian Zhang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-18", "url": "https://arxiv.org/abs/2402.11577", "pdf_url": "https://arxiv.org/pdf/2402.11577v1", "arxiv_id": "2402.11577", "doi": "10.48550/arXiv.2402.11577", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "32b69754ac1110e6a78370f3670b67cf700994f91b4963497af459a337bc7d80", "sources": ["arxiv", "semantic_scholar"], "title": "BGE Landmark Embedding: A Chunking-Free Embedding Method For Retrieval Augmented Long-Context Large Language Models", "abstract": "Large language models (LLMs) call for extension of context to handle many critical applications. However, the existing approaches are prone to expensive costs and inferior quality of context extension. In this work, we proposeExtensible Embedding, which realizes high-quality extension of LLM's context with strong flexibility and cost-effectiveness. Extensible embedding stand as an enhancement of typical token embedding, which represents the information for an extensible scope of context instead of a single token. By leveraging such compact input units of higher information density, the LLM can access to a vast scope of context even with a small context window. Extensible embedding is systematically optimized in architecture and training method, which leads to multiple advantages. 1) High flexibility of context extension, which flexibly supports ad-hoc extension of diverse context lengths. 2) Strong sample efficiency of training, which enables the embedding model to be learned in a cost-effective way. 3) Superior compatibility with the existing LLMs, where the extensible embedding can be seamlessly introduced as a plug-in component. Comprehensive evaluations on long-context language modeling and understanding tasks verify extensible embedding as an effective, efficient, flexible, and compatible method to extend the LLM's context.", "authors": ["Kun Luo", "Zheng Liu", "Shitao Xiao", "Kang Liu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-18", "url": "https://arxiv.org/abs/2402.11573", "pdf_url": "https://arxiv.org/pdf/2402.11573v1", "arxiv_id": "2402.11573", "doi": "10.48550/arXiv.2402.11573", "citation_count": 18, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.3197}
{"id": "62e229fa02f4f096bea3890dd4c9be798111dee82e4e9b10005defd9b2ebe126", "sources": ["arxiv", "semantic_scholar"], "title": "LongHeads: Multi-Head Attention is Secretly a Long Context Processor", "abstract": "Large language models (LLMs) have achieved impressive performance in numerous domains but often struggle to process lengthy inputs effectively and efficiently due to limited length generalization and attention's quadratic computational demands. Many sought to mitigate this by restricting the attention window within the pre-trained length. However, these methods introduce new issues such as ignoring the middle context and requiring additional training. To address these problems, we propose LongHeads, a training-free framework that enhances LLM's long context ability by unlocking multi-head attention's untapped potential. Instead of allowing each head to attend to the full sentence, which struggles with generalizing to longer sequences due to out-of-distribution (OOD) issues, we allow each head to process in-distribution length by selecting and attending to important context chunks. To this end, we propose a chunk selection strategy that relies on the inherent correlation between the query and the key representations, efficiently distributing context chunks to different heads. In this way, each head ensures it can effectively process attended tokens within the trained length, while different heads in different layers can collectively process longer contexts. LongHeads works efficiently in linear time, fits seamlessly with many LLMs that use relative positional encoding. LongHeads achieves 100% accuracy at the 128k length on passkey retrieval task, verifying LongHeads's efficacy in extending the usable context window for existing models. We release our code at https://github.com/LuLuLuyi/LongHeads .", "authors": ["Yi Lu", "Xin Zhou", "Wei He", "Jun Zhao", "Tao Ji", "Tao Gui", "Qi Zhang", "Xuanjing Huang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-16", "url": "https://arxiv.org/abs/2402.10685", "pdf_url": "https://arxiv.org/pdf/2402.10685v2", "arxiv_id": "2402.10685", "doi": "10.48550/arXiv.2402.10685", "citation_count": 25, "influential_citation_count": 3, "has_code": true, "code_url": "https://github.com/LuLuLuyi/LongHeads", "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.3537}
{"id": "f3f954c064bef9fd451d3c8631cb31141985e23f951512fe654806935b365c7b", "sources": ["arxiv", "semantic_scholar"], "title": "SPAR: Personalized Content-Based Recommendation via Long Engagement Attention", "abstract": "Leveraging users' long engagement histories is essential for personalized content recommendations. The success of pretrained language models (PLMs) in NLP has led to their use in encoding user histories and candidate items, framing content recommendations as textual semantic matching tasks. However, existing works still struggle with processing very long user historical text and insufficient user-item interaction. In this paper, we introduce a content-based recommendation framework, SPAR, which effectively tackles the challenges of holistic user interest extraction from the long user engagement history. It achieves so by leveraging PLM, poly-attention layers and attention sparsity mechanisms to encode user's history in a session-based manner. The user and item side features are sufficiently fused for engagement prediction while maintaining standalone representations for both sides, which is efficient for practical model deployment. Moreover, we enhance user profiling by exploiting large language model (LLM) to extract global interests from user engagement history. Extensive experiments on two benchmark datasets demonstrate that our framework outperforms existing state-of-the-art (SoTA) methods.", "authors": ["Chiyu Zhang", "Yifei Sun", "Jun Chen", "Jie Lei", "Muhammad Abdul-Mageed", "Sinong Wang", "Rong Jin", "Sem Park", "Ning Yao", "Bo Long"], "categories": ["cs.IR", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-16", "url": "https://arxiv.org/abs/2402.10555", "pdf_url": "https://arxiv.org/pdf/2402.10555v2", "arxiv_id": "2402.10555", "doi": "10.48550/arXiv.2402.10555", "citation_count": 11, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2698}
{"id": "e81aed328ea5adcff7301f32aa89d8cdd5a7195d5e737674a324d693f09bd02b", "sources": ["arxiv", "semantic_scholar"], "title": "Data Engineering for Scaling Language Models to 128K Context", "abstract": "We study the continual pretraining recipe for scaling language models' context lengths to 128K, with a focus on data engineering. We hypothesize that long context modeling, in particular \\textit{the ability to utilize information at arbitrary input locations}, is a capability that is mostly already acquired through large-scale pretraining, and that this capability can be readily extended to contexts substantially longer than seen during training~(e.g., 4K to 128K) through lightweight continual pretraining on appropriate data mixture. We investigate the \\textit{quantity} and \\textit{quality} of the data for continual pretraining: (1) for quantity, we show that 500 million to 5 billion tokens are enough to enable the model to retrieve information anywhere within the 128K context; (2) for quality, our results equally emphasize \\textit{domain balance} and \\textit{length upsampling}. Concretely, we find that naively upsampling longer data on certain domains like books, a common practice of existing work, gives suboptimal performance, and that a balanced domain mixture is important. We demonstrate that continual pretraining of the full model on 1B-5B tokens of such data is an effective and affordable strategy for scaling the context length of language models to 128K. Our recipe outperforms strong open-source long-context models and closes the gap to frontier models like GPT-4 128K.", "authors": ["Yao Fu", "Rameswar Panda", "Xinyao Niu", "Xiang Yue", "Hannaneh Hajishirzi", "Yoon Kim", "Hao Peng"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-15", "url": "https://arxiv.org/abs/2402.10171", "pdf_url": "https://arxiv.org/pdf/2402.10171v1", "arxiv_id": "2402.10171", "doi": "10.48550/arXiv.2402.10171", "citation_count": 214, "influential_citation_count": 22, "has_code": true, "code_url": "https://github.com/FranxYao/Long-Context-Data-Engineering", "venue": "International Conference on Machine Learning", "quality_score": 0.6809}
{"id": "bd9421aab330e06363ff3d435712f9e8615086cbba530540a9231002c8a90390", "sources": ["arxiv", "semantic_scholar"], "title": "A Human-Inspired Reading Agent with Gist Memory of Very Long Contexts", "abstract": "Current Large Language Models (LLMs) are not only limited to some maximum context length, but also are not able to robustly consume long inputs. To address these limitations, we propose ReadAgent, an LLM agent system that increases effective context length up to 20x in our experiments. Inspired by how humans interactively read long documents, we implement ReadAgent as a simple prompting system that uses the advanced language capabilities of LLMs to (1) decide what content to store together in a memory episode, (2) compress those memory episodes into short episodic memories called gist memories, and (3) take actions to look up passages in the original text if ReadAgent needs to remind itself of relevant details to complete a task. We evaluate ReadAgent against baselines using retrieval methods, using the original long contexts, and using the gist memories. These evaluations are performed on three long-document reading comprehension tasks: QuALITY, NarrativeQA, and QMSum. ReadAgent outperforms the baselines on all three tasks while extending the effective context window by 3.5-20x.", "authors": ["Kuang-Huei Lee", "Xinyun Chen", "Hiroki Furuta", "John Canny", "Ian Fischer"], "categories": ["cs.CL", "cs.AI", "cs.IR"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-15", "url": "https://arxiv.org/abs/2402.09727", "pdf_url": "https://arxiv.org/pdf/2402.09727v3", "arxiv_id": "2402.09727", "doi": "10.48550/arXiv.2402.09727", "citation_count": 127, "influential_citation_count": 15, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.6021}
{"id": "ccddee467610fbdcc356c967a49bc0884f98cd05e2f7d287390f31628c55a777", "sources": ["arxiv", "semantic_scholar"], "title": "InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory", "abstract": "Large language models (LLMs) have emerged as a cornerstone in real-world applications with lengthy streaming inputs (e.g., LLM-driven agents). However, existing LLMs, pre-trained on sequences with a restricted maximum length, cannot process longer sequences due to the out-of-domain and distraction issues. Common solutions often involve continual pre-training on longer sequences, which will introduce expensive computational overhead and uncontrollable change in model capabilities. In this paper, we unveil the intrinsic capacity of LLMs for understanding extremely long sequences without any fine-tuning. To this end, we introduce a training-free memory-based method, InfLLM. Specifically, InfLLM stores distant contexts into additional memory units and employs an efficient mechanism to lookup token-relevant units for attention computation. Thereby, InfLLM allows LLMs to efficiently process long sequences with a limited context window and well capture long-distance dependencies. Without any training, InfLLM enables LLMs that are pre-trained on sequences consisting of a few thousand tokens to achieve comparable performance with competitive baselines that continually train these LLMs on long sequences. Even when the sequence length is scaled to $1,024$K, InfLLM still effectively captures long-distance dependencies. Our code can be found in \\url{https://github.com/thunlp/InfLLM}.", "authors": ["Chaojun Xiao", "Pengle Zhang", "Xu Han", "Guangxuan Xiao", "Yankai Lin", "Zhengyan Zhang", "Zhiyuan Liu", "Maosong Sun"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-07", "url": "https://arxiv.org/abs/2402.04617", "pdf_url": "https://arxiv.org/pdf/2402.04617v2", "arxiv_id": "2402.04617", "doi": "10.52202/079017-3801", "citation_count": 165, "influential_citation_count": 14, "has_code": true, "code_url": "https://github.com/thunlp/InfLLM}", "venue": "Neural Information Processing Systems", "quality_score": 0.588}
{"id": "0b2df4b4193948d410be6df89b6fad3a6adcb47dd93b0b8b21d4e734a3a60a4f", "sources": ["arxiv", "semantic_scholar"], "title": "Encoding Version History Context for Better Code Representation", "abstract": "With the exponential growth of AI tools that generate source code, understanding software has become crucial. When developers comprehend a program, they may refer to additional contexts to look for information, e.g. program documentation or historical code versions. Therefore, we argue that encoding this additional contextual information could also benefit code representation for deep learning. Recent papers incorporate contextual data (e.g. call hierarchy) into vector representation to address program comprehension problems. This motivates further studies to explore additional contexts, such as version history, to enhance models' understanding of programs. That is, insights from version history enable recognition of patterns in code evolution over time, recurring issues, and the effectiveness of past solutions. Our paper presents preliminary evidence of the potential benefit of encoding contextual information from the version history to predict code clones and perform code classification. We experiment with two representative deep learning models, ASTNN and CodeBERT, to investigate whether combining additional contexts with different aggregations may benefit downstream activities. The experimental result affirms the positive impact of combining version history into source code representation in all scenarios; however, to ensure the technique performs consistently, we need to conduct a holistic investigation on a larger code base using different combinations of contexts, aggregation, and models. Therefore, we propose a research agenda aimed at exploring various aspects of encoding additional context to improve code representation and its optimal utilisation in specific situations.", "authors": ["Huy Nguyen", "Christoph Treude", "Patanamon Thongtanunam"], "categories": ["cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-06", "url": "https://arxiv.org/abs/2402.03773", "pdf_url": "https://arxiv.org/pdf/2402.03773v1", "arxiv_id": "2402.03773", "doi": "10.1145/3643991.3644929", "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "IEEE Working Conference on Mining Software Repositories", "quality_score": 0.1505}
{"id": "36588ac84e5c1b0ec394219cbda5f52e1495c0d0755e5e9f0f8b93aef3664443", "sources": ["arxiv", "semantic_scholar"], "title": "Fast Peer Adaptation with Context-aware Exploration", "abstract": "Fast adapting to unknown peers (partners or opponents) with different strategies is a key challenge in multi-agent games. To do so, it is crucial for the agent to probe and identify the peer's strategy efficiently, as this is the prerequisite for carrying out the best response in adaptation. However, exploring the strategies of unknown peers is difficult, especially when the games are partially observable and have a long horizon. In this paper, we propose a peer identification reward, which rewards the learning agent based on how well it can identify the behavior pattern of the peer over the historical context, such as the observation over multiple episodes. This reward motivates the agent to learn a context-aware policy for effective exploration and fast adaptation, i.e., to actively seek and collect informative feedback from peers when uncertain about their policies and to exploit the context to perform the best response when confident. We evaluate our method on diverse testbeds that involve competitive (Kuhn Poker), cooperative (PO-Overcooked), or mixed (Predator-Prey-W) games with peer agents. We demonstrate that our method induces more active exploration behavior, achieving faster adaptation and better outcomes than existing methods.", "authors": ["Long Ma", "Yuanfei Wang", "Fangwei Zhong", "Song-Chun Zhu", "Yizhou Wang"], "categories": ["cs.AI", "cs.LG", "cs.MA"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-04", "url": "https://arxiv.org/abs/2402.02468", "pdf_url": "https://arxiv.org/pdf/2402.02468v2", "arxiv_id": "2402.02468", "doi": "10.48550/arXiv.2402.02468", "citation_count": 12, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "International Conference on Machine Learning", "quality_score": 0.2785}
{"id": "5e037121300917e6574276496b84a4798a3fbf27542f630ef06d683a002f3f99", "sources": ["arxiv", "semantic_scholar"], "title": "Towards an Information Theoretic Framework of Context-Based Offline Meta-Reinforcement Learning", "abstract": "As a marriage between offline RL and meta-RL, the advent of offline meta-reinforcement learning (OMRL) has shown great promise in enabling RL agents to multi-task and quickly adapt while acquiring knowledge safely. Among which, context-based OMRL (COMRL) as a popular paradigm, aims to learn a universal policy conditioned on effective task representations. In this work, by examining several key milestones in the field of COMRL, we propose to integrate these seemingly independent methodologies into a unified framework. Most importantly, we show that the pre-existing COMRL algorithms are essentially optimizing the same mutual information objective between the task variable $M$ and its latent representation $Z$ by implementing various approximate bounds. Such theoretical insight offers ample design freedom for novel algorithms. As demonstrations, we propose a supervised and a self-supervised implementation of $I(Z; M)$, and empirically show that the corresponding optimization algorithms exhibit remarkable generalization across a broad spectrum of RL benchmarks, context shift scenarios, data qualities and deep learning architectures. This work lays the information theoretic foundation for COMRL methods, leading to a better understanding of task representation learning in the context of reinforcement learning. Given its generality, we envision our framework as a promising offline pre-training paradigm of foundation models for decision making.", "authors": ["Lanqing Li", "Hai Zhang", "Xinyu Zhang", "Shatong Zhu", "Yang Yu", "Junqiao Zhao", "Pheng-Ann Heng"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-02-04", "url": "https://arxiv.org/abs/2402.02429", "pdf_url": "https://arxiv.org/pdf/2402.02429v3", "arxiv_id": "2402.02429", "doi": "10.48550/arXiv.2402.02429", "citation_count": 22, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Neural Information Processing Systems", "quality_score": 0.3404}
{"id": "186a08858cf638e756430f420ce8c40e5f0aa6546c123cbf8c1f541b139aabf2", "sources": ["arxiv", "semantic_scholar"], "title": "Transformers and Cortical Waves: Encoders for Pulling In Context Across Time", "abstract": "The capabilities of transformer networks such as ChatGPT and other Large Language Models (LLMs) have captured the world's attention. The crucial computational mechanism underlying their performance relies on transforming a complete input sequence - for example, all the words in a sentence - into a long \"encoding vector\" that allows transformers to learn long-range temporal dependencies in naturalistic sequences. Specifically, \"self-attention\" applied to this encoding vector enhances temporal context in transformers by computing associations between pairs of words in the input sequence. We suggest that waves of neural activity traveling across single cortical areas or multiple regions at the whole-brain scale could implement a similar encoding principle. By encapsulating recent input history into a single spatial pattern at each moment in time, cortical waves may enable temporal context to be extracted from sequences of sensory inputs, the same computational principle used in transformers.", "authors": ["Lyle Muller", "Patricia S. Churchland", "Terrence J. Sejnowski"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science", "Medicine"], "published_date": "2024-01-25", "url": "https://arxiv.org/abs/2401.14267", "pdf_url": "https://arxiv.org/pdf/2401.14267v3", "arxiv_id": "2401.14267", "doi": "10.48550/arXiv.2401.14267", "citation_count": 20, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Trends in Neurosciences", "quality_score": 0.3306}
{"id": "313654e1a82f8e806326f2f90f845d74e31d4593a5d3734a379c665bea5e0317", "sources": ["arxiv", "semantic_scholar"], "title": "Flexibly Scaling Large Language Models Contexts Through Extensible Tokenization", "abstract": "Large language models (LLMs) are in need of sufficient contexts to handle many critical applications, such as retrieval augmented generation and few-shot learning. However, due to the constrained window size, the LLMs can only access to the information within a limited context. Although the size of context window can be extended by fine-tuning, it will result in a substantial cost in both training and inference stage. In this paper, we present Extensible Tokenization as an alternative method which realizes the flexible scaling of LLMs' context. Extensible Tokenization stands as a midware in between of the tokenized context and the LLM, which transforms the raw token embeddings into the extensible embeddings. Such embeddings provide a more compact representation for the long context, on top of which the LLM is able to perceive more information with the same context window. Extensible Tokenization is also featured by its flexibility: the scaling factor can be flexibly determined within a feasible scope, leading to the extension of an arbitrary context length at the inference time. Besides, Extensible Tokenization is introduced as a drop-in component, which can be seamlessly plugged into not only the LLM itself and but also its fine-tuned derivatives, bringing in the extended contextual information while fully preserving the LLM's existing capabilities. We perform comprehensive experiments on long-context language modeling and understanding tasks, which verify Extensible Tokenization as an effective, efficient, flexible, and compatible method to extend LLM's context. Our model and source code will be made publicly available.", "authors": ["Ninglu Shao", "Shitao Xiao", "Zheng Liu", "Peitian Zhang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-01-15", "url": "https://arxiv.org/abs/2401.07793", "pdf_url": "https://arxiv.org/pdf/2401.07793v1", "arxiv_id": "2401.07793", "doi": "10.48550/arXiv.2401.07793", "citation_count": 6, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2113}
{"id": "19eda29ca93c37b9fe0583b0ae38dccdde16436ac338bb3903af9e3914f88be8", "sources": ["arxiv", "semantic_scholar"], "title": "Extending LLMs' Context Window with 100 Samples", "abstract": "Large Language Models (LLMs) are known to have limited extrapolation ability beyond their pre-trained context window, constraining their application in downstream tasks with lengthy inputs. Recent studies have sought to extend LLMs' context window by modifying rotary position embedding (RoPE), a popular position encoding method adopted by well-known LLMs such as LLaMA, PaLM, and GPT-NeoX. However, prior works like Position Interpolation (PI) and YaRN are resource-intensive and lack comparative experiments to assess their applicability. In this work, we identify the inherent need for LLMs' attention entropy (i.e. the information entropy of attention scores) to maintain stability and introduce a novel extension to RoPE which combines adjusting RoPE's base frequency and scaling the attention logits to help LLMs efficiently adapt to a larger context window. We validate the superiority of our method in both fine-tuning performance and robustness across different context window sizes on various context-demanding tasks. Notably, our method extends the context window of LLaMA-2-7B-Chat to 16,384 with only 100 samples and 6 training steps, showcasing extraordinary efficiency. Finally, we also explore how data compositions and training curricula affect context window extension for specific downstream tasks, suggesting fine-tuning LLMs with lengthy conversations as a good starting point. We release our code and SFT data at https://github.com/GAIR-NLP/Entropy-ABF.", "authors": ["Yikai Zhang", "Junlong Li", "Pengfei Liu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2024-01-13", "url": "https://arxiv.org/abs/2401.07004", "pdf_url": "https://arxiv.org/pdf/2401.07004v1", "arxiv_id": "2401.07004", "doi": "10.48550/arXiv.2401.07004", "citation_count": 20, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/GAIR-NLP/Entropy-ABF", "venue": "arXiv.org", "quality_score": 0.3306}
{"id": "f7c4ec8e3d970b08e056cee2c1a42720ac2af8e66dcd4cffcbbfac64e83a2f6d", "sources": ["arxiv", "semantic_scholar"], "title": "Infinite-LLM: Efficient LLM Service for Long Context with DistAttention and Distributed KVCache", "abstract": "Large Language Models (LLMs) demonstrate substantial potential across a diverse array of domains via request serving. However, as trends continue to push for expanding context sizes, the autoregressive nature of LLMs results in highly dynamic behavior of the attention layers, showcasing significant differences in computational characteristics and memory requirements from the non-attention layers. This presents substantial challenges for resource management and performance optimization in service systems. Existing static model parallelism and resource allocation strategies fall short when dealing with this dynamicity. To address the issue, we propose Infinite-LLM, a novel LLM serving system designed to effectively handle dynamic context lengths. Infinite-LLM disaggregates attention layers from an LLM's inference process, facilitating flexible and independent resource scheduling that optimizes computational performance and enhances memory utilization jointly. By leveraging a pooled GPU memory strategy across a cluster, Infinite-LLM not only significantly boosts system throughput but also supports extensive context lengths. Evaluated on a dataset with context lengths ranging from a few to 2000K tokens across a cluster with 32 A100 GPUs, Infinite-LLM demonstrates throughput improvement of 1.35-3.4x compared to state-of-the-art methods, enabling efficient and elastic LLM deployment.", "authors": ["Bin Lin", "Chen Zhang", "Tao Peng", "Hanyu Zhao", "Wencong Xiao", "Minmin Sun", "Anmin Liu", "Zhipeng Zhang", "Lanbo Li", "Xiafei Qiu", "Shen Li", "Zhigang Ji", "Tao Xie", "Yong Li", "Wei Lin"], "categories": ["cs.DC", "cs.AR"], "fields_of_study": ["Computer Science"], "published_date": "2024-01-05", "url": "https://arxiv.org/abs/2401.02669", "pdf_url": "https://arxiv.org/pdf/2401.02669v2", "arxiv_id": "2401.02669", "doi": "10.48550/arXiv.2401.02669", "citation_count": 89, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4886}
{"id": "e2dc09b1295c1ef2100512614f7dd0713ba0763e604671270ca5dc2d5e381f26", "sources": ["arxiv", "semantic_scholar"], "title": "LLM Maybe LongLM: Self-Extend LLM Context Window Without Tuning", "abstract": "It is well known that LLMs cannot generalize well to long contexts whose lengths are larger than the training sequence length. This poses challenges when employing LLMs for processing long input sequences during inference. In this work, we argue that LLMs themselves have inherent capabilities to handle long contexts without fine-tuning. To achieve this goal, we propose SelfExtend to extend the context window of LLMs by constructing bi-level attention information: the grouped attention and the neighbor attention. The grouped attention captures the dependencies among tokens that are far apart, while neighbor attention captures dependencies among adjacent tokens within a specified range. The two-level attentions are computed based on the original model's self-attention mechanism during inference. With minor code modification, our SelfExtend can effortlessly extend existing LLMs' context window without any fine-tuning. We conduct comprehensive experiments on multiple benchmarks and the results show that our SelfExtend can effectively extend existing LLMs' context window length. The code can be found at \\url{https://github.com/datamllab/LongLM}.", "authors": ["Hongye Jin", "Xiaotian Han", "Jingfeng Yang", "Zhimeng Jiang", "Zirui Liu", "Chia-Yuan Chang", "Huiyuan Chen", "Xia Hu"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2024-01-02", "url": "https://arxiv.org/abs/2401.01325", "pdf_url": "https://arxiv.org/pdf/2401.01325v3", "arxiv_id": "2401.01325", "doi": "10.48550/arXiv.2401.01325", "citation_count": 176, "influential_citation_count": 11, "has_code": true, "code_url": "https://github.com/datamllab/LongLM}", "venue": "International Conference on Machine Learning", "quality_score": 0.562}
{"id": "f214699de158d9f45e18cfb696cb92b49ba00e964a989d1e42c655240dc210ec", "sources": ["arxiv", "semantic_scholar"], "title": "Structured Packing in LLM Training Improves Long Context Utilization", "abstract": "Recent advancements in long-context large language models have attracted significant attention, yet their practical applications often suffer from suboptimal context utilization. This study investigates structuring training data to enhance semantic interdependence, demonstrating that this approach effectively improves context utilization. To this end, we introduce the Structured Packing for Long Context (SPLiCe) method, which utilizes retrieval to collate mutually relevant documents into long and coherent training examples. We validate SPLiCe empirically across models of varying sizes -- 3B, 7B, and 13B -- achieving improved performance in long-context tasks, such as Qasper and HotpotQA. Remarkably, even brief fine-tuning with SPLiCe is sufficient to realize these benefits. Additionally, SPLiCe effectively mitigates the lost-in-middle phenomenon often observed in large models. Our comprehensive analysis of SPLiCe explores its design choices and reveals intriguing transfer effects; for instance, training on programming code enhances performance on natural language tasks.", "authors": ["Konrad Staniszewski", "Szymon Tworkowski", "Sebastian Jaszczur", "Yu Zhao", "Henryk Michalewski", "Łukasz Kuciński", "Piotr Miłoś"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-12-28", "url": "https://arxiv.org/abs/2312.17296", "pdf_url": "https://arxiv.org/pdf/2312.17296v9", "arxiv_id": "2312.17296", "doi": "10.48550/arXiv.2312.17296", "citation_count": 17, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "AAAI Conference on Artificial Intelligence", "quality_score": 0.3138}
{"id": "59fb8c161330e2073e4f32801176ae2c56736c1f3c2368933cb73b11e789415e", "sources": ["arxiv", "semantic_scholar"], "title": "Training With \"Paraphrasing the Original Text\" Teaches LLM to Better Retrieve in Long-context Tasks", "abstract": "As Large Language Models (LLMs) continue to evolve, more are being designed to handle long-context inputs. Despite this advancement, most of them still face challenges in accurately handling long-context tasks, often showing the \"lost in the middle\" issue. We identify that insufficient retrieval capability is one of the important reasons for this issue. To tackle this challenge, we propose a novel approach to design training data for long-context tasks, aiming at augmenting LLMs' proficiency in extracting key information from long context. Specially, we incorporate an additional part named \"paraphrasing the original text\" when constructing the answer of training samples and then fine-tuning the model. Experimenting on LongBench and NaturalQuestions Multi-document-QA dataset with models of Llama and Qwen series, our method achieves an improvement of up to 8.48% and 4.48% in average scores, respectively, showing effectiveness in improving the model's performance on long-context tasks.", "authors": ["Yijiong Yu", "Yongfeng Huang", "Zhixiao Qi", "Zhe Zhou"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-12-18", "url": "https://arxiv.org/abs/2312.11193", "pdf_url": "https://arxiv.org/pdf/2312.11193v10", "arxiv_id": "2312.11193", "doi": "10.1609/aaai.v39i24.34767", "citation_count": 5, "influential_citation_count": 1, "has_code": true, "code_url": "https://github.com/yuyijiong/train_with_paraphrasing", "venue": "AAAI Conference on Artificial Intelligence", "quality_score": 0.1945}
{"id": "8b11d38cf13299f2a92efa6c485578ec5d4362af591b313b759c600deb460378", "sources": ["arxiv", "semantic_scholar"], "title": "Zebra: Extending Context Window with Layerwise Grouped Local-Global Attention", "abstract": "This paper introduces a novel approach to enhance the capabilities of Large Language Models (LLMs) in processing and understanding extensive text sequences, a critical aspect in applications requiring deep comprehension and synthesis of large volumes of information. Recognizing the inherent challenges in extending the context window for LLMs, primarily built on Transformer architecture, we propose a new model architecture, referred to as Zebra. This architecture efficiently manages the quadratic time and memory complexity issues associated with full attention in the Transformer by employing grouped local-global attention layers. Our model, akin to a zebra's alternating stripes, balances local and global attention layers, significantly reducing computational requirements and memory consumption. Comprehensive experiments, including pretraining from scratch, continuation of long context adaptation training, and long instruction tuning, are conducted to evaluate the Zebra's performance. The results show that Zebra achieves comparable or superior performance on both short and long sequence benchmarks, while also enhancing training and inference efficiency.", "authors": ["Kaiqiang Song", "Xiaoyang Wang", "Sangwoo Cho", "Xiaoman Pan", "Dong Yu"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-12-14", "url": "https://arxiv.org/abs/2312.08618", "pdf_url": "https://arxiv.org/pdf/2312.08618v1", "arxiv_id": "2312.08618", "doi": "10.48550/arXiv.2312.08618", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "01c9d5ccd2ff6ff875c5457502d62d69315a8f4d0bd21688a12f13f34499ee78", "sources": ["arxiv", "semantic_scholar"], "title": "Fewer is More: Boosting LLM Reasoning with Reinforced Context Pruning", "abstract": "Large Language Models (LLMs) have shown impressive capabilities, yet they still struggle with math reasoning. In this work, we propose CoT-Influx, a novel approach that pushes the boundary of few-shot Chain-of-Thoughts (CoT) learning to improve LLM mathematical reasoning. Motivated by the observation that adding more concise CoT examples in the prompt can improve LLM reasoning performance, CoT-Influx employs a coarse-to-fine pruner to maximize the input of effective and concise CoT examples. The pruner first selects as many crucial CoT examples as possible and then prunes unimportant tokens to fit the context window. A math reasoning dataset with diverse difficulty levels and reasoning steps is used to train the pruner, along with a math-specialized reinforcement learning approach. As a result, by enabling more CoT examples with double the context window size in tokens, CoT-Influx significantly outperforms various prompting baselines across various LLMs (LLaMA2-7B, 13B, 70B) and 5 math datasets, achieving up to 4.55% absolute improvements. Remarkably, without any fine-tuning, LLaMA2-70B with CoT-Influx surpasses GPT-3.5 and a wide range of larger LLMs (PaLM, Minerva 540B, etc.) on the GSM8K. CoT-Influx serves as a plug-and-play module for LLMs and is compatible with most existing reasoning prompting techniques, such as self-consistency and self-verification.", "authors": ["Xijie Huang", "Li Lyna Zhang", "Kwang-Ting Cheng", "Fan Yang", "Mao Yang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-12-14", "url": "https://arxiv.org/abs/2312.08901", "pdf_url": "https://arxiv.org/pdf/2312.08901v3", "arxiv_id": "2312.08901", "doi": null, "citation_count": 19, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3253}
{"id": "08d3b900cddf4d035b778ead09ca262e9c87255bfc1454c171dab2c904780d45", "sources": ["arxiv", "semantic_scholar"], "title": "Skeleton-in-Context: Unified Skeleton Sequence Modeling with In-Context Learning", "abstract": "In-context learning provides a new perspective for multi-task modeling for vision and NLP. Under this setting, the model can perceive tasks from prompts and accomplish them without any extra task-specific head predictions or model fine-tuning. However, Skeleton sequence modeling via in-context learning remains unexplored. Directly applying existing in-context models from other areas onto skeleton sequences fails due to the inter-frame and cross-task pose similarity that makes it outstandingly hard to perceive the task correctly from a subtle context. To address this challenge, we propose Skeleton-in-Context (SiC), an effective framework for in-context skeleton sequence modeling. Our SiC is able to handle multiple skeleton-based tasks simultaneously after a single training process and accomplish each task from context according to the given prompt. It can further generalize to new, unseen tasks according to customized prompts. To facilitate context perception, we additionally propose a task-unified prompt, which adaptively learns tasks of different natures, such as partial joint-level generation, sequence-level prediction, or 2D-to-3D motion prediction. We conduct extensive experiments to evaluate the effectiveness of our SiC on multiple tasks, including motion prediction, pose estimation, joint completion, and future pose estimation. We also evaluate its generalization capability on unseen tasks such as motion-in-between. These experiments show that our model achieves state-of-the-art multi-task performance and even outperforms single-task methods on certain tasks.", "authors": ["Xinshun Wang", "Zhongbin Fang", "Xia Li", "Xiangtai Li", "Mengyuan Liu"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-12-06", "url": "https://arxiv.org/abs/2312.03703", "pdf_url": "https://arxiv.org/pdf/2312.03703v2", "arxiv_id": "2312.03703", "doi": "10.1109/CVPR52733.2024.00236", "citation_count": 33, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/fanglaosi/Skeleton-in-Context", "venue": "Computer Vision and Pattern Recognition", "quality_score": 0.3829}
{"id": "69591796ff9546386e79782b3df00f5f3f57b7ae7ace7cb2e73e4d66e1c70c25", "sources": ["arxiv", "semantic_scholar"], "title": "Adapt in Contexts: Retrieval-Augmented Domain Adaptation via In-Context Learning", "abstract": "Large language models (LLMs) have showcased their capability with few-shot inference known as in-context learning. However, in-domain demonstrations are not always readily available in real scenarios, leading to cross-domain in-context learning. Besides, LLMs are still facing challenges in long-tail knowledge in unseen and unfamiliar domains. The above limitations demonstrate the necessity of Unsupervised Domain Adaptation (UDA). In this paper, we study the UDA problem under an in-context learning setting to adapt language models from the source domain to the target domain without any target labels. The core idea is to retrieve a subset of cross-domain elements that are the most similar to the query, and elicit language model to adapt in an in-context manner by learning both target domain distribution and the discriminative task signal simultaneously with the augmented cross-domain in-context examples. We devise different prompting and training strategies, accounting for different LM architectures to learn the target distribution via language modeling. With extensive experiments on Sentiment Analysis (SA) and Named Entity Recognition (NER) tasks, we thoroughly study the effectiveness of ICL for domain transfer and demonstrate significant improvements over baseline models.", "authors": ["Quanyu Long", "Wenya Wang", "Sinno Jialin Pan"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-11-20", "url": "https://arxiv.org/abs/2311.11551", "pdf_url": "https://arxiv.org/pdf/2311.11551v1", "arxiv_id": "2311.11551", "doi": "10.48550/arXiv.2311.11551", "citation_count": 28, "influential_citation_count": 2, "has_code": false, "code_url": null, "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.3656}
{"id": "55eaa93f0c1534207b661bd8e62bdfef31ac3a3049e30dc7a8ca9b4a94b1cf10", "sources": ["arxiv", "semantic_scholar"], "title": "DocMath-Eval: Evaluating Math Reasoning Capabilities of LLMs in Understanding Long and Specialized Documents", "abstract": "Recent LLMs have demonstrated remarkable performance in solving exam-like math word problems. However, the degree to which these numerical reasoning skills are effective in real-world scenarios, particularly in expert domains, is still largely unexplored. This paper introduces DocMath-Eval, a comprehensive benchmark specifically designed to evaluate the numerical reasoning capabilities of LLMs in the context of understanding and analyzing specialized documents containing both text and tables. We conduct an extensive evaluation of 48 LLMs with Chain-of-Thought and Program-of-Thought prompting methods, aiming to comprehensively assess the capabilities and limitations of existing LLMs in DocMath-Eval. We found that even the current best-performing system (i.e., GPT-4o) still significantly lags behind human experts in solving complex numerical reasoning problems grounded in long contexts. We believe that DocMath-Eval can serve as a valuable benchmark for evaluating LLMs' capabilities in solving challenging numerical reasoning problems within expert domains.", "authors": ["Yilun Zhao", "Yitao Long", "Hongjun Liu", "Ryo Kamoi", "Linyong Nan", "Lyuhao Chen", "Yixin Liu", "Xiangru Tang", "Rui Zhang", "Arman Cohan"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-11-16", "url": "https://arxiv.org/abs/2311.09805", "pdf_url": "https://arxiv.org/pdf/2311.09805v3", "arxiv_id": "2311.09805", "doi": null, "citation_count": 53, "influential_citation_count": 5, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.4331}
{"id": "09749b1206516b3b2bec62a97b006a5e3128963588cafe3df1b16d12d1d4ffb3", "sources": ["arxiv", "semantic_scholar"], "title": "Thread of Thought Unraveling Chaotic Contexts", "abstract": "Large Language Models (LLMs) have ushered in a transformative era in the field of natural language processing, excelling in tasks related to text comprehension and generation. Nevertheless, they encounter difficulties when confronted with chaotic contexts (e.g., distractors rather than long irrelevant context), leading to the inadvertent omission of certain details within the chaotic context. In response to these challenges, we introduce the \"Thread of Thought\" (ThoT) strategy, which draws inspiration from human cognitive processes. ThoT systematically segments and analyzes extended contexts while adeptly selecting pertinent information. This strategy serves as a versatile \"plug-and-play\" module, seamlessly integrating with various LLMs and prompting techniques. In the experiments, we utilize the PopQA and EntityQ datasets, as well as a Multi-Turn Conversation Response dataset (MTCR) we collected, to illustrate that ThoT significantly improves reasoning performance compared to other prompting techniques.", "authors": ["Yucheng Zhou", "Xiubo Geng", "Tao Shen", "Chongyang Tao", "Guodong Long", "Jian-Guang Lou", "Jianbing Shen"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-11-15", "url": "https://arxiv.org/abs/2311.08734", "pdf_url": "https://arxiv.org/pdf/2311.08734v1", "arxiv_id": "2311.08734", "doi": "10.48550/arXiv.2311.08734", "citation_count": 84, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4824}
{"id": "0abb1c8725469299d95dff2c7717165fc4d6a33b507f3eab0e7b8e958a4da6bf", "sources": ["arxiv", "semantic_scholar"], "title": "LooGLE: Can Long-Context Language Models Understand Long Contexts?", "abstract": "Large language models (LLMs), despite their impressive performance in various language tasks, are typically limited to processing texts within context-window size. This limitation has spurred significant research efforts to enhance LLMs' long-context understanding with high-quality long-sequence benchmarks. However, prior datasets in this regard suffer from shortcomings, such as short context length compared to the context window of modern LLMs; outdated documents that have data leakage problems; and an emphasis on short dependency tasks rather than long dependency tasks. In this paper, we present LooGLE, a Long Context Generic Language Evaluation benchmark for LLMs' long context understanding. LooGLE features relatively new documents post-2022, with over 24,000 tokens per document and 6,000 newly generated questions spanning diverse domains. Human annotators meticulously crafted more than 1,100 high-quality question-answer pairs to meet the long dependency requirements. These pairs underwent thorough cross-validation, yielding the most precise assessment of LLMs' long dependency capabilities. The evaluation of eight state-of-the-art LLMs on LooGLE revealed key findings: (i) commercial models outperformed open-sourced models; (ii) LLMs excelled in short dependency tasks like short question-answering and cloze tasks but struggled with more intricate long dependency tasks; (iii) in-context learning and chaining thoughts offered only marginal improvements; (iv) retrieval-based techniques demonstrated substantial benefits for short question-answering, while strategies for extending context window length had limited impact on long context understanding. As such, LooGLE not only provides a systematic and comprehensive evaluation schema on long-context LLMs, but also sheds light on future development of enhanced models towards \"true long-context understanding\".", "authors": ["Jiaqi Li", "Mengmeng Wang", "Zilong Zheng", "Muhan Zhang"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-11-08", "url": "https://arxiv.org/abs/2311.04939", "pdf_url": "https://arxiv.org/pdf/2311.04939v2", "arxiv_id": "2311.04939", "doi": "10.48550/arXiv.2311.04939", "citation_count": 247, "influential_citation_count": 21, "has_code": true, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.6712}
{"id": "f49a89e2e5917c69e598f23cbc9fae572095e030dc75c900a92a83f662e85719", "sources": ["arxiv", "semantic_scholar"], "title": "Conservation law and Lie symmetry analysis of the (1+1) dimensional dispersive long-wave equation", "abstract": "In this paper, we mainly study the integrability of 1+1 dimensional dispersive long-wave equation. Firstly, the Lie symmetry analysis of the equation is carried out in the first part. And the optimal system of the equation is obtained according to the symmetry, and the invariant solution and the reduced form of the target equation are solved according to the results. Secondly, we use different methods to solve the conservation law of the target equation. To begin with, we give the adjoint determination equation and adjoint symmetry of the 1+1 dimensional dispersive long-wave equation, and use the adjoint symmetry as the equation multiplier to find several conservation laws. Then we get a Lie bracket by using the relationship between the symmetry of the equation and the adjoint symmetry. Next its strict self-adjoint property is verified, and its conservation laws are solved by Ibragimov's method. Finally, the conservation laws of the target equation are solved by Noether's theorem. Thirdly we calculate some exact solutions of the target equation by three different methods. In the end of the paper, the Hamiltonian structure of the target equation, the generalized pre-symplectic that maps symmetries into adjoint-symmetries and some of its soliton solutions are calculated. In conclusion, we use the direct construction of conservation law method, Ibragimov's method and so on to solve some new conservation laws of 1+1 dimensional dispersive long-wave equation, use the relationship between symmetry and adjoint symmetry to construct the corresponding Lie brackets, and obtain some linear soliton solutions according to the conservation law of the equation.", "authors": ["Long Ju", "Faiza Afzal", "Yufeng Zhang"], "categories": ["math-ph"], "fields_of_study": ["Physics", "Mathematics"], "published_date": "2023-10-16", "url": "https://arxiv.org/abs/2310.10426", "pdf_url": "https://arxiv.org/pdf/2310.10426v1", "arxiv_id": "2310.10426", "doi": null, "citation_count": 0, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.0}
{"id": "b9f33a1374394d57aac647cbb81eb4a9776f6226f87663c480b94f27d12fc9d2", "sources": ["arxiv", "semantic_scholar"], "title": "LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression", "abstract": "In long context scenarios, large language models (LLMs) face three main challenges: higher computational cost, performance reduction, and position bias. Research indicates that LLM performance hinges on the density and position of key information in the input prompt. Inspired by these findings, we propose LongLLMLingua for prompt compression towards improving LLMs' perception of the key information to simultaneously address the three challenges. Our extensive evaluation across various long context scenarios demonstrates that LongLLMLingua not only enhances performance but also significantly reduces costs and latency. For instance, in the NaturalQuestions benchmark, LongLLMLingua boosts performance by up to 21.4% with around 4x fewer tokens in GPT-3.5-Turbo, leading to substantial cost savings. It achieves a 94.0% cost reduction in the LooGLE benchmark. Moreover, when compressing prompts of about 10k tokens at ratios of 2x-6x, LongLLMLingua can accelerate end-to-end latency by 1.4x-2.6x. Our code is available at https://aka.ms/LongLLMLingua.", "authors": ["Huiqiang Jiang", "Qianhui Wu", "Xufang Luo", "Dongsheng Li", "Chin-Yew Lin", "Yuqing Yang", "Lili Qiu"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-10-10", "url": "https://arxiv.org/abs/2310.06839", "pdf_url": "https://arxiv.org/pdf/2310.06839v2", "arxiv_id": "2310.06839", "doi": "10.48550/arXiv.2310.06839", "citation_count": 459, "influential_citation_count": 62, "has_code": true, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.8997}
{"id": "b11158d4ddd25c04e3f53f22230754dc11455b1ca1c03a2393531f343b24fc0e", "sources": ["arxiv", "semantic_scholar"], "title": "Functional Interpolation for Relative Positions Improves Long Context Transformers", "abstract": "Preventing the performance decay of Transformers on inputs longer than those used for training has been an important challenge in extending the context length of these models. Though the Transformer architecture has fundamentally no limits on the input sequence lengths it can process, the choice of position encoding used during training can limit the performance of these models on longer inputs. We propose a novel functional relative position encoding with progressive interpolation, FIRE, to improve Transformer generalization to longer contexts. We theoretically prove that this can represent some of the popular relative position encodings, such as T5's RPE, Alibi, and Kerple. We next empirically show that FIRE models have better generalization to longer contexts on both zero-shot language modeling and long text benchmarks.", "authors": ["Shanda Li", "Chong You", "Guru Guruganesh", "Joshua Ainslie", "Santiago Ontanon", "Manzil Zaheer", "Sumit Sanghai", "Yiming Yang", "Sanjiv Kumar", "Srinadh Bhojanapalli"], "categories": ["cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-10-06", "url": "https://arxiv.org/abs/2310.04418", "pdf_url": "https://arxiv.org/pdf/2310.04418v2", "arxiv_id": "2310.04418", "doi": null, "citation_count": 17, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.3138}
{"id": "b28767500a736c34ded4ff2901de040ee64235c7e1dd96c8ec62ba32df3fd92e", "sources": ["arxiv", "semantic_scholar"], "title": "Retrieval meets Long Context Large Language Models", "abstract": "Extending the context window of large language models (LLMs) is getting popular recently, while the solution of augmenting LLMs with retrieval has existed for years. The natural questions are: i) Retrieval-augmentation versus long context window, which one is better for downstream tasks? ii) Can both methods be combined to get the best of both worlds? In this work, we answer these questions by studying both solutions using two state-of-the-art pretrained LLMs, i.e., a proprietary 43B GPT and Llama2-70B. Perhaps surprisingly, we find that LLM with 4K context window using simple retrieval-augmentation at generation can achieve comparable performance to finetuned LLM with 16K context window via positional interpolation on long context tasks, while taking much less computation. More importantly, we demonstrate that retrieval can significantly improve the performance of LLMs regardless of their extended context window sizes. Our best model, retrieval-augmented Llama2-70B with 32K context window, outperforms GPT-3.5-turbo-16k and Davinci003 in terms of average score on nine long context tasks including question answering, query-based summarization, and in-context few-shot learning tasks. It also outperforms its non-retrieval Llama2-70B-32k baseline by a margin, while being much faster at generation. Our study provides general insights on the choice of retrieval-augmentation versus long context extension of LLM for practitioners.", "authors": ["Peng Xu", "Wei Ping", "Xianchao Wu", "Lawrence McAfee", "Chen Zhu", "Zihan Liu", "Sandeep Subramanian", "Evelina Bakhturina", "Mohammad Shoeybi", "Bryan Catanzaro"], "categories": ["cs.CL", "cs.AI", "cs.IR", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-10-04", "url": "https://arxiv.org/abs/2310.03025", "pdf_url": "https://arxiv.org/pdf/2310.03025v2", "arxiv_id": "2310.03025", "doi": "10.48550/arXiv.2310.03025", "citation_count": 134, "influential_citation_count": 4, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.5326}
{"id": "41eaecf364b4eb0ed551c20da92ad2e0a87772f964c6877293983677da32ce94", "sources": ["arxiv", "semantic_scholar"], "title": "Effective Long-Context Scaling of Foundation Models", "abstract": "We present a series of long-context LLMs that support effective context windows of up to 32,768 tokens. Our model series are built through continual pretraining from Llama 2 with longer training sequences and on a dataset where long texts are upsampled. We perform extensive evaluation on language modeling, synthetic context probing tasks, and a wide range of research benchmarks. On research benchmarks, our models achieve consistent improvements on most regular tasks and significant improvements on long-context tasks over Llama 2. Notably, with a cost-effective instruction tuning procedure that does not require human-annotated long instruction data, the 70B variant can already surpass gpt-3.5-turbo-16k's overall performance on a suite of long-context tasks. Alongside these results, we provide an in-depth analysis on the individual components of our method. We delve into Llama's position encodings and discuss its limitation in modeling long dependencies. We also examine the impact of various design choices in the pretraining process, including the data mix and the training curriculum of sequence lengths -- our ablation experiments suggest that having abundant long texts in the pretrain dataset is not the key to achieving strong performance, and we empirically verify that long context continual pretraining is more efficient and similarly effective compared to pretraining from scratch with long sequences.", "authors": ["Wenhan Xiong", "Jingyu Liu", "Igor Molybog", "Hejia Zhang", "Prajjwal Bhargava", "Rui Hou", "Louis Martin", "Rashi Rungta", "Karthik Abinav Sankararaman", "Barlas Oguz", "Madian Khabsa", "Han Fang", "Yashar Mehdad", "Sharan Narang", "Kshitiz Malik", "Angela Fan", "Shruti Bhosale", "Sergey Edunov", "Mike Lewis", "Sinong Wang", "Hao Ma"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-09-27", "url": "https://arxiv.org/abs/2309.16039", "pdf_url": "https://arxiv.org/pdf/2309.16039v3", "arxiv_id": "2309.16039", "doi": "10.48550/arXiv.2309.16039", "citation_count": 342, "influential_citation_count": 22, "has_code": false, "code_url": null, "venue": "North American Chapter of the Association for Computational Linguistics", "quality_score": 0.6809}
{"id": "be703ff446ef85ac9294a6e7522478b1cbdaa7d91393253f6811e503928f1739", "sources": ["arxiv", "semantic_scholar"], "title": "LongLoRA: Efficient Fine-tuning of Long-Context Large Language Models", "abstract": "We present LongLoRA, an efficient fine-tuning approach that extends the context sizes of pre-trained large language models (LLMs), with limited computation cost. Typically, training LLMs with long context sizes is computationally expensive, requiring extensive training hours and GPU resources. For example, training on the context length of 8192 needs 16x computational costs in self-attention layers as that of 2048. In this paper, we speed up the context extension of LLMs in two aspects. On the one hand, although dense global attention is needed during inference, fine-tuning the model can be effectively and efficiently done by sparse local attention. The proposed shifted sparse attention effectively enables context extension, leading to non-trivial computation saving with similar performance to fine-tuning with vanilla attention. Particularly, it can be implemented with only two lines of code in training, while being optional in inference. On the other hand, we revisit the parameter-efficient fine-tuning regime for context expansion. Notably, we find that LoRA for context extension works well under the premise of trainable embedding and normalization. LongLoRA combines this improved LoRA with S^2-Attn. LongLoRA demonstrates strong empirical results on various tasks on Llama2 models from 7B/13B to 70B. LongLoRA extends Llama2 7B from 4k context to 100k, or Llama2 70B to 32k on a single 8x A100 machine. LongLoRA extends models' context while retaining their original architectures, and is compatible with most existing techniques, like Flash-Attention2. In addition, we further conduct supervised fine-tuning with LongLoRA and our long instruction-following LongAlpaca dataset.", "authors": ["Yukang Chen", "Shengju Qian", "Haotian Tang", "Xin Lai", "Zhijian Liu", "Song Han", "Jiaya Jia"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-09-21", "url": "https://arxiv.org/abs/2309.12307", "pdf_url": "https://arxiv.org/pdf/2309.12307v3", "arxiv_id": "2309.12307", "doi": "10.48550/arXiv.2309.12307", "citation_count": 269, "influential_citation_count": 22, "has_code": true, "code_url": "https://github.com/dvlab-research/LongLoRA", "venue": "International Conference on Learning Representations", "quality_score": 0.6809}
{"id": "6911fce80553467081d5195ba0b74aa9fc1b2304a54968ffcf80cafc5cf07eaa", "sources": ["arxiv", "semantic_scholar"], "title": "PoSE: Efficient Context Window Extension of LLMs via Positional Skip-wise Training", "abstract": "Large Language Models (LLMs) are trained with a pre-defined context length, restricting their use in scenarios requiring long inputs. Previous efforts for adapting LLMs to a longer length usually requires fine-tuning with this target length (Full-length fine-tuning), suffering intensive training cost. To decouple train length from target length for efficient context window extension, we propose Positional Skip-wisE (PoSE) training that smartly simulates long inputs using a fixed context window. This is achieved by first dividing the original context window into several chunks, then designing distinct skipping bias terms to manipulate the position indices of each chunk. These bias terms and the lengths of each chunk are altered for every training example, allowing the model to adapt to all positions within target length. Experimental results show that PoSE greatly reduces memory and time overhead compared with Full-length fine-tuning, with minimal impact on performance. Leveraging this advantage, we have successfully extended the LLaMA model to 128k tokens using a 2k training context window. Furthermore, we empirically confirm that PoSE is compatible with all RoPE-based LLMs and position interpolation strategies. Notably, our method can potentially support infinite length, limited only by memory usage in inference. With ongoing progress for efficient inference, we believe PoSE can further scale the context window beyond 128k.", "authors": ["Dawei Zhu", "Nan Yang", "Liang Wang", "Yifan Song", "Wenhao Wu", "Furu Wei", "Sujian Li"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-09-19", "url": "https://arxiv.org/abs/2309.10400", "pdf_url": "https://arxiv.org/pdf/2309.10400v3", "arxiv_id": "2309.10400", "doi": "10.48550/arXiv.2309.10400", "citation_count": 112, "influential_citation_count": 6, "has_code": false, "code_url": null, "venue": "International Conference on Learning Representations", "quality_score": 0.5133}
{"id": "ec8c74e372d3a829fd505a645226f8d71a538ce96816e8381eb5112431d7aa5f", "sources": ["arxiv", "semantic_scholar"], "title": "CoCA: Fusing Position Embedding with Collinear Constrained Attention in Transformers for Long Context Window Extending", "abstract": "Self-attention and position embedding are two key modules in transformer-based Large Language Models (LLMs). However, the potential relationship between them is far from well studied, especially for long context window extending. In fact, anomalous behaviors harming long context extrapolation exist between Rotary Position Embedding (RoPE) and vanilla self-attention unveiled by our work. To address this issue, we propose a novel attention mechanism, CoCA (Collinear Constrained Attention). Specifically, we enforce a collinear constraint between $Q$ and $K$ to seamlessly integrate RoPE and self-attention. While only adding minimal computational and spatial complexity, this integration significantly enhances long context window extrapolation ability. We provide an optimized implementation, making it a drop-in replacement for any existing transformer-based models. Extensive experiments show that CoCA performs extraordinarily well in extending context windows. A CoCA-based GPT model, trained with a context length of 512, can seamlessly extend the context window up to 32K (60$\\times$), without any fine-tuning. Additionally, by dropping CoCA in LLaMA-7B, we achieve extrapolation up to 32K within only 2K training length. Our code is publicly available at: https://github.com/codefuse-ai/Collinear-Constrained-Attention", "authors": ["Shiyi Zhu", "Jing Ye", "Wei Jiang", "Siqiao Xue", "Qi Zhang", "Yifan Wu", "Jianguo Li"], "categories": ["cs.LG", "cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-09-15", "url": "https://arxiv.org/abs/2309.08646", "pdf_url": "https://arxiv.org/pdf/2309.08646v3", "arxiv_id": "2309.08646", "doi": "10.18653/v1/2024.acl-long.233", "citation_count": 9, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/codefuse-ai/Collinear-Constrained-Attention", "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.25}
{"id": "637fe0a8f92246271266421795cb9bb374ac43e7261d5a11dca9d91571e87edc", "sources": ["arxiv", "semantic_scholar"], "title": "Support-Set Context Matters for Bongard Problems", "abstract": "Current machine learning methods struggle to solve Bongard problems, which are a type of IQ test that requires deriving an abstract \"concept\" from a set of positive and negative \"support\" images, and then classifying whether or not a new query image depicts the key concept. On Bongard-HOI, a benchmark for natural-image Bongard problems, most existing methods have reached at best 69% accuracy (where chance is 50%). Low accuracy is often attributed to neural nets' lack of ability to find human-like symbolic rules. In this work, we point out that many existing methods are forfeiting accuracy due to a much simpler problem: they do not adapt image features given information contained in the support set as a whole, and rely instead on information extracted from individual supports. This is a critical issue, because the \"key concept\" in a typical Bongard problem can often only be distinguished using multiple positives and multiple negatives. We explore simple methods to incorporate this context and show substantial gains over prior works, leading to new state-of-the-art accuracy on Bongard-LOGO (75.3%) and Bongard-HOI (76.4%) compared to methods with equivalent vision backbone architectures and strong performance on the original Bongard problem set (60.8%).", "authors": ["Nikhil Raghuraman", "Adam W. Harley", "Leonidas Guibas"], "categories": ["cs.CV", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-09-07", "url": "https://arxiv.org/abs/2309.03468", "pdf_url": "https://arxiv.org/pdf/2309.03468v2", "arxiv_id": "2309.03468", "doi": null, "citation_count": 2, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/nraghuraman/bongard-context", "venue": null, "quality_score": 0.1193}
{"id": "242fa762c453888cf8e343508c4522dd0e5ffadc7ce7c0a4ea3258e899ccac16", "sources": ["arxiv", "semantic_scholar"], "title": "YaRN: Efficient Context Window Extension of Large Language Models", "abstract": "Rotary Position Embeddings (RoPE) have been shown to effectively encode positional information in transformer-based language models. However, these models fail to generalize past the sequence length they were trained on. We present YaRN (Yet another RoPE extensioN method), a compute-efficient method to extend the context window of such models, requiring 10x less tokens and 2.5x less training steps than previous methods. Using YaRN, we show that LLaMA models can effectively utilize and extrapolate to context lengths much longer than their original pre-training would allow, while also surpassing previous the state-of-the-art at context window extension. In addition, we demonstrate that YaRN exhibits the capability to extrapolate beyond the limited context of a fine-tuning dataset. Code is available at https://github.com/jquesnelle/yarn", "authors": ["Bowen Peng", "Jeffrey Quesnelle", "Honglu Fan", "Enrico Shippole"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-31", "url": "https://arxiv.org/abs/2309.00071", "pdf_url": "https://arxiv.org/pdf/2309.00071v3", "arxiv_id": "2309.00071", "doi": "10.48550/arXiv.2309.00071", "citation_count": 562, "influential_citation_count": 60, "has_code": true, "code_url": "https://github.com/jquesnelle/yarn", "venue": "International Conference on Learning Representations", "quality_score": 0.8927}
{"id": "4e27c5be75d3ebe6a0e172255426e73f95d73aeba14bb6466d48b7f83396e397", "sources": ["arxiv", "semantic_scholar"], "title": "LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding", "abstract": "Although large language models (LLMs) demonstrate impressive performance for many language tasks, most of them can only handle texts a few thousand tokens long, limiting their applications on longer sequence inputs, such as books, reports, and codebases. Recent works have proposed methods to improve LLMs' long context capabilities by extending context windows and more sophisticated memory mechanisms. However, comprehensive benchmarks tailored for evaluating long context understanding are lacking. In this paper, we introduce LongBench, the first bilingual, multi-task benchmark for long context understanding, enabling a more rigorous evaluation of long context understanding. LongBench comprises 21 datasets across 6 task categories in both English and Chinese, with an average length of 6,711 words (English) and 13,386 characters (Chinese). These tasks cover key long-text application areas including single-doc QA, multi-doc QA, summarization, few-shot learning, synthetic tasks, and code completion. All datasets in LongBench are standardized into a unified format, allowing for effortless automatic evaluation of LLMs. Upon comprehensive evaluation of 8 LLMs on LongBench, we find that: (1) Commercial model (GPT-3.5-Turbo-16k) outperforms other open-sourced models, but still struggles on longer contexts. (2) Scaled position embedding and fine-tuning on longer sequences lead to substantial improvement on long context understanding. (3) Context compression technique such as retrieval brings improvement for model with weak ability on long contexts, but the performance still lags behind models that have strong long context understanding capability. The code and datasets are available at https://github.com/THUDM/LongBench.", "authors": ["Yushi Bai", "Xin Lv", "Jiajie Zhang", "Hongchang Lyu", "Jiankai Tang", "Zhidian Huang", "Zhengxiao Du", "Xiao Liu", "Aohan Zeng", "Lei Hou", "Yuxiao Dong", "Jie Tang", "Juanzi Li"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-28", "url": "https://arxiv.org/abs/2308.14508", "pdf_url": "https://arxiv.org/pdf/2308.14508v2", "arxiv_id": "2308.14508", "doi": "10.48550/arXiv.2308.14508", "citation_count": 1363, "influential_citation_count": 335, "has_code": true, "code_url": "https://github.com/THUDM/LongBench", "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 1.0}
{"id": "57e21f53c26e006a975dd8e0711ae7240f3f163f1fbe83e640ccd356ec9f8464", "sources": ["arxiv", "semantic_scholar"], "title": "How Much Temporal Long-Term Context is Needed for Action Segmentation?", "abstract": "Modeling long-term context in videos is crucial for many fine-grained tasks including temporal action segmentation. An interesting question that is still open is how much long-term temporal context is needed for optimal performance. While transformers can model the long-term context of a video, this becomes computationally prohibitive for long videos. Recent works on temporal action segmentation thus combine temporal convolutional networks with self-attentions that are computed only for a local temporal window. While these approaches show good results, their performance is limited by their inability to capture the full context of a video. In this work, we try to answer how much long-term temporal context is required for temporal action segmentation by introducing a transformer-based model that leverages sparse attention to capture the full context of a video. We compare our model with the current state of the art on three datasets for temporal action segmentation, namely 50Salads, Breakfast, and Assembly101. Our experiments show that modeling the full context of a video is necessary to obtain the best performance for temporal action segmentation.", "authors": ["Emad Bahrami", "Gianpiero Francesca", "Juergen Gall"], "categories": ["cs.CV", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-22", "url": "https://arxiv.org/abs/2308.11358", "pdf_url": "https://arxiv.org/pdf/2308.11358v2", "arxiv_id": "2308.11358", "doi": "10.1109/ICCV51070.2023.00950", "citation_count": 61, "influential_citation_count": 10, "has_code": false, "code_url": null, "venue": "IEEE International Conference on Computer Vision", "quality_score": 0.5207}
{"id": "e4c958588999273d2b7de5cfc7b7fbbe7761596dac20edc8329616b862db46cb", "sources": ["arxiv", "semantic_scholar"], "title": "CSM-H-R: A Context Modeling Framework in Supporting Reasoning Automation for Interoperable Intelligent Systems and Privacy Protection", "abstract": "The automation of High-Level Context (HLC) reasoning across intelligent systems at scale is imperative because of the unceasing accumulation of contextual data, the trend of the fusion of data from multiple sources (e.g., sensors, intelligent systems), and the intrinsic complexity and dynamism of context-based decision-making processes. To mitigate the challenges posed by these issues, we propose a novel Hierarchical Ontology-State Modeling (HOSM) framework CSM-H-R, which programmatically combines ontologies and states at the modeling phase and runtime phase for attaining the ability to recognize meaningful HLC. It builds on the model of our prior work on the Context State Machine (CSM) engine by incorporating the H (Hierarchy) and R (Relationship and tRansition) dimensions to take care of the dynamic aspects of context. The design of the framework supports the sharing and interoperation of context among intelligent systems and the components for handling CSMs and the management of hierarchy, relationship, and transition. Case studies are developed for IntellElevator and IntellRestaurant, two intelligent applications in a smart campus setting. The prototype implementation of the framework experiments on translating the HLC reasoning into vector and matrix computing and presents the potential of using advanced probabilistic models to reach the next level of automation in integrating intelligent systems; meanwhile, privacy protection support is achieved in the application domain by anonymization through indexing and reducing information correlation. An implementation of the framework is available at https://github.com/songhui01/CSM-H-R.", "authors": ["Songhui Yue", "Xiaoyan Hong", "Randy K. Smith"], "categories": ["cs.AI", "eess.SY"], "fields_of_study": ["Computer Science", "Engineering"], "published_date": "2023-08-21", "url": "https://arxiv.org/abs/2308.11066", "pdf_url": "https://arxiv.org/pdf/2308.11066v3", "arxiv_id": "2308.11066", "doi": "10.1109/ACCESS.2024.3446274", "citation_count": 1, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/songhui01/CSM-H-R", "venue": "IEEE Access", "quality_score": 0.0753}
{"id": "015ae60ae0bba0a5d7b08ba63beec15f84ff145ee13fd30f50ccee9887181b20", "sources": ["arxiv", "semantic_scholar"], "title": "Giraffe: Adventures in Expanding Context Lengths in LLMs", "abstract": "Modern large language models (LLMs) that rely on attention mechanisms are typically trained with fixed context lengths which enforce upper limits on the length of input sequences that they can handle at evaluation time. To use these models on sequences longer than the train-time context length, one might employ techniques from the growing family of context length extrapolation methods -- most of which focus on modifying the system of positional encodings used in the attention mechanism to indicate where tokens or activations are located in the input sequence. We conduct a wide survey of existing methods of context length extrapolation on a base LLaMA or LLaMA 2 model, and introduce some of our own design as well -- in particular, a new truncation strategy for modifying the basis for the position encoding. We test these methods using three new evaluation tasks (FreeFormQA, AlteredNumericQA, and LongChat-Lines) as well as perplexity, which we find to be less fine-grained as a measure of long context performance of LLMs. We release the three tasks publicly as datasets on HuggingFace. We discover that linear scaling is the best method for extending context length, and show that further gains can be achieved by using longer scales at evaluation time. We also discover promising extrapolation capabilities in the truncated basis. To support further research in this area, we release three new 13B parameter long-context models which we call Giraffe: 4k and 16k context models trained from base LLaMA-13B, and a 32k context model trained from base LLaMA2-13B. We also release the code to replicate our results.", "authors": ["Arka Pal", "Deep Karkhanis", "Manley Roberts", "Samuel Dooley", "Arvind Sundararajan", "Siddartha Naidu"], "categories": ["cs.AI", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-21", "url": "https://arxiv.org/abs/2308.10882", "pdf_url": "https://arxiv.org/pdf/2308.10882v1", "arxiv_id": "2308.10882", "doi": "10.48550/arXiv.2308.10882", "citation_count": 45, "influential_citation_count": 3, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.4157}
{"id": "d1c12249a0917fa8bfb6d93e0e27f9edef9cc7a3c3f64c91d8f49dc1c498be89", "sources": ["arxiv", "semantic_scholar"], "title": "A Case Study on Context Encoding in Multi-Encoder based Document-Level Neural Machine Translation", "abstract": "Recent studies have shown that the multi-encoder models are agnostic to the choice of context, and the context encoder generates noise which helps improve the models in terms of BLEU score. In this paper, we further explore this idea by evaluating with context-aware pronoun translation test set by training multi-encoder models trained on three different context settings viz, previous two sentences, random two sentences, and a mix of both as context. Specifically, we evaluate the models on the ContraPro test set to study how different contexts affect pronoun translation accuracy. The results show that the model can perform well on the ContraPro test set even when the context is random. We also analyze the source representations to study whether the context encoder generates noise. Our analysis shows that the context encoder provides sufficient information to learn discourse-level information. Additionally, we observe that mixing the selected context (the previous two sentences in this case) and the random context is generally better than the other settings.", "authors": ["Ramakrishna Appicharla", "Baban Gain", "Santanu Pal", "Asif Ekbal"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-11", "url": "https://arxiv.org/abs/2308.06063", "pdf_url": "https://arxiv.org/pdf/2308.06063v1", "arxiv_id": "2308.06063", "doi": "10.48550/arXiv.2308.06063", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Machine Translation Summit", "quality_score": 0.0753}
{"id": "8ae78c50c83b69057228cab9ce8ff90fc724e95ab3abf70b8f3f3ed85b307adf", "sources": ["arxiv", "semantic_scholar"], "title": "DegUIL: Degree-aware Graph Neural Networks for Long-tailed User Identity Linkage", "abstract": "User identity linkage (UIL), matching accounts of a person on different social networks, is a fundamental task in cross-network data mining. Recent works have achieved promising results by exploiting graph neural networks (GNNs) to capture network structure. However, they rarely analyze the realistic node-level bottlenecks that hinder UIL's performance. First, node degrees in a graph vary widely and are long-tailed. A significant fraction of tail nodes with small degrees are underrepresented due to limited structural information, degrading linkage performance seriously. The second bottleneck usually overlooked is super head nodes. It is commonly accepted that head nodes perform well. However, we find that some of them with super high degrees also have difficulty aligning counterparts, due to noise introduced by the randomness of following friends in real-world social graphs. In pursuit of learning ideal representations for these two groups of nodes, this paper proposes a degree-aware model named DegUIL to narrow the degree gap. To this end, our model complements missing neighborhoods for tail nodes and discards redundant structural information for super head nodes in embeddings respectively. Specifically, the neighboring bias is predicted and corrected locally by two modules, which are trained using the knowledge from structurally adequate head nodes. As a result, ideal neighborhoods are obtained for meaningful aggregation in GNNs. Extensive experiments demonstrate the superiority of our model. Our data and code can be found at https://github.com/Longmeix/DegUIL.", "authors": ["Meixiu Long", "Siyuan Chen", "Xin Du", "Jiahai Wang"], "categories": ["cs.SI"], "fields_of_study": ["Computer Science"], "published_date": "2023-08-10", "url": "https://arxiv.org/abs/2308.05322", "pdf_url": "https://arxiv.org/pdf/2308.05322v1", "arxiv_id": "2308.05322", "doi": "10.48550/arXiv.2308.05322", "citation_count": 11, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/Longmeix/DegUIL", "venue": null, "quality_score": 0.2698}
{"id": "e8a9a8ae6064cbb933a87c02dbf2697ba75955a9cee47cbcef9035aac8eb954c", "sources": ["arxiv", "semantic_scholar"], "title": "Pay Attention to What You Need", "abstract": "Although large language models (LLMs) have achieved significant success in natural language processing, they still struggle with long-context comprehension. Traditional approaches to mitigating this issue typically rely on fine-tuning or retraining, which is both resource-intensive and challenging to deploy in lightweight industrial settings. In this paper, we investigate the potential to accomplish this without any additional resources. Through an in-depth study of the attention mechanism in LLMs, we propose a method called Scaled ReAttention (SRA) to strengthen LLMs' ability to interpret and retrieve information by strategically manipulating their attention scores during inference. Through extensive experiments, we demonstrate that integrating SRA significantly boosts LLMs' performance on a variety of downstream tasks, highlighting its practical potential for enhancing language understanding without incurring the overhead of traditional training.", "authors": ["Yifei Gao", "Shaohong Chen", "Lei Wang", "Ruiting Dai", "Ziyun Zhang", "Kerui Ren", "Jiaji Wu", "Jun Cheng"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-07-25", "url": "https://arxiv.org/abs/2307.13365", "pdf_url": "https://arxiv.org/pdf/2307.13365v3", "arxiv_id": "2307.13365", "doi": null, "citation_count": 3, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1505}
{"id": "d2372accc58f8b92d6f64e8bb85be00dd9b65f72509b65a56ec2b757da0ba028", "sources": ["arxiv", "semantic_scholar"], "title": "Extending Context Window of Large Language Models via Positional Interpolation", "abstract": "We present Position Interpolation (PI) that extends the context window sizes of RoPE-based pretrained LLMs such as LLaMA models to up to 32768 with minimal fine-tuning (within 1000 steps), while demonstrating strong empirical results on various tasks that require long context, including passkey retrieval, language modeling, and long document summarization from LLaMA 7B to 65B. Meanwhile, the extended model by Position Interpolation preserve quality relatively well on tasks within its original context window. To achieve this goal, Position Interpolation linearly down-scales the input position indices to match the original context window size, rather than extrapolating beyond the trained context length which may lead to catastrophically high attention scores that completely ruin the self-attention mechanism. Our theoretical study shows that the upper bound of interpolation is at least $\\sim 600 \\times$ smaller than that of extrapolation, further demonstrating its stability. Models extended via Position Interpolation retain its original architecture and can reuse most pre-existing optimization and infrastructure.", "authors": ["Shouyuan Chen", "Sherman Wong", "Liangjian Chen", "Yuandong Tian"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2023-06-27", "url": "https://arxiv.org/abs/2306.15595", "pdf_url": "https://arxiv.org/pdf/2306.15595v2", "arxiv_id": "2306.15595", "doi": "10.48550/arXiv.2306.15595", "citation_count": 783, "influential_citation_count": 82, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.9595}
{"id": "1fd4d6b492a6ac3c0782a739ad80954ecd2edb4ea8b3cf2bc6b7954b23e606ba", "sources": ["arxiv", "semantic_scholar"], "title": "Guiding Language Models of Code with Global Context using Monitors", "abstract": "Language models of code (LMs) work well when the surrounding code provides sufficient context. This is not true when it becomes necessary to use types, functionality or APIs defined elsewhere in the repository or a linked library, especially those not seen during training. LMs suffer from limited awareness of such global context and end up hallucinating. Integrated development environments (IDEs) assist developers in understanding repository context using static analysis. We extend this assistance, enjoyed by developers, to LMs. We propose monitor-guided decoding (MGD) where a monitor uses static analysis to guide the decoding. We construct a repository-level dataset PragmaticCode for method-completion in Java and evaluate MGD on it. On models of varying parameter scale, by monitoring for type-consistent object dereferences, MGD consistently improves compilation rates and agreement with ground truth. Further, LMs with fewer parameters, when augmented with MGD, can outperform larger LMs. With MGD, SantaCoder-1.1B achieves better compilation rate and next-identifier match than the much larger text-davinci-003 model. We also conduct a generalizability study to evaluate the ability of MGD to generalize to multiple programming languages (Java, C# and Rust), coding scenarios (e.g., correct number of arguments to method calls), and to enforce richer semantic constraints (e.g., stateful API protocols). Our data and implementation are available at https://github.com/microsoft/monitors4codegen .", "authors": ["Lakshya A Agrawal", "Aditya Kanade", "Navin Goyal", "Shuvendu K. Lahiri", "Sriram K. Rajamani"], "categories": ["cs.CL", "cs.AI", "cs.LG", "cs.PL", "cs.SE"], "fields_of_study": ["Computer Science"], "published_date": "2023-06-19", "url": "https://arxiv.org/abs/2306.10763", "pdf_url": "https://arxiv.org/pdf/2306.10763v2", "arxiv_id": "2306.10763", "doi": "10.48550/arXiv.2306.10763", "citation_count": 35, "influential_citation_count": 4, "has_code": true, "code_url": "https://github.com/microsoft/monitors4codegen", "venue": "arXiv.org", "quality_score": 0.3891}
{"id": "b4b6f7a6eea43342fe6b6325d0ed576d114be2f3854d8c2c9085a6ac1c1b66fa", "sources": ["arxiv", "semantic_scholar"], "title": "Explore In-Context Learning for 3D Point Cloud Understanding", "abstract": "With the rise of large-scale models trained on broad data, in-context learning has become a new learning paradigm that has demonstrated significant potential in natural language processing and computer vision tasks. Meanwhile, in-context learning is still largely unexplored in the 3D point cloud domain. Although masked modeling has been successfully applied for in-context learning in 2D vision, directly extending it to 3D point clouds remains a formidable challenge. In the case of point clouds, the tokens themselves are the point cloud positions (coordinates) that are masked during inference. Moreover, position embedding in previous works may inadvertently introduce information leakage. To address these challenges, we introduce a novel framework, named Point-In-Context, designed especially for in-context learning in 3D point clouds, where both inputs and outputs are modeled as coordinates for each task. Additionally, we propose the Joint Sampling module, carefully designed to work in tandem with the general point sampling operator, effectively resolving the aforementioned technical issues. We conduct extensive experiments to validate the versatility and adaptability of our proposed methods in handling a wide range of tasks.", "authors": ["Zhongbin Fang", "Xiangtai Li", "Xia Li", "Joachim M. Buhmann", "Chen Change Loy", "Mengyuan Liu"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-06-14", "url": "https://arxiv.org/abs/2306.08659", "pdf_url": "https://arxiv.org/pdf/2306.08659v2", "arxiv_id": "2306.08659", "doi": "10.48550/arXiv.2306.08659", "citation_count": 46, "influential_citation_count": 6, "has_code": true, "code_url": "https://github.com/fanglaosi/Point-In-Context", "venue": "Neural Information Processing Systems", "quality_score": 0.4225}
{"id": "c588f313a1330151fa10620ca260c411a681c8369acb58efc461ab292c20f23b", "sources": ["arxiv", "semantic_scholar"], "title": "Context-PIPs: Persistent Independent Particles Demands Spatial Context Features", "abstract": "We tackle the problem of Persistent Independent Particles (PIPs), also called Tracking Any Point (TAP), in videos, which specifically aims at estimating persistent long-term trajectories of query points in videos. Previous methods attempted to estimate these trajectories independently to incorporate longer image sequences, therefore, ignoring the potential benefits of incorporating spatial context features. We argue that independent video point tracking also demands spatial context features. To this end, we propose a novel framework Context-PIPs, which effectively improves point trajectory accuracy by aggregating spatial context features in videos. Context-PIPs contains two main modules: 1) a SOurse Feature Enhancement (SOFE) module, and 2) a TArget Feature Aggregation (TAFA) module. Context-PIPs significantly improves PIPs all-sided, reducing 11.4% Average Trajectory Error of Occluded Points (ATE-Occ) on CroHD and increasing 11.8% Average Percentage of Correct Keypoint (A-PCK) on TAP-Vid-Kinectics. Demos are available at https://wkbian.github.io/Projects/Context-PIPs/.", "authors": ["Weikang Bian", "Zhaoyang Huang", "Xiaoyu Shi", "Yitong Dong", "Yijin Li", "Hongsheng Li"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-06-03", "url": "https://arxiv.org/abs/2306.02000", "pdf_url": "https://arxiv.org/pdf/2306.02000v2", "arxiv_id": "2306.02000", "doi": "10.52202/075280-2413", "citation_count": 11, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.2698}
{"id": "f6fb9350adc71c514555c39dc66d20d724b2f5d3df265590d9ec6eba30942772", "sources": ["arxiv", "semantic_scholar"], "title": "Stecformer: Spatio-temporal Encoding Cascaded Transformer for Multivariate Long-term Time Series Forecasting", "abstract": "Multivariate long-term time series forecasting is of great application across many domains, such as energy consumption and weather forecasting. With the development of transformer-based methods, the performance of multivariate long-term time series forecasting has been significantly improved, however, the study of spatial features extracting in transformer-based model is rare and the consistency of different prediction periods is unsatisfactory due to the large span. In this work, we propose a complete solution to address these problems in terms of feature extraction and target prediction. For extraction, we design an efficient spatio-temporal encoding extractor including a semi-adaptive graph to acquire sufficient spatio-temporal information. For prediction, we propose a Cascaded Decoding Predictor (CDP) to strengthen the correlation between different intervals, which can also be utilized as a generic component to improve the performance of transformer-based methods. The proposed method, termed as Spatio-temporal Encoding Cascaded Transformer (Stecformer), achieving a notable gap over the baseline model and is comparable with the state-of-the-art performance of transformer-based methods on five benchmark datasets. We hope our attempt will serve as a regular configuration in multivariate long-term time series forecasting in the future.", "authors": ["Zheng Sun", "Yi Wei", "Wenxiao Jia", "Long Yu"], "categories": ["cs.LG", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-25", "url": "https://arxiv.org/abs/2305.16370", "pdf_url": "https://arxiv.org/pdf/2305.16370v1", "arxiv_id": "2305.16370", "doi": "10.48550/arXiv.2305.16370", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.0753}
{"id": "4eb5330b5c8f40b7c66f75b33fa094b48d3c8da55bf66c5402964eeeccd7302f", "sources": ["arxiv", "semantic_scholar"], "title": "Cross-lingual QA: A Key to Unlocking In-context Cross-lingual Performance", "abstract": "Multilingual large language models (MLLMs) have demonstrated significant cross-lingual capabilities through in-context learning. Existing approaches typically construct monolingual in-context examples, either in the source or target language. However, translating entire in-context examples into the target language might compromise contextual integrity and be costly in the case of long-context passages. To address this, we introduce Cross-lingual QA, a cross-lingual prompting method that translates only the question and answer parts, thus reducing translation costs. Experiments on four typologically diverse multilingual benchmarks show that Cross-lingual QA prompting effectively stimulates models to elicit their cross-lingual knowledge, outperforming prior monolingual prompting approaches. Furthermore, we show that prompting open-source MLLMs with cross-lingual in-context examples enhances performance as the model scale increases.", "authors": ["Sunkyoung Kim", "Dayeon Ki", "Yireun Kim", "Jinsik Lee"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-24", "url": "https://arxiv.org/abs/2305.15233", "pdf_url": "https://arxiv.org/pdf/2305.15233v3", "arxiv_id": "2305.15233", "doi": null, "citation_count": 5, "influential_citation_count": 0, "has_code": true, "code_url": null, "venue": null, "quality_score": 0.1945}
{"id": "6fb547dfae112451214963889712dab68a9d26c562395c7856f19fc0aed8cc48", "sources": ["arxiv", "semantic_scholar"], "title": "Revisiting Parallel Context Windows: A Frustratingly Simple Alternative and Chain-of-Thought Deterioration", "abstract": "We identify two crucial limitations in the evaluation of recent parallel-integrated method Parallel Context Windows (PCW), which extends the maximum context lengths of language models, e.g., 2048 for LLaMA, by harnessing window-wise attention and positional embedding techniques. We first show that a simple yet strong baseline, weighted sum ensemble, is missing for the in-context few-shot classification. Moreover, on more challenging Chain-of-Thought (CoT) reasoning (e.g., HotpotQA), PCW would present unexpected deterioration regarding question miscomprehension and false inference. Based on our findings, we suggest that the existing PCW design may not guarantee sufficient improvement and practicality in handling lengthy documents in real-world applications. More community efforts on enabling language models' long context understanding ability should be paid.", "authors": ["Kejuan Yang", "Xiao Liu", "Kaiwen Men", "Aohan Zeng", "Yuxiao Dong", "Jie Tang"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-24", "url": "https://arxiv.org/abs/2305.15262", "pdf_url": "https://arxiv.org/pdf/2305.15262v1", "arxiv_id": "2305.15262", "doi": "10.48550/arXiv.2305.15262", "citation_count": 5, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.1945}
{"id": "7d72efe248f17fcb8fe7cfaa391b49a411bc6f33e8658eea86f75e11a1f0729b", "sources": ["arxiv", "semantic_scholar"], "title": "Multi-Level Global Context Cross Consistency Model for Semi-Supervised Ultrasound Image Segmentation with Diffusion Model", "abstract": "Medical image segmentation is a critical step in computer-aided diagnosis, and convolutional neural networks are popular segmentation networks nowadays. However, the inherent local operation characteristics make it difficult to focus on the global contextual information of lesions with different positions, shapes, and sizes. Semi-supervised learning can be used to learn from both labeled and unlabeled samples, alleviating the burden of manual labeling. However, obtaining a large number of unlabeled images in medical scenarios remains challenging. To address these issues, we propose a Multi-level Global Context Cross-consistency (MGCC) framework that uses images generated by a Latent Diffusion Model (LDM) as unlabeled images for semi-supervised learning. The framework involves of two stages. In the first stage, a LDM is used to generate synthetic medical images, which reduces the workload of data annotation and addresses privacy concerns associated with collecting medical data. In the second stage, varying levels of global context noise perturbation are added to the input of the auxiliary decoder, and output consistency is maintained between decoders to improve the representation ability. Experiments conducted on open-source breast ultrasound and private thyroid ultrasound datasets demonstrate the effectiveness of our framework in bridging the probability distribution and the semantic representation of the medical image. Our approach enables the effective transfer of probability distribution knowledge to the segmentation network, resulting in improved segmentation accuracy. The code is available at https://github.com/FengheTan9/Multi-Level-Global-Context-Cross-Consistency.", "authors": ["Fenghe Tang", "Jianrui Ding", "Lingtao Wang", "Min Xian", "Chunping Ning"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-16", "url": "https://arxiv.org/abs/2305.09447", "pdf_url": "https://arxiv.org/pdf/2305.09447v2", "arxiv_id": "2305.09447", "doi": "10.48550/arXiv.2305.09447", "citation_count": 22, "influential_citation_count": 0, "has_code": true, "code_url": "https://github.com/FengheTan9/Multi-Level-Global-Context-Cross-Consistency", "venue": "arXiv.org", "quality_score": 0.3404}
{"id": "149b52e44a3f23d8642b493c763e90608348b1a8d790aa2469e09e8ad925b056", "sources": ["arxiv", "semantic_scholar"], "title": "AxWin Transformer: A Context-Aware Vision Transformer Backbone with Axial Windows", "abstract": "Recently Transformer has shown good performance in several vision tasks due to its powerful modeling capabilities. To reduce the quadratic complexity caused by the attention, some outstanding work restricts attention to local regions or extends axial interactions. However, these methos often lack the interaction of local and global information, balancing coarse and fine-grained information. To address this problem, we propose AxWin Attention, which models context information in both local windows and axial views. Based on the AxWin Attention, we develop a context-aware vision transformer backbone, named AxWin Transformer, which outperforming the state-of-the-art methods in both classification and downstream segmentation and detection tasks.", "authors": ["Fangjian Lin", "Yizhe Ma", "Sitong Wu", "Long Yu", "Shengwei Tian"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-05-02", "url": "https://arxiv.org/abs/2305.01280", "pdf_url": "https://arxiv.org/pdf/2305.01280v1", "arxiv_id": "2305.01280", "doi": "10.48550/arXiv.2305.01280", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "8c28cd1f246454ea87baeeb6e3db428777415e6ff9f43b02ca2f94663158a78c", "sources": ["arxiv", "semantic_scholar"], "title": "RGB-D-Inertial SLAM in Indoor Dynamic Environments with Long-term Large Occlusion", "abstract": "This work presents a novel RGB-D-inertial dynamic SLAM method that can enable accurate localisation when the majority of the camera view is occluded by multiple dynamic objects over a long period of time. Most dynamic SLAM approaches either remove dynamic objects as outliers when they account for a minor proportion of the visual input, or detect dynamic objects using semantic segmentation before camera tracking. Therefore, dynamic objects that cause large occlusions are difficult to detect without prior information. The remaining visual information from the static background is also not enough to support localisation when large occlusion lasts for a long period. To overcome these problems, our framework presents a robust visual-inertial bundle adjustment that simultaneously tracks camera, estimates cluster-wise dense segmentation of dynamic objects and maintains a static sparse map by combining dense and sparse features. The experiment results demonstrate that our method achieves promising localisation and object segmentation performance compared to other state-of-the-art methods in the scenario of long-term large occlusion.", "authors": ["Ran Long", "Christian Rauch", "Vladimir Ivan", "Tin Lun Lam", "Sethu Vijayakumar"], "categories": ["cs.RO"], "fields_of_study": ["Computer Science"], "published_date": "2023-03-23", "url": "https://arxiv.org/abs/2303.13316", "pdf_url": "https://arxiv.org/pdf/2303.13316v1", "arxiv_id": "2303.13316", "doi": "10.48550/arXiv.2303.13316", "citation_count": 7, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.2258}
{"id": "6fbe81a38ddcdc21291522aaa45fc5b7f722fce904a0df5acb627a4b90965ff4", "sources": ["arxiv", "semantic_scholar"], "title": "Context-faithful Prompting for Large Language Models", "abstract": "Large language models (LLMs) encode parametric knowledge about world facts and have shown remarkable performance in knowledge-driven NLP tasks. However, their reliance on parametric knowledge may cause them to overlook contextual cues, leading to incorrect predictions in context-sensitive NLP tasks (e.g., knowledge acquisition tasks). In this paper, we seek to assess and enhance LLMs' contextual faithfulness in two aspects: knowledge conflict and prediction with abstention. We demonstrate that LLMs' faithfulness can be significantly improved using carefully designed prompting strategies. In particular, we identify opinion-based prompts and counterfactual demonstrations as the most effective methods. Opinion-based prompts reframe the context as a narrator's statement and inquire about the narrator's opinions, while counterfactual demonstrations use instances containing false facts to improve faithfulness in knowledge conflict situations. Neither technique requires additional training. We conduct experiments on three datasets of two standard NLP tasks, machine reading comprehension and relation extraction, and the results demonstrate significant improvement in faithfulness to contexts. Code and data are released at https://github.com/wzhouad/context-faithful-llm.", "authors": ["Wenxuan Zhou", "Sheng Zhang", "Hoifung Poon", "Muhao Chen"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-03-20", "url": "https://arxiv.org/abs/2303.11315", "pdf_url": "https://arxiv.org/pdf/2303.11315v2", "arxiv_id": "2303.11315", "doi": "10.48550/arXiv.2303.11315", "citation_count": 98, "influential_citation_count": 8, "has_code": true, "code_url": "https://github.com/wzhouad/context-faithful-llm", "venue": "Conference on Empirical Methods in Natural Language Processing", "quality_score": 0.4989}
{"id": "65e4153c95fac3f0f431db59ed17de352a9ac27987977fe07186b366f2381a79", "sources": ["arxiv", "semantic_scholar"], "title": "Mutual Exclusive Modulator for Long-Tailed Recognition", "abstract": "The long-tailed recognition (LTR) is the task of learning high-performance classifiers given extremely imbalanced training samples between categories. Most of the existing works address the problem by either enhancing the features of tail classes or re-balancing the classifiers to reduce the inductive bias. In this paper, we try to look into the root cause of the LTR task, i.e., training samples for each class are greatly imbalanced, and propose a straightforward solution. We split the categories into three groups, i.e., many, medium and few, according to the number of training images. The three groups of categories are separately predicted to reduce the difficulty for classification. This idea naturally arises a new problem of how to assign a given sample to the right class groups? We introduce a mutual exclusive modulator which can estimate the probability of an image belonging to each group. Particularly, the modulator consists of a light-weight module and learned with a mutual exclusive objective. Hence, the output probabilities of the modulator encode the data volume clues of the training dataset. They are further utilized as prior information to guide the prediction of the classifier. We conduct extensive experiments on multiple datasets, e.g., ImageNet-LT, Place-LT and iNaturalist 2018 to evaluate the proposed approach. Our method achieves competitive performance compared to the state-of-the-art benchmarks.", "authors": ["Haixu Long", "Xiaolin Zhang", "Yanbin Liu", "Zongtai Luo", "Jianbo Liu"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2023-02-19", "url": "https://arxiv.org/abs/2302.09498", "pdf_url": "https://arxiv.org/pdf/2302.09498v2", "arxiv_id": "2302.09498", "doi": "10.1109/CVPRW59228.2023.00517", "citation_count": 2, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.1193}
{"id": "3982ab263a5ec3fdcadb0274f1451c3949035cb8d40f6752d077129619e0cbdd", "sources": ["arxiv", "semantic_scholar"], "title": "Encoding Sentence Position in Context-Aware Neural Machine Translation with Concatenation", "abstract": "Context-aware translation can be achieved by processing a concatenation of consecutive sentences with the standard Transformer architecture. This paper investigates the intuitive idea of providing the model with explicit information about the position of the sentences contained in the concatenation window. We compare various methods to encode sentence positions into token representations, including novel methods. Our results show that the Transformer benefits from certain sentence position encoding methods on English to Russian translation if trained with a context-discounted loss (Lupo et al., 2022). However, the same benefits are not observed in English to German. Further empirical efforts are necessary to define the conditions under which the proposed approach is beneficial.", "authors": ["Lorenzo Lupo", "Marco Dinarelli", "Laurent Besacier"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2023-02-13", "url": "https://arxiv.org/abs/2302.06459", "pdf_url": "https://arxiv.org/pdf/2302.06459v2", "arxiv_id": "2302.06459", "doi": "10.48550/arXiv.2302.06459", "citation_count": 11, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "First Workshop on Insights from Negative Results in NLP", "quality_score": 0.2698}
{"id": "98dd26fc205f47106d827caaf11e34f74718052b1296b1ab41abf04bc7ba164c", "sources": ["arxiv", "semantic_scholar"], "title": "Black-box language model explanation by context length probing", "abstract": "The increasingly widespread adoption of large language models has highlighted the need for improving their explainability. We present context length probing, a novel explanation technique for causal language models, based on tracking the predictions of a model as a function of the length of available context, and allowing to assign differential importance scores to different contexts. The technique is model-agnostic and does not rely on access to model internals beyond computing token-level probabilities. We apply context length probing to large pre-trained language models and offer some initial analyses and insights, including the potential for studying long-range dependencies. The source code and an interactive demo of the method are available.", "authors": ["Ondřej Cífka", "Antoine Liutkus"], "categories": ["cs.CL", "cs.LG"], "fields_of_study": ["Computer Science"], "published_date": "2022-12-30", "url": "https://arxiv.org/abs/2212.14815", "pdf_url": "https://arxiv.org/pdf/2212.14815v3", "arxiv_id": "2212.14815", "doi": "10.18653/v1/2023.acl-short.92", "citation_count": 10, "influential_citation_count": 2, "has_code": true, "code_url": "https://github.com/cifkao/context-probing/", "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.2603}
{"id": "6996e418534fe97b1b36de85c63d14cb1622852781d4198d0ba45fa68f58ec1c", "sources": ["arxiv", "semantic_scholar"], "title": "Topical Hidden Genome: Discovering Latent Cancer Mutational Topics using a Bayesian Multilevel Context-learning Approach", "abstract": "Statistical inference on the cancer-site specificities of collective ultra-rare whole genome somatic mutations is an open problem. Traditional statistical methods cannot handle whole-genome mutation data due to their ultra-high-dimensionality and extreme data sparsity -- e.g., >30 million unique variants are observed in the ~1700 whole-genome tumor dataset considered herein, of which >99% variants are encountered only once. To harness information in these rare variants we have recently proposed the \"hidden genome model\", a formal multilevel multi-logistic model that mines information in ultra-rare somatic variants to characterize tumor types. The model condenses signals in rare variants through a hierarchical layer leveraging contexts of individual mutations. The model is currently implemented using consistent, scalable point estimation techniques that can handle 10s of millions of variants detected across thousands of tumors. Our recent publications have evidenced its impressive accuracy and attributability at scale. However, principled statistical inference from the model is infeasible due to the volume, correlation, and non-interpretability of the mutation contexts. In this paper we propose a novel framework that leverages topic models from the field of computational linguistics to induce an *interpretable dimension reduction* of the mutation contexts used in the model. The proposed model is implemented using an efficient MCMC algorithm that permits rigorous full Bayesian inference at a scale that is orders of magnitude beyond the capability of out-of-the-box high-dimensional multi-class regression methods and software. We employ our model on the Pan Cancer Analysis of Whole Genomes (PCAWG) dataset, and our results reveal interesting novel insights.", "authors": ["Saptarshi Chakraborty", "Zoe Guan", "Colin B. Begg", "Ronglai Shen"], "categories": ["stat.ME", "stat.AP", "stat.CO"], "fields_of_study": ["Mathematics", "Medicine"], "published_date": "2022-12-30", "url": "https://arxiv.org/abs/2212.14567", "pdf_url": "https://arxiv.org/pdf/2212.14567v1", "arxiv_id": "2212.14567", "doi": "10.1093/biomtc/ujae030", "citation_count": 2, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "Biometrics", "quality_score": 0.1505}
{"id": "8bd449a33e6a9ef960e324d22c1f4ce53578725688d88646da56994c676a4fe2", "sources": ["arxiv", "semantic_scholar"], "title": "Parallel Context Windows for Large Language Models", "abstract": "When applied to processing long text, Large Language Models (LLMs) are limited by their context window. Existing efforts to address this limitation involve training specialized architectures, and cannot be easily applied to off-the-shelf LLMs. We present Parallel Context Windows (PCW), a method that alleviates the context window restriction for any off-the-shelf LLM without further training. The key to the approach is to carve a long context into chunks (``windows''), restrict the attention mechanism to apply only within each window, and re-use the positional embeddings across the windows. Our main results test the PCW approach on in-context learning with models that range in size between 750 million and 178 billion parameters, and show substantial improvements for tasks with diverse input and output spaces. We show additional benefits in other settings where long context windows may be beneficial: multi-hop questions and retrieval-augmented question answering with multiple retrieved documents. Our results highlight Parallel Context Windows as a promising method for applying off-the-shelf LLMs in a range of settings that require long text sequences. We make our code publicly available at https://github.com/ai21labs/parallel-context-windows.", "authors": ["Nir Ratner", "Yoav Levine", "Yonatan Belinkov", "Ori Ram", "Inbal Magar", "Omri Abend", "Ehud Karpas", "Amnon Shashua", "Kevin Leyton-Brown", "Yoav Shoham"], "categories": ["cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2022-12-21", "url": "https://arxiv.org/abs/2212.10947", "pdf_url": "https://arxiv.org/pdf/2212.10947v3", "arxiv_id": "2212.10947", "doi": "10.18653/v1/2023.acl-long.352", "citation_count": 101, "influential_citation_count": 10, "has_code": true, "code_url": "https://github.com/ai21labs/parallel-context-windows", "venue": "Annual Meeting of the Association for Computational Linguistics", "quality_score": 0.5207}
{"id": "27ae328213b7bc017b7ce2c29190f1e66218b823b82d0f3198a243df8df476ff", "sources": ["arxiv", "semantic_scholar"], "title": "Fine-Grained Distillation for Long Document Retrieval", "abstract": "Long document retrieval aims to fetch query-relevant documents from a large-scale collection, where knowledge distillation has become de facto to improve a retriever by mimicking a heterogeneous yet powerful cross-encoder. However, in contrast to passages or sentences, retrieval on long documents suffers from the scope hypothesis that a long document may cover multiple topics. This maximizes their structure heterogeneity and poses a granular-mismatch issue, leading to an inferior distillation efficacy. In this work, we propose a new learning framework, fine-grained distillation (FGD), for long-document retrievers. While preserving the conventional dense retrieval paradigm, it first produces global-consistent representations crossing different fine granularity and then applies multi-granular aligned distillation merely during training. In experiments, we evaluate our framework on two long-document retrieval benchmarks, which show state-of-the-art performance.", "authors": ["Yucheng Zhou", "Tao Shen", "Xiubo Geng", "Chongyang Tao", "Guodong Long", "Can Xu", "Daxin Jiang"], "categories": ["cs.IR", "cs.CL"], "fields_of_study": ["Computer Science"], "published_date": "2022-12-20", "url": "https://arxiv.org/abs/2212.10423", "pdf_url": "https://arxiv.org/pdf/2212.10423v1", "arxiv_id": "2212.10423", "doi": "10.48550/arXiv.2212.10423", "citation_count": 47, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "AAAI Conference on Artificial Intelligence", "quality_score": 0.4203}
{"id": "294577f95521cda41444698852b7dc6b952b2582ccd4724311e62c9e6c00bcfe", "sources": ["arxiv", "semantic_scholar"], "title": "Ionization-induced Long-lasting Orientation of Symmetric-top Molecules", "abstract": "We theoretically consider the phenomenon of field-free long-lasting orientation of symmetric-top molecules ionized by two-color laser pulses. The anisotropic ionization produces a significant long-lasting orientation of the surviving neutral molecules. The degree of orientation increases with both the pulse intensity and, counterintuitively, with the rotational temperature. The orientation may be enhanced even further by using multiple delayed two-color pulses. The long-lasting orientation may be probed by even harmonic generation or by Coulomb-explosion-based methods. The effect may enable the study of relaxation processes in dense molecular gases, and may be useful for molecular guiding and trapping by inhomogeneous fields.", "authors": ["Long Xu", "Ilia Tutunnikov", "Yehiam Prior", "Ilya Sh. Averbukh"], "categories": ["physics.chem-ph", "physics.optics", "quant-ph"], "fields_of_study": ["Physics"], "published_date": "2022-11-16", "url": "https://arxiv.org/abs/2211.08795", "pdf_url": "https://arxiv.org/pdf/2211.08795v1", "arxiv_id": "2211.08795", "doi": "10.1103/PhysRevA.107.023111", "citation_count": 1, "influential_citation_count": 0, "has_code": false, "code_url": null, "venue": null, "quality_score": 0.0753}
{"id": "fa529a1735ed2f5fa0af6490ed0b54dcaafae19455ed9a3378efa517e459846f", "sources": ["arxiv", "semantic_scholar"], "title": "Improving Long Tailed Document-Level Relation Extraction via Easy Relation Augmentation and Contrastive Learning", "abstract": "Towards real-world information extraction scenario, research of relation extraction is advancing to document-level relation extraction(DocRE). Existing approaches for DocRE aim to extract relation by encoding various information sources in the long context by novel model architectures. However, the inherent long-tailed distribution problem of DocRE is overlooked by prior work. We argue that mitigating the long-tailed distribution problem is crucial for DocRE in the real-world scenario. Motivated by the long-tailed distribution problem, we propose an Easy Relation Augmentation(ERA) method for improving DocRE by enhancing the performance of tailed relations. In addition, we further propose a novel contrastive learning framework based on our ERA, i.e., ERACL, which can further improve the model performance on tailed relations and achieve competitive overall DocRE performance compared to the state-of-arts.", "authors": ["Yangkai Du", "Tengfei Ma", "Lingfei Wu", "Yiming Wu", "Xuhong Zhang", "Bo Long", "Shouling Ji"], "categories": ["cs.CL", "cs.AI"], "fields_of_study": ["Computer Science"], "published_date": "2022-05-21", "url": "https://arxiv.org/abs/2205.10511", "pdf_url": "https://arxiv.org/pdf/2205.10511v1", "arxiv_id": "2205.10511", "doi": "10.48550/arXiv.2205.10511", "citation_count": 5, "influential_citation_count": 1, "has_code": false, "code_url": null, "venue": "arXiv.org", "quality_score": 0.1945}
{"id": "ec401a0041446eae6abfa0ad00feafdf3cc5b46eb5c21f8864b8d5f784c25093", "sources": ["arxiv", "semantic_scholar"], "title": "Retrieval Augmented Classification for Long-Tail Visual Recognition", "abstract": "We introduce Retrieval Augmented Classification (RAC), a generic approach to augmenting standard image classification pipelines with an explicit retrieval module. RAC consists of a standard base image encoder fused with a parallel retrieval branch that queries a non-parametric external memory of pre-encoded images and associated text snippets. We apply RAC to the problem of long-tail classification and demonstrate a significant improvement over previous state-of-the-art on Places365-LT and iNaturalist-2018 (14.5% and 6.7% respectively), despite using only the training datasets themselves as the external information source. We demonstrate that RAC's retrieval module, without prompting, learns a high level of accuracy on tail classes. This, in turn, frees the base encoder to focus on common classes, and improve its performance thereon. RAC represents an alternative approach to utilizing large, pretrained models without requiring fine-tuning, as well as a first step towards more effectively making use of external memory within common computer vision architectures.", "authors": ["Alexander Long", "Wei Yin", "Thalaiyasingam Ajanthan", "Vu Nguyen", "Pulak Purkait", "Ravi Garg", "Alan Blair", "Chunhua Shen", "Anton van den Hengel"], "categories": ["cs.CV"], "fields_of_study": ["Computer Science"], "published_date": "2022-02-22", "url": "https://arxiv.org/abs/2202.11233", "pdf_url": "https://arxiv.org/pdf/2202.11233v1", "arxiv_id": "2202.11233", "doi": "10.1109/CVPR52688.2022.00683", "citation_count": 148, "influential_citation_count": 10, "has_code": false, "code_url": null, "venue": "Computer Vision and Pattern Recognition", "quality_score": 0.5433}