Daily Papers

LLM serving relies on prefix caching to improve inference performance. As growing contexts push key-value (KV) cache footprint far beyond GPU HBM and CPU DRAM capacity, KV cache is increasingly offloaded to NVMe SSDs. Unfortunately, restoring KV cache from SSDs suffers from poor I/O performance and incurs significant GPU stalls. This is primarily because the fragmented GPU memory layout results in a massive number of tiny random I/Os, rendering the low-parallelism CPU a severe bottleneck even with GPU Direct Storage (GDS), which still relies on CPU intervention to initiate each I/O and thus remains CPU-centric. This paper presents Tutti, an efficient SSD-backed KV caching solution that eliminates CPU intervention from the critical data and I/O control paths between HBM and SSDs. At the core of Tutti is a GPU-centric KV cache object store, in which the CPU is only responsible for asynchronously loading I/O kernels once per layer to the GPU. Tutti saturates NVMe SSD bandwidth and reduces GPU stalls to near zero through the following designs: (i) we provide a GPU-native object abstraction that enables bulk KV cache transfers and management; (ii) we re-architect the GPU storage stack by introducing GPU io_uring to support asynchronous GPU direct object I/O; and (iii) we propose slack-aware I/O scheduling to avoid GPU resource contention. We have implemented Tutti and integrated it to vLLM. Extensive evaluation shows that compared to the state-of-the-art GDS-enabled, SSD-backed LMCache, Tutti reduces TTFT by 78.3% under strict SLO constraints and improves the achievable request rate by 2x. The serving cost is reduced by 27%. Tutti achieves nearly the same inference performance as DRAM-backed LMCache, while providing almost infinite capacity.

In LLM serving, reusing the KV cache of prompts across requests is critical for reducing TTFT and serving costs. Cache-affinity scheduling, which co-locates requests with the same prompt prefix to maximize KV cache reuse, often conflicts with load-balancing scheduling that distributes requests evenly across compute instances. Existing schedulers fail to reconcile this trade-off as they operate within a single mapping space, typically applying cache-affinity routing to a subset of requests and load-balanced routing to the rest, without a unified solution to achieve both goals. To address this limitation, we propose DualMap, a dual-mapping scheduling strategy for distributed LLM serving that achieves both cache affinity and load balancing. Its key idea is to map each request to two candidate instances via two independent hash functions based on the request prompt, then intelligently select the better candidate based on current system states. This design increases the likelihood that requests with shared prefixes are co-located, while evenly dispersing distinct prefixes across the cluster via ``the power of two choices''. To make DualMap robust under dynamic and skewed real-world workloads, we incorporate three techniques: 1) SLO-aware request routing, which prioritizes cache affinity but switches to load-aware scheduling when TTFT exceeds the SLO, enhancing load balance without sacrificing cache reuse; 2) hotspot-aware rebalancing, which dynamically migrates requests from overloaded to underloaded instances, mitigating hotspots and rebalancing the system; 3) lightweight dual-hash-ring scaling, which leverages a dual-hash-ring mapping to support fast and low-overhead instance scaling without costly global remapping. Experiments on real-world workloads show that DualMap improves effective request capacity by up to 2.25times under the same TTFT SLO constraints compared with SOTA work.

Large language models (LLMs) have demonstrated remarkable performance across a wide range of industrial applications, from search and recommendation systems to generative tasks. Although scaling laws indicate that larger models generally yield better generalization and performance, their substantial computational requirements often render them impractical for many real-world scenarios at scale. In this paper, we present a comprehensive set of insights for training and deploying small language models (SLMs) that deliver high performance for a variety of industry use cases. We focus on two key techniques: (1) knowledge distillation and (2) model compression via structured pruning and quantization. These approaches enable SLMs to retain much of the quality of their larger counterparts while significantly reducing training/serving costs and latency. We detail the impact of these techniques on a variety of use cases in a large professional social network platform and share deployment lessons, including hardware optimization strategies that improve speed and throughput for both predictive and reasoning-based applications in Recommendation Systems.

Retrieval-augmented generation (RAG) improves large language model (LLM) answer quality by grounding generation in external evidence, but processing retrieved contexts makes the prefill stage a dominant serving cost. RAG cache fusion reduces this cost by reusing precomputed key-value (KV) caches for retrieved chunks and selectively recomputing tokens under the current prompt. Existing selectors, however, face a dilemma between quality and efficiency: fast query-agnostic or final-layer query-to-context selectors can miss request-relevant evidence, whereas full-view query-aware selectors require broad context and layer visibility before recomputation and therefore stall the layer-wise cache-fusion pipeline. We present QCFuse, a compressed-view query-aware selector for RAG cache fusion. QCFuse uses chunk-anchor query probing to condition user-query states on compact per-chunk anchors and critical-layer profiling to identify recomputation tokens without all-layer inspection. We implement QCFuse in SGLang and evaluate it on four open-weight LLMs across six datasets. QCFuse reaches full-prefill-level quality. At matched quality, QCFuse achieves an average prefill-time speedup of 1.7x over full prefill and 1.5x over ProphetKV, the strongest quality-preserving baseline.

Reducing serving cost and latency is a fundamental concern for the deployment of language models (LMs) in business applications. To address this, cascades of LMs offer an effective solution that conditionally employ smaller models for simpler queries. Cascaded systems are typically built with independently trained models, neglecting the advantages of considering inference-time interactions of the cascaded LMs during training. In this paper, we present cascade-aware training(CAT), an approach to optimizing the overall quality-cost performance tradeoff of a cascade of LMs. We achieve inference-time benefits by training the small LM with awareness of its place in a cascade and downstream capabilities. We demonstrate the value of the proposed method with over 60 LM tasks of the SuperGLUE, WMT22, and FLAN2021 datasets.

Quantization effectively reduces the serving costs of Large Language Models (LLMs) by speeding up data movement through compressed parameters and enabling faster operations via integer arithmetic. However, activating integer arithmetic requires quantizing both weights and activations, which poses challenges due to the significant outliers in LLMs that increase quantization error. In this work, we investigate these outliers with an emphasis on their effect on layer-wise quantization error, then examine how smoothing and rotation transform the observed values. Our primary contributions include introducing a new metric to measure and visualize quantization difficulty based on channel magnitudes, as well as proposing a hybrid approach that applies channel-wise scaling before rotation, supported by a mathematical formulation of its benefits.

Leech-lattice vector quantization holds the strongest reported 2-bit quality under its own evaluation protocol. Its kernel decodes one shell; we found no implementation of the multi-shell decoder the rate requires. This paper supplies one and measures its serving cost for decode-phase GEMV at batch 1. First, a serving path for the full 301-class codebook: an offline expansion into GPU layouts and a fused dequantize-plus-matvec kernel reading them without warp divergence, verified against f64. Second, the in-VRAM rate is a design axis distinct from the on-disk rate. Four bit-exact layouts timed in one process show binary bit planes beating one-hot masks on size and speed at constant bandwidth (4.80 bits per weight, 2.15x FP16). Below 4.3 bits a second, irregular stream enters; at 3.6 the decode stops being shifts and masks. Third, deployed four-bit (AWQ) and two-bit (QTIP) GEMV kernels run in the same process. The trellis kernel reads 2.40x fewer bytes than our served layout and runs 2.27x faster at near-equal fractions of their byte bounds: the time gap tracks the traffic gap, the price of unfolding a codebook too large for a lookup table. Fourth, the validity envelope: the trellis kernel outruns our no-weights control, so our launch geometry sets that floor, and on a second memory hierarchy every lattice arm falls below FP16. With the output head held identical across arms, the kernel-and-format path gains 1.11x, 1.29x and 1.41x end to end at 4B, 8B and 14B; with an int8 output head the served 4B reaches 87.0 tok/s in 2.60 GB. The quality cost, 1.38x perplexity and 14.7 MMLU points at 4B, shrinks across the three sizes measured.

Serving transformer language models with high throughput requires caching Key-Values (KVs) to avoid redundant computation during autoregressive generation. The memory footprint of KV caching is significant and heavily impacts serving costs. This work proposes to lessen these memory requirements. While recent work has largely addressed KV cache reduction via compression and eviction along the temporal axis, we argue that the depth dimension offers an orthogonal and robust avenue for optimization. Although prior research suggests that a full cache for every layer is redundant, implementing cross-layer cache sharing remains a practical challenge; existing methods typically suffer from reduced throughput or increased time-to-first-token. In this paper, we demonstrate that dropping a layer's cache offers efficient optimization without information loss. We propose a simple training approach: random cross-layer attention. During training, layers randomly choose to attend either to their own KV states or those of a preceding layer. This stochastic process adapts the model to be robust to various depth-wise cache sharing strategies, ensuring flexibility for unknown hardware constraints at deployment time. Our evaluations show that applying this scheme during pre-training or fine-tuning enables depth-wise cache sharing for various model families. Furthermore, for larger models in data-constrained settings, this approach is suggestive of a regularization-like effect, frequently preserving or improving performance while significantly reducing the cache's memory footprint.

Algorithmic efficiency techniques such as distillation (hinton2015distillation) are useful in improving model quality without increasing serving costs, provided a larger teacher model is available for a smaller student model to learn from during training. Standard distillation methods are limited to only forcing the student to match the teacher's outputs. Given the costs associated with training a large model, we believe we should be extracting more useful information from a teacher model than by just making the student match the teacher's outputs. In this paper, we introduce \guide (Guided Initialization and Distillation of Embeddings). \guide can be considered a distillation technique that forces the student to match the teacher in the parameter space. Using \guide we show 25-26\% reduction in the teacher-student quality gap when using large student models (400M - 1B parameters) trained on approx 20B tokens. We also present a thorough analysis demonstrating that \guide can be combined with knowledge distillation with near additive improvements. Furthermore, we show that applying \guide alone leads to substantially better model quality than applying knowledge distillation by itself. Most importantly, \guide introduces no training or inference overhead and hence any model quality gains from our method are virtually free.

We investigate a failure mode of large language models (LLMs) in which plain-text prompts elicit excessive outputs, a phenomenon we term Overflow. Unlike jailbreaks or prompt injection, Overflow arises under ordinary interaction settings and can lead to elevated serving cost, latency, and cross-user performance degradation, particularly when scaled across many requests. Beyond usability, the stakes are economic and environmental: unnecessary tokens increase per-request cost and energy consumption, compounding into substantial operational spend and carbon footprint at scale. Moreover, Overflow represents a practical vector for compute amplification and service degradation in shared environments. We introduce BenchOverflow, a model-agnostic benchmark of nine plain-text prompting strategies that amplify output volume without adversarial suffixes or policy circumvention. Using a standardized protocol with a fixed budget of 5000 new tokens, we evaluate nine open- and closed-source models and observe pronounced rightward shifts and heavy tails in length distributions. Cap-saturation rates (CSR@1k/3k/5k) and empirical cumulative distribution functions (ECDFs) quantify tail risk; within-prompt variance and cross-model correlations show that Overflow is broadly reproducible yet heterogeneous across families and attack vectors. A lightweight mitigation-a fixed conciseness reminder-attenuates right tails and lowers CSR for all strategies across the majority of models. Our findings position length control as a measurable reliability, cost, and sustainability concern rather than a stylistic quirk. By enabling standardized comparison of length-control robustness across models, BenchOverflow provides a practical basis for selecting deployments that minimize resource waste and operating expense, and for evaluating defenses that curb compute amplification without eroding task performance.

Reasoning-focused LLMs improve answer quality by generating longer reasoning traces, but the additional tokens dramatically increase serving cost, motivating inference optimization. We extend and apply Puzzle, a post-training neural architecture search (NAS) framework, to gpt-oss-120B to produce gpt-oss-puzzle-88B, a deployment-optimized derivative. Our approach combines heterogeneous MoE expert pruning, selective replacement of full-context attention with window attention, FP8 KV-cache quantization with calibrated scales, and post-training reinforcement learning to recover accuracy, while maintaining low generation length. In terms of per-token speeds, on an 8XH100 node we achieve 1.63X and 1.22X throughput speedups in long-context and short-context settings, respectively. gpt-oss-puzzle-88B also delivers throughput speedups of 2.82X on a single NVIDIA H100 GPU. However, because token counts can change with reasoning effort and model variants, per-token throughput (tok/s) and latency (ms/token) do not necessarily lead to end-to-end speedups: a 2X throughput gain is erased if traces grow 2X. Conversely, throughput gains can be spent on more reasoning tokens to improve accuracy; we therefore advocate request-level efficiency metrics that normalize throughput by tokens generated and trace an accuracy--speed frontier across reasoning efforts. We show that gpt-oss-puzzle-88B improves over gpt-oss-120B along the entire frontier, delivering up to 1.29X higher request-level efficiency. Across various benchmarks, gpt-oss-puzzle-88B matches or slightly exceeds the parent on suite-average accuracy across reasoning efforts, with retention ranging from 100.8% (high) to 108.2% (low), showing that post-training architecture search can substantially reduce inference costs without sacrificing quality.

LLM agents are rapidly becoming production software, deployed to handle customer service, adjudicate disputes, and operate internal systems. Notably, the work of building them is increasingly handed to coding agents, yet existing benchmarks say little about whether an AI system can deliver one under the conditions of a real client engagement. We introduce τ^τ-bench (pronounced hyper-tau-bench), a benchmark that makes agent construction the task. A developer agent is given the records a business actually keeps, a client who holds requirements, a production API that operations must run through, a codebase to inherit, and limits on serving cost and models: the same starting point a real engagement provides. From these it must deliver a complete customer-service agent, scored by deploying that agent against held-out simulated users. Across 53 tasks spanning four domains, the strongest configuration, Claude Opus 5 under Claude Code, passes just 23.9% of evaluation simulations. Meanwhile, an expert-authored reference ceiling scores 82.2%. The failures mirror ones human agent developers see: models issue shallow queries in place of deep comprehension of the records, communicate almost nothing to the client, and experiment too little with agent architecture and serving spend, shipping the first design that runs. We aim for τ^τ-bench to turn the work of cooperative agent building into a measurable target for coding agents.

Long-context agentic workflows have emerged as a defining use case for large language models, making attention efficiency critical for both inference speed and serving cost. Sparse attention addresses this challenge effectively, and DeepSeek Sparse Attention (DSA) is a representative production-grade solution: a lightweight lightning indexer selects the top-k most relevant tokens per query, reducing core attention from O(L^2) to O(Lk). However, the indexer itself retains O(L^2) complexity and must run independently at every layer, despite the fact that the resulting top-k selections are highly similar across consecutive layers. We present IndexCache, which exploits this cross-layer redundancy by partitioning layers into a small set of Full layers that run their own indexers and a majority of Shared layers that simply reuse the nearest Full layer's top-k indices. We propose two complementary approaches to determine and optimize this configuration. Training-free IndexCache applies a greedy search algorithm that selects which layers to retain indexers by directly minimizing language modeling loss on a calibration set, requiring no weight updates. Training-aware IndexCache introduces a multi-layer distillation loss that trains each retained indexer against the averaged attention distributions of all layers it serves, enabling even simple interleaved patterns to match full-indexer accuracy. Experimental results on a 30B DSA model show that IndexCache can remove 75% of indexer computations with negligible quality degradation, achieving up to 1.82times prefill speedup and 1.48times decode speedup compared to standard DSA. These positive results are further confirmed by our preliminary experiments on the production-scale GLM-5 model (Figure 1).

Data-residency constraints force enterprises to self-host LLMs, but continuous adoption of newer models without decommissioning their predecessors expands the serving fleet, fragmenting a finite GPU pool. We consolidate traffic from over 200 internal applications onto a single model by closing quality gaps identified through production error analysis along three axes: instruction following, function-calling, and internal task distribution. Quality is tracked by offline benchmarks stratified to production traffic and scored by deterministic verifiers or calibrated LLM judges. Rather than optimising all objectives jointly, which introduces cross-domain reward interference, we train a separate GRPO expert per axis and merge them via two-stage SLERP. Each expert's reward exposes a distinct failure mode, namely semantic collapse, over-calling, and verbosity hacking, each requiring a domain-specific fix. In non-reasoning mode the recipe surpasses a {sim}7times larger by total parameters baseline on the in-house Arena with 69.6 to 65.8, instruction following with 0.85 to 0.83, and function-calling with 0.79 to 0.77, while lifting general dialogue benchmarks. The model absorbs 50% of platform traffic, 116M requests per month, at a fraction of the serving cost.

Multimodal representation learning is a cornerstone of modern AI. By encoding multimodal queries and targets into vectors, it powers industrial search and recommendation and underpins modern agents. Real-world platforms with complex modalities and massive-scale content, such as Douyin, Xiaohongshu, and YouTube, demand both efficiency under billion-scale indexing and fine-grained discrimination for hard matching. Existing MLLM embedding models rarely satisfy both. Contrastive models are efficient but rely on pair-level supervision too coarse for fine-grained distinctions, while CoT-based models improve discrimination through explicit generation impractical to serve online. We present Douyin Multimodal Embedding (DME), a model trained in two stages to combine both strengths. Stage 1 performs large-scale contrastive pre-training that establishes a unified multimodal embedding space with broad modality and task coverage. Stage 2 supplements semantic sufficiency, the property that an embedding is grounded in retrieval-relevant evidence and preserves fine-grained counterpart-side semantics, via two mechanisms. Evidence-Grounded Typed Latent Reasoning organizes retrieval evidence through hidden-space latent reasoning, and Cross-Conditional Reconstruction enforces counterpart-side semantics through cross-directional autoregressive reconstruction. Both act only during training and add only marginal query-side overhead, so DME serves as efficiently as a standard contrastive encoder. On MMEB-v2, DME reaches state-of-the-art results at comparable scales for its 2B and 9B variants (74.8 and 78.4), with especially strong video and visual-document tasks. In production, DME delivers a 2.92% relative gain on Douyin's in-house offline evaluation set, is deployed across Douyin scenarios such as generative, image, and AI search, and yields a 0.1% Lifetime (LT) gain in online A/B testing on Douyin search.

Model merging aims to combine multiple expert models into a more capable single model, offering benefits such as reduced storage and serving costs, improved generalization, and support for decentralized model development. Despite its promise, previous studies have primarily focused on merging a few small models. This leaves many unanswered questions about the effect of scaling model size and how it interplays with other key factors -- like the base model quality and number of expert models -- , to affect the merged model's performance. This work systematically evaluates the utility of model merging at scale, examining the impact of these different factors. We experiment with merging fully fine-tuned models using 4 popular merging methods -- Averaging, Task~Arithmetic, Dare, and TIES -- across model sizes ranging from 1B-64B parameters and merging up to 8 different expert models. We evaluate the merged models on both held-in tasks, i.e., the expert's training tasks, and zero-shot generalization to unseen held-out tasks. Our experiments provide several new insights about model merging at scale and the interplay between different factors. First, we find that merging is more effective when experts are created from strong base models, i.e., models with good zero-shot performance. Second, larger models facilitate easier merging. Third merging consistently improves generalization capabilities. Notably, when merging 8 large expert models, the merged models often generalize better compared to the multitask trained models. Fourth, we can better merge more expert models when working with larger models. Fifth, different merging methods behave very similarly at larger scales. Overall, our findings shed light on some interesting properties of model merging while also highlighting some limitations. We hope that this study will serve as a reference point on large-scale merging for upcoming research.

Financial AI agents often fail for a simple reason: they make users carry the complexity. A user must repeatedly restate goals, risk preferences, portfolio context, past judgments, and shifting market assumptions, while the agent answers, retrieves, acts, and forgets. In finance, this is not just inconvenient. In tasks such as market analysis, copy-trading review, and trade preparation, forgotten context and stale memory can create latency, repeated errors, weak auditability, and unsafe decisions. We propose the interaction-native knowledge harness (InKH), an architecture for financial LLM agents that absorbs complexity into the system. InKH converts user, market, portfolio, and tool events into structured operational knowledge. It uses passive knowledge injection to assemble a bounded working context buffer before the main model step, temporal graph memory for low-latency retrieval, a wiki audit surface for human-readable governance, and background extraction with maturity, decay, and write-time invalidation. We evaluate InKH on a reproducible controlled synthetic benchmark with 24 random seeds, 4 rounds, 80 episodes per round, and 6 baselines, producing 46,080 baseline-conditioned evaluations. InKH achieves mean task quality of 0.815 at 900 ms latency. Compared with agent-driven wiki-walk memory, it reduces latency by 82.95 percent, token cost by 82.29 percent, and stale-knowledge usage by 96.58 percent, while improving quality by 0.108 and traceability by 0.461. Compared with a temporal-graph system without invalidation, it improves quality by 0.050 and reduces stale-memory usage by 96.58 percent with comparable serving cost. The results support a design thesis for financial AI: adoption happens when complexity is absorbed by the system rather than transferred to the user. The benchmark validates architecture-level behavior, not live trading performance.

Visual document retrieval has become essential for accessing information in visually rich documents. Existing approaches fall into two camps. Late-interaction retrievers achieve strong quality through fine-grained token-level matching but store hundreds of vectors per page, incurring large index footprints and high serving costs. By contrast, dense single-vector retrievers retain storage and latency advantages but consistently lag in quality because they compress all information into a single final-layer embedding. In this work, we first conduct a layerwise diagnostic on single-vector retrievers, revealing that retrieval-relevant signal resides in internal representations. Motivated by these findings, we propose MINER (Mining Multimodal Internal RepreseNtation for Efficient Retrieval), a lightweight plug-in module that probes and fuses internal signals across transformer layers into a single compact embedding without modifying the backbone or sacrificing single-vector efficiency. The first Retrieval-Aligned Layer Probing stage attaches a lightweight probe at each layer, surfacing which dimensions carry retrieval-relevant information. The subsequent Adaptive Sparse Multi-Layer Fusion stage applies performance-adaptive neuron-level masking to the selected layers and fuses the surviving signals into the final dense vector. Across ViDoRe V1/V2/V3, MINER outperforms existing dense single-vector retrievers on the majority of benchmarks, with up to 4.5% nDCG@5 improvement over its corresponding backbone. Compared to strong late-interaction baselines, in some settings MINER substantially narrows the nDCG@5 gap to 0.2 while preserving the storage and serving advantages of dense retrieval.

Large language models (LLMs) have brought exciting new advances to mobile UI agents, a long-standing research field that aims to complete arbitrary natural language tasks through mobile UI interactions. However, existing UI agents usually demand high reasoning capabilities of powerful large models that are difficult to be deployed locally on end-users' devices, which raises huge concerns about user privacy and centralized serving cost. One way to reduce the required model size is to customize a smaller domain-specific model with high-quality training data, e.g. large-scale human demonstrations of diverse types of apps and tasks, while such datasets are extremely difficult to obtain. Inspired by the remarkable coding abilities of recent small language models (SLMs), we propose to convert the UI task automation problem to a code generation problem, which can be effectively solved by an on-device SLM and efficiently executed with an on-device code interpreter. Unlike normal coding tasks that can be extensively pretrained with public datasets, generating UI automation code is challenging due to the diversity, complexity, and variability of target apps. Therefore, we adopt a document-centered approach that automatically builds fine-grained API documentation for each app and generates diverse task samples based on this documentation. By guiding the agent with the synthetic documents and task samples, it learns to generate precise and efficient scripts to complete unseen tasks. Based on detailed comparisons with state-of-the-art mobile UI agents, our approach effectively improves the mobile task automation with significantly higher success rates and lower latency/token consumption. Code will be open-sourced.

User modeling in industrial recommender systems typically produces dense embeddings, which suffer from representational constraints inherent to fixed-dimensional vectors. An emerging alternative for discrete user representation -- using LLMs to generate text-based user tokens -- captures topical co-occurrences rather than deep sequential behavior dynamics and produces outputs that are difficult to ground to item attributes. Meanwhile, Semantic ID (SID) based item tokenization has proven effective for improving generalization in generative recommendation, yet discrete SID-based representations for users remain largely unexplored. We propose TokenMinds, an industrial-scale system that extends the PLUM framework from item retrieval to user modeling, generating both discrete SID-based user tokens and dense user embeddings via an encoder-decoder architecture adapted from pre-trained LLMs. This dual-output design provides the complementary benefits of discrete, semantically grounded user representations while maintaining compatibility with existing downstream models that rely on dense embeddings. Additionally, the shared SID vocabulary naturally extends to cross-scenario modeling: by unifying long-form and short-form video behaviors into a single model, we substantially reduce training and serving costs. We validate TokenMinds through extensive offline experiments and live launches on multiple YouTube surfaces, served on full user traffic (billions of users) via an asynchronous infrastructure that decouples representation generation from downstream scoring. Focusing on ranking as the primary downstream use case, our results confirm the practical viability of SID-based user tokens at industrial scale and demonstrate that tokens and dense embeddings provide complementary value across different production ranking systems.

While foundation models update slowly due to resource-intensive training requirements, domain-specific models evolve between updates. Model merging aims to combine multiple expert models into a single, more capable model, thereby reducing storage and serving costs while supporting decentralized model development. Despite its potential, previous studies have primarily focused on merging visual classification models or Large Language Models (LLMs) for code and math tasks. Multimodal Large Language Models (MLLMs), which extend the capabilities of LLMs through large-scale multimodal training, have gained traction. However, there lacks a benchmark for model merging research that clearly divides the tasks for MLLM training and evaluation. In this paper, (i) we introduce the model merging benchmark for MLLMs, which includes multiple tasks such as VQA, Geometry, Chart, OCR, and Grounding, providing both LoRA and full fine-tuning models. Moreover, we explore how model merging can combine different modalities (e.g., vision-language, audio-language, and video-language models), moving toward the Omni-language model. (ii) We implement 10 model merging algorithms on the benchmark. Furthermore, we propose a novel method that removes noise from task vectors and robustly optimizes the merged vector based on a loss defined over task vector interactions, achieving an average performance gain of 2.48%. (iii) We find that model merging offers a promising way for building improved MLLMs without requiring data training. Our results also demonstrate that the complementarity among multiple modalities outperforms individual modalities.

LLM-based agentic applications have shown increasingly remarkable capabilities in complex workflows but incur substantial costs due to extensive planning and reasoning requirements. Existing LLM caching techniques (like context caching and semantic caching), primarily designed for serving chatbots, are insufficient for agentic applications where outputs depend on external data or environmental contexts. We propose agentic plan caching, a novel approach that extracts, stores, adapts, and reuses structured plan templates from planning stages of agentic applications across semantically similar tasks to reduce the cost of serving. Unlike traditional semantic caching, our system extracts plan templates from completed agent executions at test-time, employs keyword extraction to match new requests against cached plans, and utilizes lightweight models to adapt these templates to task-specific plans with contexts. Evaluation across multiple real-world agentic applications shows that our system can reduce costs by 46.62% on average while maintaining performance, offering a more efficient solution for serving LLM-based agents that complements existing LLM serving infrastructures.

While model serving has unlocked unprecedented capabilities, the high cost of serving large-scale models continues to be a significant barrier to widespread accessibility and rapid innovation. Compiler optimizations have long driven substantial performance improvements, but existing compilers struggle with neural workloads due to the exponentially large and highly interdependent space of possible transformations. Although existing stochastic search techniques can be effective, they are often sample-inefficient and fail to leverage the structural context underlying compilation decisions. We set out to investigate the research question of whether reasoning with large language models (LLMs), without any retraining, can leverage the context-aware decision space of compiler optimizations to significantly improve sample efficiency. To that end, we introduce a novel compilation framework (dubbed REASONING COMPILER) that formulates optimization as a sequential, context-aware decision process guided by a large language model and structured Monte Carlo tree search (MCTS). The LLM acts as a proposal mechanism, suggesting hardware-informed transformations that reflect the current program state and accumulated performance feedback. MCTS incorporates the LLM-generated proposals to balance exploration and exploitation, facilitating a structured, context-sensitive traversal of the expansive compiler optimization space. By achieving substantial speedups with markedly fewer samples than leading neural compilers, our approach demonstrates the potential of LLM-guided reasoning to transform the landscape of compiler optimization.

Many applications must provide low-latency LLM service to users or risk unacceptable user experience. However, over-provisioning resources to serve fluctuating request patterns is often prohibitively expensive. In this work, we present a best-effort serving system that employs deep reinforcement learning to adjust service quality based on the task distribution and system load. Our best-effort system can maintain availability with over 10x higher client request rates, serves above 96% of peak performance 4.1x more often, and serves above 98% of peak performance 2.3x more often than static serving on unpredictable workloads. Our learned router is robust to shifts in both the arrival and task distribution. Compared to static serving, learned best-effort serving allows for cost-efficient serving through increased hardware utility. Additionally, we argue that learned best-effort LLM serving is applicable in wide variety of settings and provides application developers great flexibility to meet their specific needs.

We introduce AIBrix, a cloud-native, open-source framework designed to optimize and simplify large-scale LLM deployment in cloud environments. Unlike traditional cloud-native stacks, AIBrix follows a co-design philosophy, ensuring every layer of the infrastructure is purpose-built for seamless integration with inference engines like vLLM. AIBrix introduces several key innovations to reduce inference costs and enhance performance including high-density LoRA management for dynamic adapter scheduling, LLM-specific autoscalers, and prefix-aware, load-aware routing. To further improve efficiency, AIBrix incorporates a distributed KV cache, boosting token reuse across nodes, leading to a 50% increase in throughput and a 70% reduction in inference latency. AIBrix also supports unified AI runtime which streamlines model management while maintaining vendor-agnostic engine compatibility. For large-scale multi-node inference, AIBrix employs hybrid orchestration -- leveraging Kubernetes for coarse-grained scheduling and Ray for fine-grained execution -- to balance efficiency and flexibility. Additionally, an SLO-driven GPU optimizer dynamically adjusts resource allocations, optimizing heterogeneous serving to maximize cost efficiency while maintaining service guarantees. Finally, AIBrix enhances system reliability with AI accelerator diagnostic tools, enabling automated failure detection and mock-up testing to improve fault resilience. AIBrix is available at https://github.com/vllm-project/aibrix.

Semantic caching cuts LLM inference costs by serving a cached response to semantically similar queries. Standard practice evaluates these systems using PR-AUC, a metric that only measures how well scores rank and ignores whether they are usable at a fixed threshold. We show this mismatch leads to systematically poor deployment choices, as models with the highest PR-AUC are often the worst in operation. We introduce Precision-Cache Hit Ratio (P-CHR) AUC, a cache-aware metric that measures precision across cache utilization levels, and Calibration Retention Rate (CRR), which captures how much offline ranking quality survives at deployment. We decompose the operational gap between offline and deployed quality into a recoverable calibration component and an irreducible structural component fixed by the dataset's positive rate. Our experiments show that the calibration gap is governed by the training objective rather than data scale, and post-hoc calibration only partially closes it. Ultimately, model selection for semantic caching is a calibration problem, not a ranking one, and measuring it is the first step to closing the gap.

Transformer-based large language models (LLMs) exhibit impressive performance in generative tasks but introduce significant challenges in real-world serving due to inefficient use of the expensive, computation-optimized accelerators. This mismatch arises from the autoregressive nature of LLMs, where the generation phase comprises operators with varying resource demands. Specifically, the attention operator is memory-intensive, exhibiting a memory access pattern that clashes with the strengths of modern accelerators, especially as context length increases. To enhance the efficiency and cost-effectiveness of LLM serving, we introduce the concept of attention offloading. This approach leverages a collection of cheap, memory-optimized devices for the attention operator while still utilizing high-end accelerators for other parts of the model. This heterogeneous setup ensures that each component is tailored to its specific workload, maximizing overall performance and cost efficiency. Our comprehensive analysis and experiments confirm the viability of splitting the attention computation over multiple devices. Also, the communication bandwidth required between heterogeneous devices proves to be manageable with prevalent networking technologies. To further validate our theory, we develop Lamina, an LLM inference system that incorporates attention offloading. Experimental results indicate that Lamina can provide 1.48x-12.1x higher estimated throughput per dollar than homogeneous solutions.

Post-training quantization (PTQ) is widely used to reduce the cost of serving large language models (LLMs), but its accuracy cost is uneven and is often tuned per model. We study where quantization damage occurs and how to allocate a small additional precision budget. Using causal mixed-precision intervention as ground truth (raise each layer to 8-bit in turn and measure the accuracy it recovers) across 9 open-weight models in 4 architecture families, we test 3 intuitive hypotheses: that quantization damage lives in task circuits, where the model computes, or in weight statistics. None of them predicts which layers benefit from restored precision. Recovery is instead diffuse: for 8 of 9 models, recovering 75% of the gap takes roughly half the layers; the lone exception, Qwen3-8B, is sharply concentrated. At a matched precision budget, spending it globally on finer quantization granularity beats locally repairing the most recoverable layers for all 8 group-128-compatible models (all but OpenLLaMA, whose width rules out group-128), by 21-52 points, including the concentrated Qwen3-8B. We report 2 secondary findings: the residual is budget-limited (8-bit is near-lossless in our evaluation across RTN, GPTQ, and AWQ), and the location of peak recovery correlates with architecture within a family, though not across families. Within this budget setting, global granularity is a better default than selectively protecting critical layers. More broadly, cheap signals that correlate with quantization damage do not necessarily identify where restoring precision improves accuracy; this must be tested with causal intervention.

Investment Analysis is a cornerstone of the Financial Services industry. The rapid integration of advanced machine learning techniques, particularly Large Language Models (LLMs), offers opportunities to enhance the equity rating process. This paper explores the application of LLMs to generate multi-horizon stock ratings by ingesting diverse datasets. Traditional stock rating methods rely heavily on the expertise of financial analysts, and face several challenges such as data overload, inconsistencies in filings, and delayed reactions to market events. Our study addresses these issues by leveraging LLMs to improve the accuracy and consistency of stock ratings. Additionally, we assess the efficacy of using different data modalities with LLMs for the financial domain. We utilize varied datasets comprising fundamental financial, market, and news data from January 2022 to June 2024, along with GPT-4-32k (v0613) (with a training cutoff in Sep. 2021 to prevent information leakage). Our results show that our benchmark method outperforms traditional stock rating methods when assessed by forward returns, specially when incorporating financial fundamentals. While integrating news data improves short-term performance, substituting detailed news summaries with sentiment scores reduces token use without loss of performance. In many cases, omitting news data entirely enhances performance by reducing bias. Our research shows that LLMs can be leveraged to effectively utilize large amounts of multimodal financial data, as showcased by their effectiveness at the stock rating prediction task. Our work provides a reproducible and efficient framework for generating accurate stock ratings, serving as a cost-effective alternative to traditional methods. Future work will extend to longer timeframes, incorporate diverse data, and utilize newer models for enhanced insights.

Modern deployments require LLMs to enforce safety policies at scale, yet many controls rely on inference-time interventions that add recurring compute cost and serving complexity. Activation steering is widely used, but it requires runtime hooks and scales cost with the number of generations; conditional variants improve selectivity by gating when steering is applied but still retain an inference-time control path. We ask whether selective refusal can be moved entirely offline: can a mechanistic understanding of category-specific refusal be distilled into a circuit-restricted weight update that deploys as a standard checkpoint? We propose C-Δθ: Circuit Restricted Weight Arithmetic, which (i) localizes refusal-causal computation as a sparse circuit using EAP-IG and (ii) computes a constrained weight update ΔθC supported only on that circuit (typically <5% of parameters). Applying ΔθC yields a drop-in edited checkpoint with no inference-time hooks, shifting cost from per-request intervention to a one-time offline update. We evaluate category-targeted selectivity and capability retention on refusal and utility benchmarks.

Large language models (LLMs) are reshaping recommender systems by enabling richer modeling of users, content, and context directly in natural language. At Netflix, we are exploring this direction through GenRec, an LLM-backed recommendation ranker built on top of an in-house foundational LLM. GenRec follows a two-phase framework: Phase 1 adapts an open-source LLM to Netflix data, developing deep understanding of the catalog and member behavior while balancing capabilities such as content understanding and instruction following. Phase 2 post-trains this foundation model with recommendation-ranking specific data, labels, and reward signals, aiming to align the ranker with business requirements and long-term member satisfaction. This paper focuses on Phase 2 and the transition from a traditional discriminative ranker with thousands of engineered features to an LLM-backed ranker driven by verbalized user histories and context. We describe our design for input verbalization and context engineering, post-training data construction, reward integration, model architecture, and a cost-constrained serving design based on a prefill-only inference approach. We report results from a large-scale A/B test comparing GenRec against the current production ranker model, where we show that a GenRec model trained with substantially fewer Phase-2 labeled training examples and input signals can achieve statistically significant gains in offline and online metrics. We discuss how LLM-backed recommenders could shift the recommendation paradigm: from feature engineering to context engineering, and from bespoke architectures to shared foundation backbones. We also outline practical lessons for serving such systems under real-world resource constraints.

Generative Retrieval (GR) has emerged as a promising paradigm for modern search systems. Compared to multi-stage cascaded architecture, it offers advantages such as end-to-end joint optimization and high computational efficiency. OneSearch, as a representative industrial-scale deployed generative search framework, has brought significant commercial and operational benefits. However, its inadequate understanding of complex queries, inefficient exploitation of latent user intents, and overfitting to narrow historical preferences have limited its further performance improvement. To address these challenges, we propose OneSearch-V2, a latent reasoning enhanced self-distillation generative search framework. It contains three key innovations: (1) a thought-augmented complex query understanding module, which enables deep query understanding and overcomes the shallow semantic matching limitations of direct inference; (2) a reasoning-internalized self-distillation training pipeline, which uncovers users' potential yet precise e-commerce intentions beyond log-fitting through implicit in-context learning; (3) a behavior preference alignment optimization system, which mitigates reward hacking arising from the single conversion metric, and addresses personal preference via direct user feedback. Extensive offline evaluations demonstrate OneSearch-V2's strong query recognition and user profiling capabilities. Online A/B tests further validate its business effectiveness, yielding +3.98\% item CTR, +3.05\% buyer conversion rate, and +2.11\% order volume. Manual evaluation further confirms gains in search experience quality, with +1.65\% in page good rate and +1.37\% in query-item relevance. More importantly, OneSearch-V2 effectively mitigates common search system issues such as information bubbles and long-tail sparsity, without incurring additional inference costs or serving latency.

Large language models (LLMs) are increasingly integrated into many online services. However, a major challenge in deploying LLMs is their high cost, due primarily to the use of expensive GPU instances. To address this problem, we find that the significant heterogeneity of GPU types presents an opportunity to increase GPU cost efficiency and reduce deployment costs. The broad and growing market of GPUs creates a diverse option space with varying costs and hardware specifications. Within this space, we show that there is not a linear relationship between GPU cost and performance, and identify three key LLM service characteristics that significantly affect which GPU type is the most cost effective: model request size, request rate, and latency service-level objective (SLO). We then present M\'elange, a framework for navigating the diversity of GPUs and LLM service specifications to derive the most cost-efficient set of GPUs for a given LLM service. We frame the task of GPU selection as a cost-aware bin-packing problem, where GPUs are bins with a capacity and cost, and items are request slices defined by a request size and rate. Upon solution, M\'elange derives the minimal-cost GPU allocation that adheres to a configurable latency SLO. Our evaluations across both real-world and synthetic datasets demonstrate that M\'elange can reduce deployment costs by up to 77% as compared to utilizing only a single GPU type, highlighting the importance of making heterogeneity-aware GPU provisioning decisions for LLM serving. Our source code is publicly available at https://github.com/tyler-griggs/melange-release.

Efficient deployment of large language models (LLMs) in production forces a trade-off between accuracy and cost. Operators often default to a single model that is either expensive for easy queries or insufficient for hard ones. To address this challenge, we propose a two-stage cascaded solution. Stage 1 clusters incoming queries and assigns each cluster to its most cost-effective model. The cost budget for this routing process is set by an interpretable hyperparameter, tuned offline. Stage 2 adds a quality estimation (QE) cascade; when an output from Stage 1 is judged low-quality, the query is escalated to a stronger model. This ensures only hard or low-confidence cases reach the expensive models. On the test datasets, the cascaded system retains 97-99% of the strongest model's accuracy while reducing Time Per Output Token (TPOT). It requires only task-correctness labels and adapts to changes in the model pool without manual reconfiguration.

We present RLM-Cascade, a proxy-layer system that applies speculative decoding at the response level to reduce LLM API costs without requiring model architecture access or a shared vocabulary. A fast, inexpensive draft model generates a candidate response; a capable verify model accepts, enhances, or is bypassed entirely depending on a lightweight complexity router. On a real-world agentic coding workload (Claude Code), RLM-Cascade achieves a draft-use rate of 88.8% across 125 production requests, reducing API cost by 45.8% relative to a direct Opus baseline. Counter-intuitively, the proxy also reduces end-to-end latency: median response time is 2,026 ms versus 3,698 ms for Native Opus -- a 1.83X speedup at p50 -- because the SKIPPED path (DeepSeek only, no Opus call) dominates the workload distribution. Quality matches or exceeds the Opus baseline: 100% pass rate on a 20-task Code/Math/Instruct benchmark versus 95% for Native Opus. We further describe a rule-based complexity router that selects the SKIPPED path for simple agentic turns and a hybrid tool-call strategy that bypasses the speculative pipeline for schema-critical tool-selection turns. RLM-Cascade is deployed in production as an enterprise AI infrastructure component and published as open source with a live metrics dashboard and Prometheus endpoint.

As AI datacenters retire functional GPUs, vast quantities of still capable accelerators enter secondary markets. This paper investigates whether these retired GPUs can find a productive afterlife to form a DumpsterCluster that can serve modern LLM inference, and under what conditions such repurposing is economically viable and environmentally sustainable. We physically built a 128-GPU DumpsterCluster from scratch using only second-hand components and ran it for one year. At current market prices (\22K for the DumpsterCluster vs. 600K for an 8-GPU B200 system), the economic advantages are substantial. Through pipeline-parallel optimizations, our V100 based DumpsterCluster achieves competitive LLaMA-70B throughput, validating production viability. However, our deployment reveals critical context dependencies. Older GPUs consume significantly more energy per token, making total cost of ownership favorable only in regions with inexpensive electricity. Under grid-average carbon intensity, second-hand systems can produce approximately 4x higher total carbon emissions per token for 8B models, and over 40x for 70B models, compared to current-generation hardware. These findings show that GPU afterlife is not universally sustainable - hardware repurposing must be strategically coupled with low carbon energy sources. When deployed in regions with favourable energy economics and clean electricity, second-hand GPUs offer a viable pathway for expanding AI capacity while advancing affordability, energy security, and environmental responsibility.

Serverless computing has grown rapidly for serving Large Language Model (LLM) inference due to its pay-as-you-go pricing, fine-grained GPU usage, and rapid scaling. However, our analysis reveals that current serverless can effectively serve general LLM but fail with Low-Rank Adaptation (LoRA) inference due to three key limitations: 1) massive parameter redundancy among functions where 99% of weights are unnecessarily duplicated, 2) costly artifact loading latency beyond LLM loading, and 3) magnified resource contention when serving multiple LoRA LLMs. These inefficiencies lead to massive GPU wastage, increased Time-To-First-Token (TTFT), and high monetary costs. We propose ServerlessLoRA, a novel serverless inference system designed for faster and cheaper LoRA LLM serving. ServerlessLoRA enables secure backbone LLM sharing across isolated LoRA functions to reduce redundancy. We design a pre-loading method that pre-loads comprehensive LoRA artifacts to minimize cold-start latency. Furthermore, ServerlessLoRA employs contention aware batching and offloading to mitigate GPU resource conflicts during bursty workloads. Experiment on industrial workloads demonstrates that ServerlessLoRA reduces TTFT by up to 86% and cuts monetary costs by up to 89% compared to state-of-the-art LLM inference solutions.

Large language models (LLMs) have excelled in various applications, yet serving them at scale is challenging due to their substantial resource demands and high latency. Our real-world studies reveal that over 70% of user requests to LLMs have semantically similar counterparts, suggesting the potential for knowledge transfer among requests. However, naively caching and reusing past responses leads to a big quality drop. In this paper, we introduce IC-Cache, a caching system that enables live LLM capability augmentation to improve serving efficiency: by leveraging historical request-response pairs from larger models as in-context examples, IC-Cache empowers small LLMs to imitate and even exceed the compositional abilities (e.g., reasoning) of their larger counterparts, enabling selective offloading of requests to reduce cost and latency. Achieving this live augmentation at scale introduces intricate trade-offs between response quality, latency, and system throughput. For a new request, IC-Cache efficiently selects similar, high-utility examples to prepend them to the new request's input. At scale, it adaptively routes requests across LLMs of varying capabilities, accounting for response quality and serving loads. IC-Cache employs a cost-aware cache replay mechanism that refines example quality offline to maximize online cache utility and efficiency. Evaluations on millions of realistic requests demonstrate that IC-Cache improves LLM serving throughput by 1.4-5.9x and reduces latency by 28-71% without hurting response quality.

The exploration and application of Large Language Models (LLMs) is thriving. To reduce deployment costs, continuous batching has become an essential feature in current service frameworks. The effectiveness of continuous batching relies on an accurate estimate of the memory requirements of requests. However, due to the diversity in request output lengths, existing frameworks tend to adopt aggressive or conservative schedulers, which often result in significant overestimation or underestimation of memory consumption. Consequently, they suffer from harmful request evictions or prolonged queuing times, failing to achieve satisfactory throughput under strict Service Level Agreement (SLA) guarantees (a.k.a. goodput), across various LLM application scenarios with differing input-output length distributions. To address this issue, we propose a novel Past-Future scheduler that precisely estimates the peak memory resources required by the running batch via considering the historical distribution of request output lengths and calculating memory occupancy at each future time point. It adapts to applications with all types of input-output length distributions, balancing the trade-off between request queuing and harmful evictions, thereby consistently achieving better goodput. Furthermore, to validate the effectiveness of the proposed scheduler, we developed a high-performance LLM serving framework, LightLLM, that implements the Past-Future scheduler. Compared to existing aggressive or conservative schedulers, LightLLM demonstrates superior goodput, achieving up to 2-3times higher goodput than other schedulers under heavy loads. LightLLM is open source to boost the research in such direction (https://github.com/ModelTC/lightllm).

In retrieval-augmented generation and agentic LLM serving, prompts are assembled from independent segments into long contexts, making the prefill stage dominate per-request cost. Two directions have emerged to reduce this cost: position-independent caching (PIC) admits KV reuse for non-contiguous segments shared across requests, while hybrid-attention models cut computation by replacing most full-attention layers with linear attention. However, they cannot coexist: applying existing PIC methods to hybrid-attention models breaks down because per-token KV-cache reuse primitives do not transfer to the per-request recurrent state. We present Hypic, the first system to accelerate hybrid-attention LLM serving with position-independent caching. For linear-attention layers, we identify the segment-cumulative transition operator as the missing algebraic primitive and cache it alongside each segment's zero-start end-state, enabling near-exact and constant-time composition of independently cached segments. For the remaining full-attention layers, existing PIC methods also fail because linear layers do not expose the per-token hidden states needed for selective recomputation. We show that the largest deviations concentrate at segment beginnings and construct a small seam window that propagates hidden states through the hybrid-attention stack to repair cross-segment attention. Finally, Hypic introduces segment parallelism, which exploits PIC's segment-level self-containment to parallelize cache-miss prefill across instances, turning long cold requests into an accelerable workload. Evaluated across four hybrid-attention models and five workloads, Hypic reduces time-to-first-token by 3.25times on average and improves QPS by 1.66times over Prefix Cache, while preserving task quality with a 1.71-point gap from Full Recompute.

Deploying multiple models within shared GPU clusters is a key strategy to improve resource efficiency in large language model (LLM) serving. Existing multi-LLM serving systems improve GPU utilization at the cost of degraded inference performance, particularly time-to-first-token (TTFT). We attribute this degradation to the lack of awareness regarding future workload characteristics. In contrast, recent analyses have shown the strong periodicity and long-term predictability of real-world LLM serving workloads. In this paper, we propose one-for-many GPU prewarming, which proactively loads parameters from multiple models onto GPUs based on workload forecasts. These prewarmed weights enable the system to promptly instantiate serving instances upon encountering request bursts. We design and implement WarmServe, a multi-LLM serving system incorporating three key techniques: (1) a model placement algorithm that optimizes prewarming decisions to minimize cross-model prewarming interference, (2) a KV cache reservation strategy that repurposes idle KV cache space on running GPUs for prewarming new models, and (3) an efficient GPU memory switching mechanism for tensor management. Evaluation on real-world datasets shows that WarmServe reduces tail TTFT by up to 50.8times compared to the state-of-the-art autoscaling-based system, while supporting up to 2.5times higher request throughput than the GPU-sharing system.

Increasing demand for Large Language Models (LLMs) services imposes substantial deployment and computation costs on providers. LLM routing offers a cost-efficient solution by directing queries to the optimal LLM based on model and query features. However, existing works primarily focus on offline scenarios and struggle to adapt to online settings with high query volume and constrained token budgets. In this work, we introduce the first training-free algorithm for online routing scenarios. Our algorithm leverages approximate nearest neighbor search to efficiently estimate query features and performs a one-time optimization over a small set of initial queries to learn a routing strategy that guides future routing. We provide theoretical guarantees demonstrating that our algorithm achieves a competitive ratio of 1 - o(1) under natural assumptions, which is further validated by extensive experiments across 3 benchmark datasets and 8 baselines, showing an average improvement of 3.55times in overall performance, 1.85times in cost efficiency, and nearly 4.25times in throughput. Our code is available at https://github.com/fzwark/PORT.

Large Language Models (LLMs) have gained immense success in revolutionizing various applications, including content generation, search and recommendation, and AI-assisted operation. To reduce high training costs, Mixture-of-Experts (MoE) architecture has become a popular backbone for modern LLMs. However, despite the benefits, serving MoE-based LLMs experience severe memory inefficiency due to sparsely activated experts. Recent studies propose to offload inactive experts from GPU memory to CPU memory to improve the serving efficiency of MoE models. However, they either incur high inference latency or high model memory footprints due to coarse-grained designs. To tame the latency-memory trade-off in MoE serving, we present fMoE, a fine-grained expert offloading system for MoE serving that achieves low inference latency with memory efficiency. We design fMoE to extract fine-grained expert selection patterns from MoE models and semantic hints from input prompts to efficiently guide expert prefetching, caching, and offloading decisions. fMoE is prototyped on top of HuggingFace Transformers and deployed on a six-GPU testbed. Experiments with open-source MoE models and real-world workloads show that fMoE reduces inference latency by 47% and improves expert hit rate by 36% over state-of-the-art solutions.

We survey the large language model (LLM) serving area to understand the intricate dynamics between cost-efficiency and accuracy, which is magnified by the growing need for longer contextual understanding when deploying models at a massive scale. Our findings reveal that works in this space optimize along three distinct but conflicting goals: improving serving context length (C), improving serving accuracy (A), and improving serving performance (P). Drawing inspiration from the CAP theorem in databases, we propose a CAP principle for LLM serving, which suggests that any optimization can improve at most two of these three goals simultaneously. Our survey categorizes existing works within this framework. We find the definition and continuity of user-perceived measurement metrics are crucial in determining whether a goal has been met, akin to prior CAP databases in the wild. We recognize the CAP principle for LLM serving as a guiding principle, rather than a formal theorem, to inform designers of the inherent and dynamic trade-offs in serving models. As serving accuracy and performance have been extensively studied, this survey focuses on works that extend serving context length and address the resulting challenges.

Transformer-based pre-trained language models (PLMs) mostly suffer from excessive overhead despite their advanced capacity. For resource-constrained devices, there is an urgent need for a spatially and temporally efficient model which retains the major capacity of PLMs. However, existing statically compressed models are unaware of the diverse complexities between input instances, potentially resulting in redundancy and inadequacy for simple and complex inputs. Also, miniature models with early exiting encounter challenges in the trade-off between making predictions and serving the deeper layers. Motivated by such considerations, we propose a collaborative optimization for PLMs that integrates static model compression and dynamic inference acceleration. Specifically, the PLM is slenderized in width while the depth remains intact, complementing layer-wise early exiting to speed up inference dynamically. To address the trade-off of early exiting, we propose a joint training approach that calibrates slenderization and preserves contributive structures to each exit instead of only the final layer. Experiments are conducted on GLUE benchmark and the results verify the Pareto optimality of our approach at high compression and acceleration rate with 1/8 parameters and 1/19 FLOPs of BERT.

This paper presents MoE-Infinity, a cost-efficient mixture-of-expert (MoE) serving system that realizes activation-aware expert offloading. MoE-Infinity features sequence-level expert activation tracing, a new approach adept at identifying sparse activations and capturing the temporal locality of MoE inference. By analyzing these traces, MoE-Infinity performs novel activation-aware expert prefetching and caching, substantially reducing the latency overheads usually associated with offloading experts for improved cost performance. Extensive experiments in a cluster show that MoE-Infinity outperforms numerous existing systems and approaches, reducing latency by 4 - 20X and decreasing deployment costs by over 8X for various MoEs. MoE-Infinity's source code is publicly available at https://github.com/TorchMoE/MoE-Infinity

Multi-turn LLM agents interleave model calls with external tool invocations, shifting serving from stateless request processing to stateful program execution. Serving these workloads requires scheduling, KV-cache management, and routing policies that use program-level context, including turn dependencies, tool-induced gaps, and reusable KV state. Evaluating such policies directly on real systems is costly, since each design point may require dedicated accelerator time across arrival rates, model scales, serving-instance counts, and memory hierarchies. Simulation offers a scalable alternative, but existing LLM serving simulators target stateless request-level workloads and therefore omit the core dynamics of agent serving: multi-turn program execution, cross-turn cache locality, and KV-cache residency during tool gaps. We present AGENTSERVESIM, a hardware-aware simulator for multi-turn LLM agent serving. AGENTSERVESIM evaluates serving policies at program granularity through composable modules: a Program Orchestrator preserves program identity and turn order, a Tool Simulator materializes tool-induced gaps, a Session-Aware Router maintains program-to-instance affinity for cache-aware dispatch, and a KV Residency Model tracks policy-defined KV placement across HBM, host DRAM/CXL, and eviction. Across real serving deployments and hardware configurations, AGENTSERVESIM reproduces real-system behavior within 6% error across key performance metrics while running entirely on commodity CPUs. These results show that AGENTSERVESIM enables controlled, repeatable exploration of agent-serving policies without requiring exhaustive deployment on costly accelerators.

Large Language Models (LLMs) demonstrate exceptional performance but entail significant memory and computational costs, restricting their practical deployment. While existing INT4/INT8 quantization reduces these costs, they often degrade accuracy or lack optimal efficiency. INT6 quantization offers a superior trade-off between model accuracy and inference efficiency, but lacks hardware support in modern GPUs, forcing emulation via higher-precision arithmetic units that limit acceleration. In this paper, we propose FlexQ, a novel post-training INT6 quantization framework combining algorithmic innovation with system-level optimizations. FlexQ employs uniform 6-bit weight quantization across all layers, with adaptive retention of 8-bit activations in layers identified through layer-wise sensitivity analysis. To maximize hardware efficiency, we develop a specialized high-performance GPU kernel supporting matrix multiplication for W6A6 and W6A8 representations via Binary Tensor Core (BTC) equivalents, effectively bypassing the lack of native INT6 tensor cores. Evaluations on LLaMA models show FlexQ maintains near-FP16 accuracy, with perplexity increases of no more than 0.05. The proposed kernel achieves an average 1.39times speedup over ABQ-LLM on LLaMA-2-70B linear layers. End-to-end, FlexQ delivers 1.33times inference acceleration and 1.21times memory savings over SmoothQuant. Code is released at https://github.com/FlyFoxPlayer/FlexQ.

Speculative inference is a promising paradigm employing small speculative models (SSMs) as drafters to generate draft tokens, which are subsequently verified in parallel by the target large language model (LLM). This approach enhances the efficiency of inference serving by reducing LLM inference latency and costs while preserving generation quality. However, existing speculative methods face critical challenges, including inefficient resource utilization and limited draft acceptance, which constrain their scalability and overall effectiveness. To overcome these obstacles, we present CoSine, a novel speculative inference system that decouples sequential speculative decoding from parallel verification, enabling efficient collaboration among multiple nodes. Specifically, CoSine routes inference requests to specialized drafters based on their expertise and incorporates a confidence-based token fusion mechanism to synthesize outputs from cooperating drafters, ensuring high-quality draft generation. Additionally, CoSine dynamically orchestrates the execution of speculative decoding and verification in a pipelined manner, employing batch scheduling to selectively group requests and adaptive speculation control to minimize idle periods. By optimizing parallel workflows through heterogeneous node collaboration, CoSine balances draft generation and verification throughput in real-time, thereby maximizing resource utilization. Experimental results demonstrate that CoSine achieves superior performance compared to state-of-the-art speculative approaches. Notably, with equivalent resource costs, CoSine achieves up to a 23.2% decrease in latency and a 32.5% increase in throughput compared to baseline methods.

Coding agents repeatedly search, navigate, and retain context from evolving repositories, but disconnected indexes, language servers, and task-local histories force repeated discovery and obscure lifecycle costs. CodeNib builds reusable lexical, dense, and structural views per repository commit, maps outputs to repository-relative source ranges, maintains selected views across edits, and serves ranked search, symbol navigation, and bounded context through one runtime. Across 100 snapshots, we map quality-cost frontiers across the repository-context lifecycle. When outputs match an independent rebuild, graph and vector updates are 8.7times and 25.4times faster at the median. On the static-navigation subset matching normalized live-server locations (63% of 1,000 requests), the median per-request live/static latency ratio is 4.7times. Across five models, selected context policies preserve localization with 50--87% fewer trajectory tokens than paired grep/read. Together, these results support multi-view repository-context serving with explicit, operation-specific validity boundaries.

Streaming video generation is emerging as a new serving workload in which users interact with long-lived sessions that generate video progressively, chunk by chunk. Unlike offline video generation or typical LLM serving, streaming video generation must preserve session state across active and idle periods, repeatedly schedule ongoing sessions, and deliver each chunk under a tight latency target. This creates two key serving challenges in multi-user, multi-GPU environments: session duration heterogeneity, where long-running sessions make placement decisions suboptimal over time, and temporal user-demand heterogeneity, where the number of active sessions fluctuates sharply across bursts and idle periods. We present TurboServe, the first serving system designed specifically for streaming video generation workloads. TurboServe formulates serving as an online scheduling problem that jointly coordinates session placement and GPU provisioning. Its closed-loop scheduling algorithm combines a migration-aware placement controller, which rebalances sessions across GPUs to reduce the maximum per-chunk latency, with a load-driven autoscaling controller, which adapts the GPU budget to workload variation for improved cost efficiency. To support these decisions at runtime, TurboServe implements coalesced chunk processing for batching concurrent active sessions on the same GPU, GPU-CPU offloading for session suspension and resumption, and NCCL-based GPU-GPU migration for online rebalancing. We evaluate TurboServe on real-world production traces from Shengshu Technology across multiple model sizes and GPU clusters with up to 64 NVIDIA B300 GPUs. Compared with baseline serving configurations, TurboServe reduces worst-case per-chunk latency by 37.5% and total GPU operating cost by 37.2% on average. Our code is publicly available at https://github.com/shengshu-ai/TurboServe.

The rapid evolution of large language models (LLMs) has unlocked their capabilities in advanced reasoning tasks like mathematical problem-solving, code generation, and legal analysis. Central to this progress are inference-time reasoning algorithms, which refine outputs by exploring multiple solution paths, at the cost of increasing compute demands and response latencies. Existing serving systems fail to adapt to the scaling behaviors of these algorithms or the varying difficulty of queries, leading to inefficient resource use and unmet latency targets. We present Dynasor, a system that optimizes inference-time compute for LLM reasoning queries. Unlike traditional engines, Dynasor tracks and schedules requests within reasoning queries and uses Certaindex, a proxy that measures statistical reasoning progress based on model certainty, to guide compute allocation dynamically. Dynasor co-adapts scheduling with reasoning progress: it allocates more compute to hard queries, reduces compute for simpler ones, and terminates unpromising queries early, balancing accuracy, latency, and cost. On diverse datasets and algorithms, Dynasor reduces compute by up to 50% in batch processing and sustaining 3.3x higher query rates or 4.7x tighter latency SLOs in online serving.

LLMs are widely adopted in production, pushing inference systems to their limits. Disaggregated LLM serving (e.g., PD separation and KV state disaggregation) improves scalability and cost efficiency, but it also turns KV into an explicit payload crossing network and storage boundaries, making KV a dominant end-to-end bottleneck. Existing KV compression are typically static runtime configurations, despite production service context varies over time in workload mix, bandwidth, and SLO/quality budgets. As a result, a fixed choice can be suboptimal or even increase latency. We present KVServe, the first service-aware and adaptive KV communication compression framework for disaggregated LLM serving: KVServe (1) unifies KV compression into a modular strategy space with new components and cross-method recomposition; (2) introduces Bayesian Profiling Engine that efficiently searches this space and distills a 3D Pareto candidate set, reducing 50times offline search overhead; and (3) deploys a Service-Aware Online Controller that combines an analytical latency model with a lightweight bandit to select profiles under constraints and correct offline-to-online mismatch. Integrated into vLLM and evaluated across datasets, models, GPUs and networks, KVServe achieves up to 9.13times JCT speedup in PD-separated serving and up to 32.8times TTFT reduction in KV-disaggregated serving.

Heterogeneous LLM serving stacks split scheduling into two layers that optimize in isolation: model routers pick a model from quality and cost signals while ignoring instance load, and serving load balancers optimize queues while ignoring quality. We present RouteBalance, a serving-aware scheduling layer that fuses both into a single online assignment over concrete model instances, jointly trading off quality, latency, and cost. A batched in-process predictor stack and dead-reckoned instance state keep the joint decision cheap on the request hot path (approx32 ms at 12 req/s). On a 13-instance, 28-GPU heterogeneous cluster serving four model sizes, a single deployed RouteBalance stack traces the upper region of the three-way quality-cost-throughput frontier. Sweeping one weight vector reaches both the highest routing-decision quality (DeepEval 0.419, +0.013 over the strongest baseline, 95% CI [{+}0.005,{+}0.022]; the ordering holds when a second judge re-scores the actually served text) and, at its cost-priority corner, per-request cost that ties the cheapest baseline. With router engineering equalized against concurrent-scoring baseline variants we build, its balanced preset serves at 2.8 s and 30 req/s, leading 2.6 to 4.1times ahead of enhanced BEST-Route at high load. (Deploying those routers as published, one serial scoring call per request, makes them collapse 23times under load, a deployment-architecture effect we isolate separately, not the routing result.) A four-arm isolation shows the benefit follows from pricing latency at model-selection time; the learned predictors contribute calibration and SLO headroom rather than the headline frontier. Code: https://github.com/AKafakA/route-balance

Linear attention has recently gained significant attention for long-context inference due to its constant decoding cost with respect to context length. However, existing serving systems typically serve linear attention by recurrently computing and updating a large linear attention state in every decoding step. Since the state is much larger than the per-token key and value, recurrent decoding incurs substantial memory access and becomes inefficient for serving linear attention. In this paper, we propose KVBuffer, an IO-aware serving mechanism for linear attention. By buffering recent keys and values, KVBuffer enables serving systems to compute linear attention outputs in more flexible and memory-efficient ways. For decoding, KVBuffer enables chunkwise computation, which reduces average memory access and decoding latency by deferring state updates and applying them in batch. For speculative decoding, KVBuffer verifies draft tokens in parallel and avoids storing temporary states. For short contexts, KVBuffer computes attention outputs directly from buffered keys and values, without creating or updating the linear attention state. We implement KVBuffer in SGLang for Qwen3-Next. Our evaluations show that KVBuffer can reduce linear attention decoding latency by up to 45.17% and increase the maximum number of serving requests by 5x for speculative decoding when verifying four draft tokens.

Agentic workflows have emerged as a powerful paradigm for solving complex, multi-stage tasks, but serving them at scale is computationally expensive given the many LLM inferences that each request must pass through. Configuration selection, or the cost-aware assignment of workflow agents to specific LLMs, can reduce these costs, but existing approaches bind configuration decisions before request execution, making them ill-suited for the heterogeneous and lengthy execution of workflows. Specifically, system loads can fluctuate rapidly and substantially during a request's lifetime, causing fixed configurations to quickly become suboptimal. We present Aragog, a system that progressively adapts a request's configuration throughout its execution to match runtime dynamics. To make this practical despite the massive space of workflow configurations, Aragog decouples the problem into two core elements -- a one-time routing step that identifies all accuracy-preserving configurations, and a cheap per-stage scheduler that selects among them using up-to-date system observations -- and introduces novel strategies to accelerate each. Across diverse workflows and model families, Aragog increases maximum serving throughput by 50.0--217.0\% and reduces median latency by 32.5--78.9\% at peak request rates, while maintaining accuracy comparable to the most expensive configurations.

Large Multimodal Models (LMMs) extend Large Language Models (LLMs) by handling diverse inputs such as images, audio, and video, but at the cost of adding a multimodal encoding stage that increases both computational and memory overhead. This step negatively affects key Service Level Objectives (SLOs), such as time to first token (TTFT) and time per output token (TPOT). We introduce Encode-Prefill-Decode (EPD) Disaggregation, a novel framework that separates the encoding, prefill, and decode stages onto dedicated resources. Unlike current systems, which bundle encoding and prefill together, our approach decouples these steps, unlocking new opportunities and optimizations. These include a mechanism to cache multimedia tokens for efficient transfer, a novel way to parallelize the encoding load within a request, a module for optimal resource allocation for disaggregated serving, and a novel role-switching method to handle changing workload characteristics. Experimental evaluations with popular LMMs show substantial gains in memory efficiency (up to 15x lower peak memory utilization), batch sizes (up to 22x larger), 10x more images per request, and 2.2x larger KV caches. Furthermore, it leads to significant improvements in SLO attainment (up to 90-100% improvement) and TTFT (up to 71% reduction), compared to systems that do not disaggregate. The code is available at https://github.com/vbdi/epdserve.

We study linear dueling bandits in volatile environments characterized by the simultaneous presence of post-serving contexts, delayed feedback, and adversarial corruption. Feedback is subject to unknown stochastic or adversarial delays and a cumulative corruption budget C. To address these challenges, we propose e RCDP-UCB, which integrates a learned approximator that predicts post-serving contexts from pre-serving information. It further employs an adaptive weighting strategy that clips feature vectors to mitigate the impact of corrupted and delayed observations simultaneously. Under standard regularity conditions and a parametric post-serving mapping, we rigorously establish that our algorithm is delay-regime-agnostic, achieving a regret upper bound of mathcal{O}(d(T + C + D)), where d is the total feature dimension and D encapsulates the delay complexity. Crucially, our analysis reveals an additive cost structure between corruption and delay, avoiding the multiplicative degradation typical of prior works. We further establish lower bounds that nearly match our upper bounds up to a d factor for adversarial delays in the absence of post-serving contexts. Code is available at https://github.com/youngmin0oh/rcdp-public.

Large multimodal models (LMMs) demonstrate impressive capabilities in understanding images, videos, and audio beyond text. However, efficiently serving LMMs in production environments poses significant challenges due to their complex architectures and heterogeneous characteristics across their multi-stage inference pipelines. We present the first comprehensive systems analysis of two prominent LMM architectures, decoder-only and cross-attention, across six representative open-source models, revealing key systems design implications. We also present an in-depth analysis of production LMM inference traces, uncovering unique workload characteristics, including variable, heavy-tailed request distributions and bursty traffic patterns. Based on these insights, we propose ModServe, a modular LMM serving system that decouples stages for independent optimization and adaptive scaling. ModServe dynamically reconfigures stages and handles bursty traffic with modality-aware scheduling and autoscaling to meet tail latency SLOs while minimizing costs. ModServe achieves 3.3-5.5x higher throughput (leading to 25-41.3% cost saving) while meeting SLOs on a 128-GPU cluster with production traces.

Large Language Models (LLMs) are revolutionizing numerous industries, but their substantial computational demands create challenges for efficient deployment, particularly in cloud environments. Traditional approaches to inference serving often struggle with resource inefficiencies, leading to high operational costs, latency issues, and limited scalability. This article explores how Cloud Native technologies, such as containerization, microservices, and dynamic scheduling, can fundamentally improve LLM inference serving. By leveraging these technologies, we demonstrate how a Cloud Native system enables more efficient resource allocation, reduces latency, and enhances throughput in high-demand scenarios. Through real-world evaluations using Kubernetes-based autoscaling, we show that Cloud Native architectures can dynamically adapt to workload fluctuations, mitigating performance bottlenecks while optimizing LLM inference serving performance. This discussion provides a broader perspective on how Cloud Native frameworks could reshape the future of scalable LLM inference serving, offering key insights for researchers, practitioners, and industry leaders in cloud computing and artificial intelligence.

Novel View Synthesis (NVS) is the task of generating new images of a scene from viewpoints that were not part of the original input. Diffusion-based NVS can generate high-quality, temporally consistent images, however, remains computationally prohibitive. Conversely, regression-based NVS offers suboptimal generation quality despite requiring significantly lower compute; leaving the design objective of a high-quality, inference-efficient NVS framework an open challenge. To close this critical gap, we present Sphinx, a training-free hybrid inference framework that achieves diffusion-level fidelity at a significantly lower compute. Sphinx proposes to use regression-based fast initialization to guide and reduce the denoising workload for the diffusion model. Additionally, it integrates selective refinement with adaptive noise scheduling, allowing more compute to uncertain regions and frames. This enables Sphinx to provide flexible navigation of the performance-quality trade-off, allowing adaptation to latency and fidelity requirements for dynamically changing inference scenarios. Our evaluation shows that Sphinx achieves an average 1.8x speedup over diffusion model inference with negligible perceptual degradation of less than 5%, establishing a new Pareto frontier between quality and latency in NVS serving.

The growing demand for Large Language Models (LLMs) across diverse applications has prompted a paradigm shift in the design of deep learning serving systems. Deploying LLMs, especially in multi-tenant environments, presents considerable challenges due to their high computational and memory demands. We present BlockLLM, a serving system that exploits the potential of sharing components among fine-tuned LLM models to offer an efficient and flexible solution for LLM workloads. BlockLLM partitions the models into finer-grained blocks to enable the reuse of model components and independent provisioning to improve the computation efficiency. BlockLLM consists of an offline block zoo, for storing the blocks, and an online system to serve the requests through chains of blocks. It offers multi-fold flexibility: (1) Adaptive assembly of block chains on-the-fly is achieved with the help of equivalence evaluation among blocks in the zoo. (2) We enable per-block batch size and configure best-effort KV cache coordination at individual block level. (3) We adopt speculative execution and locality-aware block placement to mitigate the communication costs from dynamic block resource allocation. Our evaluation demonstrates that BlockLLM reduces memory and storage footprints and improves computation efficiency, outperforming existing serving approach in 95\%ile latency and GPU utilization by 33.5\% and 20.1\%, respectively.

Six-bit quantization (FP6) can effectively reduce the size of large language models (LLMs) and preserve the model quality consistently across varied applications. However, existing systems do not provide Tensor Core support for FP6 quantization and struggle to achieve practical performance improvements during LLM inference. It is challenging to support FP6 quantization on GPUs due to (1) unfriendly memory access of model weights with irregular bit-width and (2) high runtime overhead of weight de-quantization. To address these problems, we propose TC-FPx, the first full-stack GPU kernel design scheme with unified Tensor Core support of float-point weights for various quantization bit-width. We integrate TC-FPx kernel into an existing inference system, providing new end-to-end support (called FP6-LLM) for quantized LLM inference, where better trade-offs between inference cost and model quality are achieved. Experiments show that FP6-LLM enables the inference of LLaMA-70b using only a single GPU, achieving 1.69x-2.65x higher normalized inference throughput than the FP16 baseline. The source code will be publicly available soon.

The rapid advancement of Large Language Models (LLMs) has driven the need for more efficient serving strategies. In this context, efficiency refers to the proportion of requests that meet their Service Level Objectives (SLOs), particularly for Time To First Token (TTFT) and Time Between Tokens (TBT). However, existing systems often prioritize one metric at the cost of the other. We present Ascendra, an LLM serving system designed to meet both TTFT and TBT SLOs simultaneously. The core insight behind Ascendra is that a request's urgency evolves as it approaches its deadline. To leverage this, Ascendra partitions GPU resources into two types of instances: low-priority and high-priority. Low-priority instances maximize throughput by processing requests out of arrival order, but at the risk of request starvation. To address this, Ascendra employs a performance model to predict requests at risk of missing their SLOs and proactively offloads them to high-priority instances. High-priority instances are optimized for low-latency execution and handle urgent requests nearing their deadlines. This partitioned architecture enables Ascendra to effectively balance high throughput and low latency. Extensive evaluation shows that Ascendra improves system throughput by up to 1.7x compared to vLLM and Sarathi-Serve while meeting both TTFT and TBT SLOs.

Retrieval-augmented language models (RaLM) have demonstrated the potential to solve knowledge-intensive natural language processing (NLP) tasks by combining a non-parametric knowledge base with a parametric language model. Instead of fine-tuning a fully parametric model, RaLM excels at its low-cost adaptation to the latest data and better source attribution mechanisms. Among various RaLM approaches, iterative RaLM delivers a better generation quality due to a more frequent interaction between the retriever and the language model. Despite the benefits, iterative RaLM usually encounters high overheads due to the frequent retrieval step. To this end, we propose RaLMSpec, a speculation-inspired framework that provides generic speed-up over iterative RaLM while preserving the same model outputs through speculative retrieval and batched verification. By further incorporating prefetching, optimal speculation stride scheduler, and asynchronous verification, RaLMSpec can automatically exploit the acceleration potential to the fullest. For naive iterative RaLM serving, extensive evaluations over three language models on four downstream QA datasets demonstrate that RaLMSpec can achieve a speed-up ratio of 1.75-2.39x, 1.04-1.39x, and 1.31-1.77x when the retriever is an exact dense retriever, approximate dense retriever, and sparse retriever respectively compared with the baseline. For KNN-LM serving, RaLMSpec can achieve a speed-up ratio up to 7.59x and 2.45x when the retriever is an exact dense retriever and approximate dense retriever, respectively, compared with the baseline.

GPU Confidential Computing (GPU-CC) now preserves GPU-local performance: on NVIDIA B300, BF16 matmul runs at 0.998x of non-confidential performance. Yet LLM serving under Intel TDX plus GPU-CC still loses 13-27% of throughput, and KV-cache restore latency can more than double. This paper studies that gap on two Blackwell platforms, RTX Pro 6000 and B300 HGX, and identifies its dominant cause: the confidential VM-GPU bridge, not GPU compute. We find that GPU-CC turns host/device movement into a serialized, high-setup-cost channel. Secure copies do not gain CUDA-stream concurrency within a context, asynchronous transfers block at the runtime boundary, and small crossings pay a fixed toll. This violates the assumptions of modern inference runtimes, where DMA is expected to be cheap, concurrent, and asynchronous. In vLLM dense decode, the gap closes around 44x-slower small alloc-and-copy operations; targeted patches reject alternative explanations. A scheduling flag recovers 57% of the gap, while a worker-thread drain recovers up to 92% in qualified high-concurrency runs. The same bridge model explains a +131% KV-restore penalty and a 34x model-load slowdown. Blackwell also changes the confidential tenancy unit. We qualify confidential multi-GPU NVSwitch tenants on B300, including 510 GB/s NVLink P2P inside a CVM and concurrent isolated tenants, and identify the remaining fabric-attestation gap for production confidential AI platforms.

Large Language Models (LLMs) are increasingly deployed in both latency-sensitive online services and cost-sensitive offline workloads. Co-locating these workloads on shared serving instances can improve resource utilization, but directly applying this approach to Prefill/Decode (P/D) disaggregated systems introduces severe load imbalance, as fluctuating request mixes alter the intrinsic P/D ratio. Existing dynamic adjustment techniques cannot keep up with the bursty traffic patterns of online services. We propose a latency-constraint disaggregated architecture, which separates cluster resources into latency-strict and latency-relaxed pools based on task latency requirements. This design enables flexible placement of offline decode tasks, mitigating P/D imbalance while preserving online performance. To fully exploit this flexibility, we propose (1) a bottleneck-based scheduler guided by a Roofline-based performance model for performance bottleneck based scheduling, and (2) a fast preemption mechanism that strictly enforces Service Level Objectives (SLOs) for online requests. Experiments on real-world traces show that compared to existing offline system approaches, our method improves offline throughput by up to 3x, while maintaining online request SLOs.

Edge deployment of large language models (LLMs) increasingly relies on libraries of lightweight LoRA adapters, yet GPU/DRAM can keep only a small resident subset at a time. Serving a request through a non-resident adapter requires paging its weights from storage, incurring measurable latency. This creates a two-timescale online control problem: on a slow timescale, the system selects which adapters remain resident in fast memory, while on a fast timescale it routes each request to an adapter whose context-dependent utility is unknown a priori. The two decisions are tightly coupled: the cache determines the cost of exploration, and the router determines which adapters receive informative feedback. We formulate this joint caching-and-routing problem as a two-timescale contextual bandit and propose POLAR (Paging and Online Learning for Adapter Routing). POLAR pairs a cache-aware LinUCB router with an epoch-based cache controller. We study two variants. A fixed-epoch version provides a robust baseline with worst-case regret guarantees under arbitrary contexts. An epoch-doubling version, POLAR+, adds forced exploration and improved cache optimization to achieve mathcal{O}(dNT+KT) sublinear regret under stochastic regularity and cacheability conditions, where N is the adapter count, K the cache size, d the context dimension, and T the horizon. The routing term matches the standard contextual-bandit rate up to logarithmic factors, showing that the memory hierarchy does not fundamentally slow routing learning. Experiments using 15 real LoRA adapters for Qwen2.5-7B together with measured GPU paging latencies show that adaptive cache control substantially outperforms non-adaptive baselines and exhibits scaling trends consistent with the theory.

Serving numerous users and requests concurrently requires good fairness in Large Language Models (LLMs) serving system. This ensures that, at the same cost, the system can meet the Service Level Objectives (SLOs) of more users , such as time to first token (TTFT) and time between tokens (TBT), rather than allowing a few users to experience performance far exceeding the SLOs. To achieve better fairness, the preemption-based scheduling policy dynamically adjusts the priority of each request to maintain balance during runtime. However, existing systems tend to overly prioritize throughput, overlooking the overhead caused by preemption-induced context switching, which is crucial for maintaining fairness through priority adjustments. In this work, we identify three main challenges that result in this overhead. 1) Inadequate I/O utilization. 2) GPU idleness. 3) Unnecessary I/O transmission during multi-turn conversations. Our key insight is that the block-based KV cache memory policy in existing systems, while achieving near-zero memory waste, leads to discontinuity and insufficient granularity in the KV cache memory. To respond, we introduce FastSwitch, a fairness-aware serving system that not only aligns with existing KV cache memory allocation policy but also mitigates context switching overhead. Our evaluation shows that FastSwitch outperforms the state-of-the-art LLM serving system vLLM with speedups of 1.4-11.2x across different tail TTFT and TBT.

Inference serving for large language models (LLMs) is the key to unleashing their potential in people's daily lives. However, efficient LLM serving remains challenging today because the requests are inherently heterogeneous and unpredictable in terms of resource and latency requirements, as a result of the diverse applications and the dynamic execution nature of LLMs. Existing systems are fundamentally limited in handling these characteristics and cause problems such as severe queuing delays, poor tail latencies, and SLO violations. We introduce Llumnix, an LLM serving system that reacts to such heterogeneous and unpredictable requests by runtime rescheduling across multiple model instances. Similar to context switching across CPU cores in modern operating systems, Llumnix reschedules requests to improve load balancing and isolation, mitigate resource fragmentation, and differentiate request priorities and SLOs. Llumnix implements the rescheduling with an efficient and scalable live migration mechanism for requests and their in-memory states, and exploits it in a dynamic scheduling policy that unifies the multiple rescheduling scenarios elegantly. Our evaluations show that Llumnix improves tail latencies by an order of magnitude, accelerates high-priority requests by up to 1.5x, and delivers up to 36% cost savings while achieving similar tail latencies, compared against state-of-the-art LLM serving systems. Llumnix is publicly available at https://github.com/AlibabaPAI/llumnix.

Quantization can accelerate large language model (LLM) inference. Going beyond INT8 quantization, the research community is actively exploring even lower precision, such as INT4. Nonetheless, state-of-the-art INT4 quantization techniques only accelerate low-batch, edge LLM inference, failing to deliver performance gains in large-batch, cloud-based LLM serving. We uncover a critical issue: existing INT4 quantization methods suffer from significant runtime overhead (20-90%) when dequantizing either weights or partial sums on GPUs. To address this challenge, we introduce QoQ, a W4A8KV4 quantization algorithm with 4-bit weight, 8-bit activation, and 4-bit KV cache. QoQ stands for quattuor-octo-quattuor, which represents 4-8-4 in Latin. QoQ is implemented by the QServe inference library that achieves measured speedup. The key insight driving QServe is that the efficiency of LLM serving on GPUs is critically influenced by operations on low-throughput CUDA cores. Building upon this insight, in QoQ algorithm, we introduce progressive quantization that can allow low dequantization overhead in W4A8 GEMM. Additionally, we develop SmoothAttention to effectively mitigate the accuracy degradation incurred by 4-bit KV quantization. In the QServe system, we perform compute-aware weight reordering and take advantage of register-level parallelism to reduce dequantization latency. We also make fused attention memory-bound, harnessing the performance gain brought by KV4 quantization. As a result, QServe improves the maximum achievable serving throughput of Llama-3-8B by 1.2x on A100, 1.4x on L40S; and Qwen1.5-72B by 2.4x on A100, 3.5x on L40S, compared to TensorRT-LLM. Remarkably, QServe on L40S GPU can achieve even higher throughput than TensorRT-LLM on A100. Thus, QServe effectively reduces the dollar cost of LLM serving by 3x. Code is available at https://github.com/mit-han-lab/qserve.

Practical large language model (LLM) services may involve a long system prompt, which specifies the instructions, examples, and knowledge documents of the task and is reused across numerous requests. However, the long system prompt causes throughput/latency bottlenecks as the cost of generating the next token grows w.r.t. the sequence length. This paper aims to improve the efficiency of LLM services that involve long system prompts. Our key observation is that handling these system prompts requires heavily redundant memory accesses in existing causal attention computation algorithms. Specifically, for batched requests, the cached hidden states (i.e., key-value pairs) of system prompts are transferred from off-chip DRAM to on-chip SRAM multiple times, each corresponding to an individual request. To eliminate such a redundancy, we propose RelayAttention, an attention algorithm that allows reading these hidden states from DRAM exactly once for a batch of input tokens. RelayAttention is a free lunch: it maintains the generation quality while requiring no model retraining, as it is based on a mathematical reformulation of causal attention.

Reducing the inference latency of large language models (LLMs) is crucial, and speculative decoding (SD) stands out as one of the most effective techniques. Rather than letting the LLM generate all tokens directly, speculative decoding employs effective proxies to predict potential outputs, which are then verified by the LLM without compromising the generation quality. Yet, deploying SD in real online LLM serving systems (with continuous batching) does not always yield improvement -- under higher request rates or low speculation accuracy, it paradoxically increases latency. Furthermore, there is no best speculation length work for all workloads under different system loads. Based on the observations, we develop a dynamic framework SmartSpec. SmartSpec dynamically determines the best speculation length for each request (from 0, i.e., no speculation, to many tokens) -- hence the associated speculative execution costs -- based on a new metric called goodput, which characterizes the current observed load of the entire system and the speculation accuracy. We show that SmartSpec consistently reduces average request latency by up to 3.2x compared to non-speculative decoding baselines across different sizes of target models, draft models, request rates, and datasets. Moreover, SmartSpec can be applied to different styles of speculative decoding, including traditional, model-based approaches as well as model-free methods like prompt lookup and tree-style decoding.

Despite the rapid advancements of large language models (LLMs), LLM serving systems remain memory-intensive and costly. The key-value (KV) cache, which stores KV tensors during autoregressive decoding, is crucial for enabling low-latency, high-throughput LLM inference serving. In this survey, we focus on system-aware KV infrastructure for serving LLMs (abbreviated as sKis). We revisit recent work from a system behavior perspective, organizing existing efforts into three dimensions: execution and scheduling (temporal), placement and migration (spatial), and representation and retention (structural). Furthermore, we analyze cross-behavior co-design affinity and behavior-objective links, highlighting future opportunities. Our work systematizes a rapidly evolving area, providing a foundation for understanding and innovating KV cache designs in modern LLM serving infrastructure.

Mixture-of-Experts (MoE) model architecture has emerged as a promising solution for scaling transformer models efficiently, offering sparse activation that reduces computational costs while increasing model capacity. However, as MoE models scale, they need to be distributed across GPU devices, thus face critical performance bottlenecks due to their large memory footprint. Expert parallelism distributes experts across GPUs, however, faces key challenges including an unbalanced token routing and expert activation, resulting in communication tail latency and processing inefficiencies. While existing solutions address some of these issues, they fail to resolve the dual challenges of load imbalance and communication skew. The imbalance in token processing load across experts causes uneven processing times on different GPUs, while communication skew between GPUs leads to unbalanced inter-GPU data transfers. These factors degrade the performance of MoE models by increasing tail latency and reducing overall throughput. To address these limitations, we propose an Integer Linear Programming (ILP) formulation to optimize expert placement by jointly considering token load, communication, and computation costs. We exploit the property that there is a token routing dependency across layers, where tokens routed to a specific expert in one layer are likely to be routed to a limited set of experts in the subsequent layer. Our solution, MoETuner, offers an optimal expert-to-GPU assignment that minimizes inter-GPU token routing costs and balances token processing across devices, thereby reducing tail latency and end-to-end execution time. Experimental results demonstrate 9.3% and 17.5% of end-to-end speedups for single-node and multi-node inference respectively, showcasing the potential of our ILP-based optimization for offering expert parallel solutions for next-generation MoEs.

Imitation learning from human demonstrations has shown impressive performance in robotics. However, most results focus on table-top manipulation, lacking the mobility and dexterity necessary for generally useful tasks. In this work, we develop a system for imitating mobile manipulation tasks that are bimanual and require whole-body control. We first present Mobile ALOHA, a low-cost and whole-body teleoperation system for data collection. It augments the ALOHA system with a mobile base, and a whole-body teleoperation interface. Using data collected with Mobile ALOHA, we then perform supervised behavior cloning and find that co-training with existing static ALOHA datasets boosts performance on mobile manipulation tasks. With 50 demonstrations for each task, co-training can increase success rates by up to 90%, allowing Mobile ALOHA to autonomously complete complex mobile manipulation tasks such as sauteing and serving a piece of shrimp, opening a two-door wall cabinet to store heavy cooking pots, calling and entering an elevator, and lightly rinsing a used pan using a kitchen faucet. Project website: https://mobile-aloha.github.io

Long-context modeling is a pivotal capability for Large Language Models, yet the quadratic complexity of attention remains a critical bottleneck, particularly during the compute-intensive prefilling phase. Our previous work, FlashPrefill, mitigates this cost through instantaneous pattern discovery and max-based dynamic thresholding; however, it remains an algorithmic prototype that is still distant from production deployment. In this paper, we present FlashPrefill V2, which evolves FlashPrefill from a prototype toward practical long-context serving along three dimensions. First, we introduce a mean correction term that effectively suppresses the approximation error, keeping performance degradation manageable even at extreme sparsity levels. Second, we redesign the sparse attention operator with PackGQA memory access, warp specialization, and pingpong pipelining, fully aligning with the latest FlashAttention-3/4 implementations and supporting FP8 inference to meet practical quantization requirements. Third, FlashPrefill V2 natively supports paged KV cache and continuous batching, allowing integration as an attention backend in modern inference frameworks such as SGLang. Extensive evaluations on NVIDIA H20 GPUs---among the most widely deployed inference accelerators---demonstrate that FlashPrefill V2 delivers up to 47.26x and 27.19x speedups over FlashAttention-2 at 128K context length under FP8 and BF16 precision, respectively, and, in FP8, still achieves a 30.49x speedup against an FA3/4-aligned dense baseline.

Increasingly, LLM inference services proxy client requests to engine replicas distributed globally. Load-balancing policies must jointly account for factors including KV-cache locality, replica load, and variable network latency when optimizing for metrics like latency and TTFT. However, existing systems only evaluate a subset of these factors in their cost model, leading to uneven concentrations of load and KV-cache across replicas. We present GORGO, a proxy architecture that holistically factors network latency, prefill cost, and queueing delay using tunable parameters. Since open-source chat datasets such as LMSYS-Chat1M and WildChat-4.8M lack long-context, high prefix-reuse data, we release a synthetic dataset, ART-Chat-2.5M, from long-context production metadata. On a tuning window from ART-Chat-2.5M, evolutionary strategies guide the GORGO policy's parameters to directly optimize p95 TTFT. During held-out evaluation windows, we fix the parameter values learned from tuning and improve p95 TTFT by 6.9-15.5% and p95 end-to-end (E2E) latency by 14.3-30.9% over baseline load-balancing policies such as simple session affinity and prefix-cache. The code and ART-Chat-2.5M dataset can be found at https://github.com/Arcadia-Research-Team/GORGO.

Video Diffusion Transformer (DiT) models are a dominant approach for high-quality video generation but suffer from high inference cost due to iterative denoising. Existing caching approaches primarily exploit similarity within the diffusion process of a single request to skip redundant denoising steps. In this paper, we introduce Chorus, a caching approach that leverages similarity across requests to accelerate video diffusion model serving. Chorus achieves up to 45\% speedup on industrial 4-step distilled models, where prior intra-request caching approaches are ineffective. Particularly, Chorus employs a three-stage caching strategy along the denoising process. Stage 1 performs full reuse of latent features from similar requests. Stage 2 exploits inter-request caching in specific latent regions during intermediate denoising steps. This stage is combined with Token-Guided Attention Amplification to improve semantic alignment between the generated video and the conditional prompts, thereby extending the applicability of full reuse to later denoising steps.

For years, we have built LLM serving systems like any other critical infrastructure: a single general-purpose stack, hand-tuned over many engineer-years, meant to support every model and workload. In this paper, we take the opposite bet: a multi-agent loop that automatically synthesizes bespoke serving systems for different usage scenarios. We propose VibeServe, the first agentic loop that generates entire LLM serving stacks end-to-end. VibeServe uses an outer loop to plan and track the search over system designs, and an inner loop to implement candidates, check correctness, and measure performance on the target benchmark. In the standard deployment setting, where existing stacks are highly optimized, VibeServe remains competitive with vLLM, showing that generation-time specialization need not come at the cost of performance. More interestingly, in non-standard scenarios, VibeServe outperforms existing systems by exploiting opportunities that generic systems miss in six scenarios involving non-standard model architectures, workload knowledge, and hardware-specific optimizations. Together, these results suggest a different point in the design space for infrastructure software: generation-time specialization rather than runtime generality. Code is available at https://github.com/uw-syfi/vibe-serve.

Multi-agent LLM applications organize execution in synchronized rounds where a central scheduler gathers outputs from all agents and redistributes the combined context. This All-Gather communication pattern creates massive KV Cache redundancy, because every agent's prompt contains the same shared output blocks, yet existing reuse methods fail to exploit it efficiently. We present TokenDance, a system that scales the number of concurrent agents by exploiting the All-Gather pattern for collective KV Cache sharing. TokenDance's KV Collector performs KV Cache reuse over the full round in one collective step, so the cost of reusing a shared block is paid once regardless of agent count. Its Diff-Aware Storage encodes sibling caches as block-sparse diffs against a single master copy, achieving 11-17x compression on representative workloads. Evaluation on GenerativeAgents and AgentSociety shows that TokenDance supports up to 2.7x more concurrent agents than vLLM with prefix caching under SLO requirement, reduces per-agent KV Cache storage by up to 17.5x, and achieves up to 1.9x prefill speedup over per-request position-independent caching.

Large Language Model (LLM) inference in production must meet stringent service-level objectives for both time-to-first-token (TTFT) and time-between-token (TBT) while maximizing throughput under fixed compute, memory, and interconnect budgets. Modern serving systems adopt stall-free scheduling techniques such as chunked prefill, which splits long prompt processing along the token dimension and interleaves prefill with ongoing decode iterations. While effective at stabilizing TBT, chunked prefill incurs substantial overhead in Mixture-of-Experts (MoE) models: redundant expert weight loads increase memory traffic by up to 39% and inflate energy consumption. We propose layered prefill, a new scheduling paradigm that treats transformer layer groups as the primary scheduling unit. By vertically partitioning the model into contiguous layer groups and interleaving prefill and decode across the groups, layered prefill sustains stall-free decoding while eliminating chunk-induced MoE weight reloads. It reduces off-chip bandwidth demand, lowering TTFT by up to 70%, End-to-End latency by 41% and per-token energy by up to 22%. Evaluations show that layered prefill consistently improves the TTFT--TBT Pareto frontier over chunked prefill, reducing expert-load traffic and energy cost while maintaining stall-free decoding. Overall, shifting the scheduling axis from tokens to layers unlocks a new operating regime for high-efficiency, energy-aware LLM serving in co-located environments.

Deploying LLMs efficiently requires testing hundreds of serving configurations, but evaluating each one on a GPU cluster takes hours and costs thousands of dollars. Discrete-event simulators are faster and cheaper, but they require re-implementing the serving system's control logic -- a burden that compounds as frameworks evolve. We present Revati, a time-warp emulator that enables performance modeling by directly executing real serving system code at simulation-like speed. The system intercepts CUDA API calls to virtualize device management, allowing serving frameworks to run without physical GPUs. Instead of executing GPU kernels, it performs time jumps -- fast-forwarding virtual time by predicted kernel durations. We propose a coordination protocol that synchronizes these jumps across distributed processes while preserving causality. On vLLM and SGLang, Revati achieves less than 5% prediction error across multiple models and parallelism configurations, while running 5-17x faster than real GPU execution.

Multi-provider LLM gateways reverse proxies that route, load-balance, and rate-limit requests across foundation-model APIs have become critical production infrastructure. Yet the failure modes specific to this architectural layer remain undocumented, scattered across issue trackers and post-mortems with no unifying framework. We introduce , a two-axis taxonomy that classifies failures by their origin layer (Network/Transport, Streaming/Protocol, State/Session, Model~Behavior, Governance/Cost) and their detectability (Loud vs.\ Silent). We populate this taxonomy with five verified catalog entries sourced from public bug reports and first-hand stress testing, each accompanied by a mechanistic root-cause analysis. Three entries include standalone reproduction scripts. Our principal finding is that the most operationally severe failures are silent: they return HTTP~200, pass every standard health check, and corrupt application state in ways that require semantic-level observability to detect. Two such silent failures a concurrency race condition causing history loss and a streaming index collision corrupting tool-call payloads were discovered first-hand during evaluation campaigns.

Multimodal large language models (MLLMs) are typically trained in multiple stages, with video-based supervised fine-tuning (Video-SFT) serving as a key step for improving visual understanding. Yet its effect on the fine-grained evolution of visual capabilities, particularly the balance between spatial and temporal understanding, remains poorly understood. In this paper, we systematically study how Video-SFT reshapes visual capabilities in MLLMs. Across architectures, parameter scales, and frame sampling settings, we observe a consistent pattern: Video-SFT reliably improves video performance, but often yields limited gains or even degradation on static image benchmarks. We further show that this trade-off is closely tied to temporal budget: increasing the number of sampled frames generally improves video performance, but does not reliably improve static image performance. Motivated by this finding, we study an instruction-aware Hybrid-Frame strategy that adaptively allocates frame counts and partially mitigates the image-video trade-off. Our results indicate that Video-SFT is not a free lunch for MLLMs, and preserving spatial understanding remains a central challenge in joint image-video training.

Large language models (LLMs) often incorporate multiple text chunks in their inputs to provide the necessary contexts. To speed up the prefill of the long LLM inputs, one can pre-compute the KV cache of a text and re-use the KV cache when the context is reused as the prefix of another LLM input. However, the reused text chunks are not always the input prefix, and when they are not, their precomputed KV caches cannot be directly used since they ignore the text's cross-attention with the preceding text in the LLM input. Thus, the benefits of reusing KV caches remain largely unrealized. This paper tackles just one question: when an LLM input contains multiple text chunks, how to quickly combine their precomputed KV caches in order to achieve the same generation quality as the expensive full prefill (i.e., without reusing KV cache)? We present CacheBlend, a scheme that reuses the pre-computed KV caches, regardless prefix or not, and selectively recomputes the KV values of a small subset of tokens to partially update each reused KV cache. In the meantime,the small extra delay for recomputing some tokens can be pipelined with the retrieval of KV caches within the same job,allowing CacheBlend to store KV caches in slower devices with more storage capacity while retrieving them without increasing the inference delay. By comparing CacheBlend with the state-of-the-art KV cache reusing schemes on three open-source LLMs of various sizes and four popular benchmark datasets of different tasks, we show that CacheBlend reduces time-to-first-token (TTFT) by 2.2-3.3X and increases the inference throughput by 2.8-5X, compared with full KV recompute, without compromising generation quality or incurring more storage cost.

Mixture of Experts (MoE) LLMs, characterized by their sparse activation patterns, offer a promising approach to scaling language models while avoiding proportionally increasing the inference cost. However, their large parameter sizes present deployment challenges in resource-constrained environments with limited GPU memory capacity, as GPU memory is often insufficient to accommodate the full set of model weights. Consequently, typical deployments rely on CPU-GPU hybrid execution: the GPU handles compute-intensive GEMM operations, while the CPU processes the relatively lightweight attention mechanism. This setup introduces a key challenge: how to effectively optimize resource utilization across CPU and GPU? Prior work has designed system optimizations based on performance models with limited scope. Specifically, such models do not capture the complex interactions between hardware properties and system execution mechanisms. Therefore, previous approaches neither identify nor achieve the hardware limit. This paper presents MoE-Lens, a high-throughput MoE LLM inference system designed through holistic performance modeling for resource-constrained environments. Our performance model thoroughly analyzes various fundamental system components, including CPU memory capacity, GPU compute power, and workload characteristics, to understand the theoretical performance upper bound of MoE inference. Furthermore, it captures the system execution mechanisms to identify the key hardware bottlenecks and accurately predict the achievable throughput. Informed by our performance model, MoE-Lens introduces an inference system approaching hardware limits. Evaluated on diverse MoE models and datasets, MoE-Lens outperforms the state-of-the-art solution by 4.6x on average (up to 25.5x), with our theoretical model predicting performance with an average 94% accuracy.

Large language models (LLMs), while driving a new wave of interactive AI applications across numerous domains, suffer from high inference costs and heavy cloud dependency. Motivated by the redundancy phenomenon in linguistics, we propose a progressive inference paradigm over cloud and edge, i.e., firstly generating the sketch of the answer by LLMs at cloud, and then conducting parallel extension to fill in details by small models (SLMs) at edge. Progressive inference offers potential benefits to improve throughput and reduce inference latency while facing key implementation challenges, including decreased response quality from SLMs, a tradeoff between the brevity and comprehensiveness of sketches, as well as increased latency caused by network transmission and edge inference. In this work, we propose and implement PICE, an LLM serving system with semantic-level cloud-edge collaboration, enhancing inference throughput and quality through dynamic inference task scheduling, ensemble learning, and parallel edge inference. Extensive testbed experiments illustrate that our approach achieves 1.5-2times throughput enhancement and up to 43% latency reduction, while also potentially enhancing the quality compared to SOTA systems.

Efficiently serving Large Language Models (LLMs) requires selecting an optimal parallel execution plan, balancing computation, memory, and communication overhead. However, determining the best strategy is challenging due to varying parallelism techniques (data, pipeline, tensor) and workload characteristics (e.g., compute-intensive tasks with long prompts vs. memory-intensive tasks with long generation). We propose APEX, an LLM serving system simulator that efficiently identifies optimal parallel execution plans by considering key factors of LLM serving systems, such as memory usage, batching behavior, etc. APEX performs dynamism-aware simulation to model iteration-level batching, and leverages LLMs' repetitive structure to reduce design space, scaling efficiently to trillion-scale models. APEX abstracts the key components of LLM serving systems, including the model, batching module, quantization formats, and device clusters, enabling the simulator to be general and extensible. Simulating on a CPU, APEX evaluates execution plans for various device clusters, covering diverse LLMs and workloads. APEX finds plans up to 3.37x faster than heuristics, and also plans that reduce energy consumption by up to 45% compared to latency-optimal plans. APEX performs comprehensive evaluations, reporting key system metrics like time per output token and time to first token, which can help service providers meet SLOs. APEX identifies an optimal plan within 15 minutes on a CPU, making it 71x faster and 1234x more cost-effective than cloud-based GPU deployment. APEX can be accessed at https://github.com/microsoft/apex_plus

Serving Large Language Models (LLMs) is costly. However, post-training weight quantization can address this problem by both compressing their sizes for limited memory and saving bandwidth for acceleration. As not all weight dimensions are equally important, those methods typically rely on a sensitivity metric, which indicates the element-wise influence of weights on loss function and is used to preprocess original weights for better quantization. In this work, we conduct an empirical study on the accuracy of the sensitivity metric, and find that existing gradient and Hessian based metrics are very inaccurate: they underestimate quantization's impact on the loss function by orders of magnitude, mainly due to the small convergence radius of local 2nd order approximation, \ie, gradient and Hessian term in Taylor's formula. To tackle this problem, we propose Post-quantization Integral (PQI), an accurate metric to estimate posterior sensitivity in a fine-grained manner. To leverage this accurate metric, we further propose ReQuant, a simple yet powerful framework that mainly consists of two Dense-and-Sparse detach components: self-adaptive outlier selection and step-wise significant weights detach. Results show that ReQuant boosts state-of-the-art post-training quantization methods, with a pronounced improvement of 2.66 perplexity gain on Llama 3.2 1B with QTIP.

Serving large language models (LLMs) in production can incur substantial costs, which has prompted recent advances in inference system optimizations. Today, these systems are evaluated against conventional latency and throughput metrics (eg. TTFT, TBT, Normalised Latency and TPOT). However, these metrics fail to fully capture the nuances of LLM inference, leading to an incomplete assessment of user-facing performance crucial for real-time applications such as chat and translation. In this paper, we first identify the pitfalls of current performance metrics in evaluating LLM inference systems. We then propose Etalon, a comprehensive performance evaluation framework that includes fluidity-index -- a novel metric designed to reflect the intricacies of the LLM inference process and its impact on real-time user experience. Finally, we evaluate various existing open-source platforms and model-as-a-service offerings using Etalon, discussing their strengths and weaknesses. Etalon is available at https://github.com/project-etalon/etalon.

The capabilities and adoption of deep neural networks (DNNs) grow at an exhilarating pace: Vision models accurately classify human actions in videos and identify cancerous tissue in medical scans as precisely than human experts; large language models answer wide-ranging questions, generate code, and write prose, becoming the topic of everyday dinner-table conversations. Even though their uses are exhilarating, the continually increasing model sizes and computational complexities have a dark side. The economic cost and negative environmental externalities of training and serving models is in evident disharmony with financial viability and climate action goals. Instead of pursuing yet another increase in predictive performance, this dissertation is dedicated to the improvement of neural network efficiency. Specifically, a core contribution addresses the efficiency aspects during online inference. Here, the concept of Continual Inference Networks (CINs) is proposed and explored across four publications. CINs extend prior state-of-the-art methods developed for offline processing of spatio-temporal data and reuse their pre-trained weights, improving their online processing efficiency by an order of magnitude. These advances are attained through a bottom-up computational reorganization and judicious architectural modifications. The benefit to online inference is demonstrated by reformulating several widely used network architectures into CINs, including 3D CNNs, ST-GCNs, and Transformer Encoders. An orthogonal contribution tackles the concurrent adaptation and computational acceleration of a large source model into multiple lightweight derived models. Drawing on fusible adapter networks and structured pruning, Structured Pruning Adapters achieve superior predictive accuracy under aggressive pruning using significantly fewer learned weights compared to fine-tuning with pruning.

Efficiently serving large language models (LLMs) requires batching many requests together to reduce the cost per request. Yet, the key-value (KV) cache, which stores attention keys and values to avoid re-computations, significantly increases memory demands and becomes the new bottleneck in speed and memory usage. This memory demand increases with larger batch sizes and longer context lengths. Additionally, the inference speed is limited by the size of KV cache, as the GPU's SRAM must load the entire KV cache from the main GPU memory for each token generated, causing the computational core to be idle during this process. A straightforward and effective solution to reduce KV cache size is quantization, which decreases the total bytes taken by KV cache. However, there is a lack of in-depth studies that explore the element distribution of KV cache to understand the hardness and limitation of KV cache quantization. To fill the gap, we conducted a comprehensive study on the element distribution in KV cache of popular LLMs. Our findings indicate that the key cache should be quantized per-channel, i.e., group elements along the channel dimension and quantize them together. In contrast, the value cache should be quantized per-token. From this analysis, we developed a tuning-free 2bit KV cache quantization algorithm, named KIVI. With the hardware-friendly implementation, KIVI can enable Llama (Llama-2), Falcon, and Mistral models to maintain almost the same quality while using 2.6times less peak memory usage (including the model weight). This reduction in memory usage enables up to 4times larger batch size, bringing 2.35times sim 3.47times throughput on real LLM inference workload. The source code is available at https://github.com/jy-yuan/KIVI.

The high cost of ownership of AI compute infrastructure and challenges of robust serving of large language models (LLMs) has led to a surge in managed Model-as-a-service deployments. Even when enterprises choose on-premises deployments, the compute infrastructure is typically shared across many teams in order to maximize the return on investment. In both scenarios the deployed models operate only on plaintext data, and so enterprise data owners must allow their data to appear in plaintext on a shared or multi-tenant compute infrastructure. This results in data owners with private or sensitive data being hesitant or restricted in what data they use with these types of deployments. In this work we introduce the Stained Glass Transform, a learned, stochastic, and sequence dependent transformation of the word embeddings of an LLM which information theoretically provides privacy to the input of the LLM while preserving the utility of model. We theoretically connect a particular class of Stained Glass Transforms to the theory of mutual information of Gaussian Mixture Models. We then calculate a-postiori privacy estimates, based on mutual information, and verify the privacy and utility of instances of transformed embeddings through token level metrics of privacy and standard LLM performance benchmarks.

Self-attention is an essential component of large language models(LLMs) but a significant source of inference latency for long sequences. In multi-tenant LLMs serving scenarios, the compute and memory operation cost of self-attention can be optimized by using the probability that multiple LLM requests have shared system prompts in prefixes. In this paper, we introduce ChunkAttention, a prefix-aware self-attention module that can detect matching prompt prefixes across multiple requests and share their key/value tensors in memory at runtime to improve the memory utilization of KV cache. This is achieved by breaking monolithic key/value tensors into smaller chunks and structuring them into the auxiliary prefix tree. Consequently, on top of the prefix-tree based KV cache, we design an efficient self-attention kernel, where a two-phase partition algorithm is implemented to improve the data locality during self-attention computation in the presence of shared system prompts. Experiments show that ChunkAttention can speed up the self-attention kernel by 3.2-4.8times compared to the start-of-the-art implementation, with the length of the system prompt ranging from 1024 to 4096.

AI Agents can perform complex operations at great speed, but just like all the humans we have ever hired, their intelligence remains fallible. Miscommunications aren't noticed, systemic biases have no counter-action, and inner monologues are rarely written down. We did not come to fire them for their mistakes, but to hire them and provide a safe productive working environment. We posit that we can reuse a common corporate organizational structure: teams of independent AI agents with strict role boundaries can work with common goals, but opposing incentives. Multiple models serving as a team of rivals can catch and minimize errors within the final product at a small cost to the velocity of actions. In this paper we demonstrate that we can achieve reliability without acquiring perfect components, but through careful orchestration of imperfect ones. This paper describes the architecture of such a system in practice: specialized agent teams (planners, executors, critics, experts), organized into an organization with clear goals, coordinated through a remote code executor that keeps data transformations and tool invocations separate from reasoning models. Rather than agents directly calling tools and ingesting full responses, they write code that executes remotely; only relevant summaries return to agent context. By preventing raw data and tool outputs from contaminating context windows, the system maintains clean separation between perception (brains that plan and reason) and execution (hands that perform heavy data transformations and API calls). We demonstrate the approach achieves over 90% internal error interception prior to user exposure while maintaining acceptable latency tradeoffs. A survey from our traces shows that we only trade off cost and latency to achieve correctness and incrementally expand capabilities without impacting existing ones.

Coaching, which involves classroom observation and expert feedback, is a widespread and fundamental part of teacher training. However, the majority of teachers do not have access to consistent, high quality coaching due to limited resources and access to expertise. We explore whether generative AI could become a cost-effective complement to expert feedback by serving as an automated teacher coach. In doing so, we propose three teacher coaching tasks for generative AI: (A) scoring transcript segments based on classroom observation instruments, (B) identifying highlights and missed opportunities for good instructional strategies, and (C) providing actionable suggestions for eliciting more student reasoning. We recruit expert math teachers to evaluate the zero-shot performance of ChatGPT on each of these tasks for elementary math classroom transcripts. Our results reveal that ChatGPT generates responses that are relevant to improving instruction, but they are often not novel or insightful. For example, 82% of the model's suggestions point to places in the transcript where the teacher is already implementing that suggestion. Our work highlights the challenges of producing insightful, novel and truthful feedback for teachers while paving the way for future research to address these obstacles and improve the capacity of generative AI to coach teachers.

Realistic evaluation of LLM serving systems requires online workloads, dynamic arrivals, queueing, and the serving engine's local scheduling for execution batching, but running such experiments on GPUs is expensive. Existing simulators reduce this cost, but often operate offline or in time-warped mode, re-implement serving-engine schedulers, or require accurate operator/kernel-level latency models. We present LLM-Emu, a serving-native emulator for vLLM that preserves the production HTTP, scheduling, KV-cache, and output-processing paths while replacing only GPU forward execution with profile-sampled latency and synthetic output tokens. Tested on two different GPUs, four model variants, two model families, two attention backends, and both Poisson and bursty ShareGPT workloads, LLM-Emu closely tracks real vLLM serving behavior: TPOT and ITL stay within 4.8% absolute error, E2E latency within 5.3%, and output throughput within 1.9%; TTFT is less stable, with maximum error 10.4%, reflecting its sensitivity to admission and queue state. These results suggest that lightweight, serving-native emulation can support practical online experimentation for LLM-serving systems. LLM-Emu is open sourced at https://github.com/AKafakA/llm-emu.

Deploying a large language model (LLM) inference service remains costly because centralized serving depends on specialized GPU clusters and high-bandwidth interconnects in datacenters. An appealing alternative is to leverage collaborative decentralized GPU pools. However, heterogeneity in GPU and limited interconnected network bandwidth, along with potentially dynamic availability, make efficient scheduling the central challenge in this scenario. In this paper, we present Parallax, a decentralized LLM serving system that turns a pool of heterogeneous GPUs into an efficient inference platform via a two-phase scheduler. Parallax decomposes planning into (i) model allocation, which places layers of each replica across diverse GPUs to jointly optimize latency and throughput under memory and link-bandwidth constraints, and (ii) request-time GPU pipeline selection, which stitches layers from different replicas into end-to-end execution chains that balance load and adapt to current conditions. We implement Parallax and evaluate it on open-source LLMs deployed over real volunteer nodes. Parallax consistently reduces latency and increases throughput relative to decentralized baselines, demonstrating that principled scheduling can make volunteer compute a practical, affordable substrate for LLM inference. Github Repo at: https://github.com/GradientHQ/parallax.

Video streaming analytics is a crucial workload for vision-language model serving, but the high cost of multimodal inference limits scalability. Prior systems reduce inference cost by exploiting temporal and spatial redundancy in video streams, but they target either the vision transformer (ViT) or the LLM with a limited view, leaving end-to-end opportunities untapped. Moreover, existing methods incur significant overhead to identify redundancy, either through offline profiling and training or costly online computation, making them ill-suited for dynamic real-time streams. We present CodecSight, a codec-guided streaming video analytics system, built on a key observation that video codecs already extract the temporal and spatial structure of each stream as a byproduct of compression. CodecSight treats this codec metadata as a low-cost runtime signal to unify optimization across video decoding, visual processing, and LLM prefilling, with transmission reduction as an inherent benefit of operating directly on compressed bitstreams. This drives codec-guided patch pruning before ViT encoding and selective key-value cache refresh during LLM prefilling, both of which are fully online and do not require offline training. Experiments show that CodecSight achieves an improvement in throughput of up to 3times, and a reduction of up to 87% in GPU compute over state-of-the-art baselines, maintaining competitive accuracy with only 0sim8% F1 drop.

We show that verifier-free evolution is bottlenecked by both diversity and efficiency: without external correction, repeated evolution accelerates collapse toward narrow modes, while the uniform use of a high-cost model wastes compute and quickly becomes economically impractical. We introduce Squeeze Evolve, a unified multi-model orchestration framework for verifier-free evolutionary inference. Our approach is guided by a simple principle: allocate model capability where it has the highest marginal utility. Stronger models are reserved for high-impact stages, while cheaper models handle the other stages at much lower costs. This principle addresses diversity and cost-efficiency jointly while remaining lightweight. Squeeze Evolve naturally supports open-source, closed-source, and mixed-model deployments. Across AIME 2025, HMMT 2025, LiveCodeBench V6, GPQA-Diamond, ARC-AGI-V2, and multimodal vision benchmarks, such as MMMU-Pro and BabyVision, Squeeze Evolve consistently improves the cost-capability frontier over single-model evolution and achieves new state-of-the-art results on several tasks. Empirically, Squeeze Evolve reduces API cost by up to sim3times and increases fixed-budget serving throughput by up to sim10times. Moreover, on discovery tasks, Squeeze Evolve is the first verifier-free evolutionary method to match, and in some cases exceed, the performance of verifier-based evolutionary methods.

With the ubiquitous use of modern large language models (LLMs) across industries, the inference serving for these models is ever expanding. Given the high compute and memory requirements of modern LLMs, more and more top-of-the-line GPUs are being deployed to serve these models. Energy availability has come to the forefront as the biggest challenge for data center expansion to serve these models. In this paper, we present the trade-offs brought up by making energy efficiency the primary goal of LLM serving under performance SLOs. We show that depending on the inputs, the model, and the service-level agreements, there are several knobs available to the LLM inference provider to use for being energy efficient. We characterize the impact of these knobs on the latency, throughput, as well as the energy. By exploring these trade-offs, we offer valuable insights into optimizing energy usage without compromising on performance, thereby paving the way for sustainable and cost-effective LLM deployment in data center environments.

Large Language Models (LLMs) like ChatGPT are capable of successfully performing many language processing tasks zero-shot (without the need for training data). If this capacity also applies to the coding of social phenomena like persuasiveness and political ideology, then LLMs could effectively transform Computational Social Science (CSS). This work provides a road map for using LLMs as CSS tools. Towards this end, we contribute a set of prompting best practices and an extensive evaluation pipeline to measure the zero-shot performance of 13 language models on 24 representative CSS benchmarks. On taxonomic labeling tasks (classification), LLMs fail to outperform the best fine-tuned models but still achieve fair levels of agreement with humans. On free-form coding tasks (generation), LLMs produce explanations that often exceed the quality of crowdworkers' gold references. We conclude that today's LLMs can radically augment the CSS research pipeline in two ways: (1) serving as zero-shot data annotators on human annotation teams, and (2) bootstrapping challenging creative generation tasks (e.g., explaining the hidden meaning behind text). In summary, LLMs can significantly reduce costs and increase efficiency of social science analysis in partnership with humans.

Large language models have demonstrated promising capabilities upon scaling up parameters. However, serving large language models incurs substantial computation and memory movement costs due to their large scale. Quantization methods have been employed to reduce service costs and latency. Nevertheless, outliers in activations hinder the development of INT4 weight-activation quantization. Existing approaches separate outliers and normal values into two matrices or migrate outliers from activations to weights, suffering from high latency or accuracy degradation. Based on observing activations from large language models, outliers can be classified into channel-wise and spike outliers. In this work, we propose Rotated Runtime Smooth (RRS), a plug-and-play activation smoother for quantization, consisting of Runtime Smooth and the Rotation operation. Runtime Smooth (RS) is introduced to eliminate channel-wise outliers by smoothing activations with channel-wise maximums during runtime. The rotation operation can narrow the gap between spike outliers and normal values, alleviating the effect of victims caused by channel-wise smoothing. The proposed method outperforms the state-of-the-art method in the LLaMA and Qwen families and improves WikiText-2 perplexity from 57.33 to 6.66 for INT4 inference.

The evolution of recommender systems has shifted preference storage from rating matrices and dense embeddings to semantic memory in the agentic era. Yet existing agents rely on isolated memory, overlooking crucial collaborative signals. Bridging this gap is hindered by the dual challenges of distilling vast graph contexts without overwhelming reasoning agents with cognitive load, and evolving the collaborative memory efficiently without incurring prohibitive computational costs. To address this, we propose MemRec, a framework that architecturally decouples reasoning from memory management to enable efficient collaborative augmentation. MemRec introduces a dedicated, cost-effective LM_Mem to manage a dynamic collaborative memory graph, serving synthesized, high-signal context to a downstream LLM_Rec. The framework operates via a practical pipeline featuring efficient retrieval and cost-effective asynchronous graph propagation that evolves memory in the background. Extensive experiments on four benchmarks demonstrate that MemRec achieves state-of-the-art performance. Furthermore, architectural analysis confirms its flexibility, establishing a new Pareto frontier that balances reasoning quality, cost, and privacy through support for diverse deployments, including local open-source models. Code:https://github.com/rutgerswiselab/memrec and Homepage: https://memrec.weixinchen.com

Two widely adopted techniques for LLM inference serving systems today are hybrid batching and disaggregated serving. A hybrid batch combines prefill and decode tokens of different requests in the same batch to improve resource utilization and throughput at the cost of increased latency per token. In contrast, disaggregated serving decouples compute-bound prefill and bandwidth-bound decode phases to optimize for service level objectives (SLOs) at the cost of resource under-utilization and KV-cache transfer overheads. To address the limitations of these techniques, we propose RAPID-Serve: a technique to concurrently execute prefill and decode on the same GPU(s) to meet latency SLOs while maintaining high throughput and efficient resource utilization. Furthermore, we propose Adaptive Resource Management for runtime compute resource allocation, optionally leveraging CU masking (a fine-grained Compute Unit partitioning feature on AMD InstinctTM GPUs). RAPID-Serve provides up to 4.1x (average 1.7x) unconstrained throughput improvement and 32x and higher (average 4.9x) throughput improvement under SLO constraints, showing it as an effective strategy compared to the state-of-the-art approaches, particularly in resource-constrained environments.

Recently, as Large Language Models (LLMs) have fundamentally impacted various fields, the methods for incorporating up-to-date information into LLMs or adding external knowledge to construct domain-specific models have garnered wide attention. Retrieval-Augmented Generation (RAG), serving as an inference-time scaling method, is notable for its low cost and minimal effort for parameter tuning. However, due to heterogeneous training data and model architecture, the variant embedding models used in RAG exhibit different benefits across various areas, often leading to different similarity calculation results and, consequently, varying response quality from LLMs. To address this problem, we propose and examine two approaches to enhance RAG by combining the benefits of multiple embedding models, named Mixture-Embedding RAG and Confident RAG. Mixture-Embedding RAG simply sorts and selects retrievals from multiple embedding models based on standardized similarity; however, it does not outperform vanilla RAG. In contrast, Confident RAG generates responses multiple times using different embedding models and then selects the responses with the highest confidence level, demonstrating average improvements of approximately 10% and 5% over vanilla LLMs and RAG, respectively. The consistent results across different LLMs and embedding models indicate that Confident RAG is an efficient plug-and-play approach for various domains. We will release our code upon publication.

Transformers are mostly relying on softmax attention, which introduces quadratic complexity with respect to sequence length and remains a major bottleneck for efficient inference. Prior work on linear or hybrid attention typically replaces softmax attention uniformly across all layers, often leading to significant performance degradation or requiring extensive retraining to recover model quality. This work proposes LayerBoost, a layer-aware attention reduction method that selectively modifies the attention mechanism based on the sensitivity of individual transformer layers. It first performs a systematic sensitivity analysis on a pretrained model to identify layers that are critical for maintaining performance. Guided by this analysis, three distinct strategies can be applied: retaining standard softmax attention in highly sensitive layers, replacing it with linear sliding window attention in moderately sensitive layers, and removing attention entirely in layers that exhibit low sensitivity. To recover performance after these architectural modifications, we introduce a lightweight distillation-based healing phase requiring only 10M additional training tokens. LayerBoost reduces inference latency and improves throughput by up to 68% at high concurrency, while maintaining competitive model quality. It matches base model performance on several benchmarks, exhibits only minor degradations on others, and significantly outperforms state-of-the-art attention linearization methods. These efficiency gains make our method particularly well-suited for high-concurrency serving and hardware-constrained deployment scenarios, where inference cost and memory footprint are critical bottlenecks.

Large Language Models (LLMs) have achieved remarkable performance in Machine Translation (MT), but deploying them at scale remains prohibitively expensive. A widely adopted remedy is the hybrid system paradigm, which balances cost and quality by serving most requests with a small model and selectively routing a fraction to a large model. However, existing routing strategies often rely on heuristics, external predictors, or absolute quality estimation, which fail to capture whether the large model actually provides a worthwhile improvement over the small one. In this paper, we formulate routing as a budget allocation problem and identify marginal gain, i.e., the large model's improvement over the small model, as the optimal signal for budgeted decisions. Building on this, we propose RouteLMT (routing for LLM-based MT), an efficient in-model router that predicts this expected gain by probing the small translators prompt-token representation, without requiring external models or hypothesis decoding. Extensive experiments demonstrate that our RouteLMT outperforms heuristics, quality/difficulty estimation baselines, achieving a superior quality-budget Pareto frontier. Furthermore, we analyze regression risks and show that a simple guarded variant can mitigate severe quality losses.

Large Language Models (LLMs) have demonstrated remarkable abilities in tackling a wide range of complex tasks. However, their huge computational and memory costs raise significant challenges in deploying these models on resource-constrained devices or efficiently serving them. Prior approaches have attempted to alleviate these problems by permanently removing less important model structures, yet these methods often result in substantial performance degradation due to the permanent deletion of model parameters. In this work, we tried to mitigate this issue by reducing the number of active parameters without permanently removing them. Specifically, we introduce a differentiable dynamic pruning method that pushes dense models to maintain a fixed number of active parameters by converting their MLP layers into a Mixture of Experts (MoE) architecture. Our method, even without fine-tuning, consistently outperforms previous structural pruning techniques across diverse model families, including Phi-2, LLaMA-2, LLaMA-3, and Qwen-2.5.

Mixture-of-Expert (MoE) based large language models (LLMs), such as the recent Mixtral and DeepSeek-MoE, have shown great promise in scaling model size without suffering from the quadratic growth of training cost of dense transformers. Like dense models, training MoEs requires answering the same question: given a training budget, what is the optimal allocation on the model size and number of tokens? We study the scaling law of MoE-based LLMs regarding the relations between the model performance, model size, dataset size, and the expert degree. Echoing previous research studying MoE in different contexts, we observe the diminishing return of increasing the number of experts, but this seems to suggest we should scale the number of experts until saturation, as the training cost would remain constant, which is problematic during inference time. We propose to amend the scaling law of MoE by introducing inference efficiency as another metric besides the validation loss. We find that MoEs with a few (4/8) experts are the most serving efficient solution under the same performance, but costs 2.5-3.5x more in training. On the other hand, training a (16/32) expert MoE much smaller (70-85%) than the loss-optimal solution, but with a larger training dataset is a promising setup under a training budget.

As large-scale visual-document corpora such as arXiv papers and enterprise PDFs continue to grow, visual-document retrieval has gained increasing attention; yet it still lacks a deployable system that lexically indexes visual documents to serve queries without neural encoding at scale. Existing methods either achieve strong retrieval quality with VLM-based dense or multi-vector models but require neural query encoding at serving time, or avoid query encoding with OCR- or caption-based BM25 at the cost of time-consuming text extraction or generation. To fill this missing serving regime, we present V-SPLADE, an inference-free sparse retriever for visual-document retrieval. However, such inference-free multimodal learned sparse retrieval systems remain underexplored and have not yet shown dense-level effectiveness under high sparsity. We attribute this limitation to a lexical grounding problem: visual sparse representations often fail to capture the lexical content embedded in document images. To address this problem, we introduce caption-gated token supervision, a training-only signal that uses VLM-generated captions as lexical cues to activate retrieval-relevant vocabulary dimensions. With this supervision, V-SPLADE improves average NDCG@5 across six visual-document retrieval benchmarks by +13.8pp over the same-scale dense baseline and by up to +6.3pp over OCR- or caption-based BM25 baselines. On an 18.7M-document corpus, it more than doubles R@5 over the same-scale dense baseline and further improves competing retrievers through score fusion by up to +2.4pp R@5. Code will be released soon at https://github.com/naver/v-splade.

Speculative decoding is a pivotal technique to accelerate the inference of large language models (LLMs) by employing a smaller draft model to predict the target model's outputs. However, its efficacy can be limited due to the low predictive accuracy of the draft model, particularly when faced with diverse text inputs and a significant capability gap between the draft and target models. We introduce online speculative decoding (OSD) to address this challenge. The main idea is to continually update (multiple) draft model(s) on observed user query data using the abundant excess computational power in an LLM serving cluster. Given that LLM inference is memory-bounded, the surplus computational power in a typical LLM serving cluster can be repurposed for online retraining of draft models, thereby making the training cost-neutral. Since the query distribution of an LLM service is relatively simple, retraining on query distribution enables the draft model to more accurately predict the target model's outputs, particularly on data originating from query distributions. As the draft model evolves online, it aligns with the query distribution in real time, mitigating distribution shifts. We develop a prototype of online speculative decoding based on online knowledge distillation and evaluate it using both synthetic and real query data on several popular LLMs. The results show a substantial increase in the token acceptance rate by 0.1 to 0.65, which translates into 1.22x to 3.06x latency reduction.

Named Entity Recognition (NER) is a key component in industrial information extraction pipelines, where systems must satisfy strict latency and throughput constraints in addition to strong accuracy. State-of-the-art NER accuracy is often achieved by span-based frameworks, which construct span representations from token encodings and classify candidate spans. However, many span-based methods enumerate large numbers of candidates and process each candidate with marker-augmented inputs, substantially increasing inference cost and limiting scalability in large-scale deployments. In this work, we propose SpanDec, an efficient span-based NER framework that targets this bottleneck. Our main insight is that span representation interactions can be computed effectively at the final transformer stage, avoiding redundant computation in earlier layers via a lightweight decoder dedicated to span representations. We further introduce a span filtering mechanism during enumeration to prune unlikely candidates before expensive processing. Across multiple benchmarks, SpanDec matches competitive span-based baselines while improving throughput and reducing computational cost, yielding a better accuracy-efficiency trade-off suitable for high-volume serving and on-device applications.

We introduce "ivrit.ai", a comprehensive Hebrew speech dataset, addressing the distinct lack of extensive, high-quality resources for advancing Automated Speech Recognition (ASR) technology in Hebrew. With over 3,300 speech hours and a over a thousand diverse speakers, ivrit.ai offers a substantial compilation of Hebrew speech across various contexts. It is delivered in three forms to cater to varying research needs: raw unprocessed audio; data post-Voice Activity Detection, and partially transcribed data. The dataset stands out for its legal accessibility, permitting use at no cost, thereby serving as a crucial resource for researchers, developers, and commercial entities. ivrit.ai opens up numerous applications, offering vast potential to enhance AI capabilities in Hebrew. Future efforts aim to expand ivrit.ai further, thereby advancing Hebrew's standing in AI research and technology.

LLM agents often degrade over long episodes: as trajectories grow, they revisit explored states, repeat failed actions, and lose strategies that previously worked. Test-time training (TTT) offers a way to adapt model weights to the evolving task state, but existing LLM TTT methods largely adapt once to a fixed input. We study continuous TTT in multi-turn agent episodes, where each update changes the policy that generates later training text. This creates a self-training loop that helps when new trajectory information appears, but can amplify drift when the agent gets stuck and repeatedly trains on similar text. We find that update-text repetition distinguishes these regimes and introduce Agentic Test-Time Training (aTTT), a token-level reweighting method that downweights the loss on tokens appearing in repeated n-grams from prior updates while leaving novel tokens fully weighted. To run such updates inside live episodes, we build a concurrent serving system using vLLM's runtime LoRA API, limiting overhead to 1.9times the no-TTT cost. aTTT improves success by up to 5.0 points on ALFWorld and 4.9 points on SWE-bench Lite. The gains concentrate where models already have task competence but drift over long trajectories, suggesting that aTTT mainly preserves existing competence rather than teaching new abilities.

AI systems can now cheaply generate plausible scientific artifacts such as papers, reviews, and surveys. This creates a risk of epistemic pollution in our scientific systems, where unreliable but plausible-looking artifacts can accumulate faster than the system can filter them out. The problem is structural: the epistemic infrastructure of science was calibrated to a world where producing a plausible artifact required substantial expertise, labor, and time, so generation cost itself served as a rough filter; AI weakens that filter without comparably lowering verification cost. We argue that AI-era science should treat this as an engineering problem: redesigning epistemic infrastructure to rebalance the costs of generation and verification. The current paper-centered system makes verification expensive: papers compress long-context scientific logic into prose, forcing reviewers, human or AI, to reconstruct underlying argument structure before they can evaluate it. As one step in this direction, we propose blueprints as preliminary epistemic infrastructure: structured, decomposed research artifacts that represent claims, evidence, assumptions, and definitions as typed graph components. Blueprints are designed to trade an upfront generation cost for cheaper, more local, more distributed verification downstream. We have instantiated the proposal in a proof-of-concept prototype.

Transformers use the dense self-attention mechanism which gives a lot of flexibility for long-range connectivity. Over multiple layers of a deep transformer, the number of possible connectivity patterns increases exponentially. However, very few of these contribute to the performance of the network, and even fewer are essential. We hypothesize that there are sparsely connected sub-networks within a transformer, called information pathways which can be trained independently. However, the dynamic (i.e., input-dependent) nature of these pathways makes it difficult to prune dense self-attention during training. But the overall distribution of these pathways is often predictable. We take advantage of this fact to propose Stochastically Subsampled self-Attention (SSA) - a general-purpose training strategy for transformers that can reduce both the memory and computational cost of self-attention by 4 to 8 times during training while also serving as a regularization method - improving generalization over dense training. We show that an ensemble of sub-models can be formed from the subsampled pathways within a network, which can achieve better performance than its densely attended counterpart. We perform experiments on a variety of NLP, computer vision and graph learning tasks in both generative and discriminative settings to provide empirical evidence for our claims and show the effectiveness of the proposed method.

Standard fine-tuning of large pre-trained language models (PLMs) for downstream tasks requires updating hundreds of millions to billions of parameters, and storing a large copy of the PLM weights for every task resulting in increased cost for storing, sharing and serving the models. To address this, parameter-efficient fine-tuning (PEFT) techniques were introduced where small trainable components are injected in the PLM and updated during fine-tuning. We propose AdaMix as a general PEFT method that tunes a mixture of adaptation modules -- given the underlying PEFT method of choice -- introduced in each Transformer layer while keeping most of the PLM weights frozen. For instance, AdaMix can leverage a mixture of adapters like Houlsby or a mixture of low rank decomposition matrices like LoRA to improve downstream task performance over the corresponding PEFT methods for fully supervised and few-shot NLU and NLG tasks. Further, we design AdaMix such that it matches the same computational cost and the number of tunable parameters as the underlying PEFT method. By only tuning 0.1-0.2% of PLM parameters, we show that AdaMix outperforms SOTA parameter-efficient fine-tuning and full model fine-tuning for both NLU and NLG tasks.

Multimodal Large Language Models (MLLMs) have tremendous potential to improve the accuracy, availability, and cost-effectiveness of healthcare by providing automated solutions or serving as aids to medical professionals. Despite promising first steps in developing medical MLLMs in the past few years, their capabilities and limitations are not well-understood. Recently, many benchmark datasets have been proposed that test the general medical knowledge of such models across a variety of medical areas. However, the systematic failure modes and vulnerabilities of such models are severely underexplored with most medical benchmarks failing to expose the shortcomings of existing models in this safety-critical domain. In this paper, we introduce MediConfusion, a challenging medical Visual Question Answering (VQA) benchmark dataset, that probes the failure modes of medical MLLMs from a vision perspective. We reveal that state-of-the-art models are easily confused by image pairs that are otherwise visually dissimilar and clearly distinct for medical experts. Strikingly, all available models (open-source or proprietary) achieve performance below random guessing on MediConfusion, raising serious concerns about the reliability of existing medical MLLMs for healthcare deployment. We also extract common patterns of model failure that may help the design of a new generation of more trustworthy and reliable MLLMs in healthcare.