id
stringlengths 64
64
| published
stringlengths 19
25
| title
stringlengths 7
262
| description
stringlengths 6
54.4k
| link
stringlengths 31
227
| category
stringclasses 6
values | image
stringlengths 3
247
|
|---|---|---|---|---|---|---|
ddd458190a4299939b992115330bc571edff52bdd4c476ae6532067d2d9d873a
|
2026-01-21T00:00:00-05:00
|
Conversational Context Classification: A Representation Engineering Approach
|
arXiv:2601.12286v1 Announce Type: new Abstract: The increasing prevalence of Large Language Models (LLMs) demands effective safeguards for their operation, particularly concerning their tendency to generate out-of-context responses. A key challenge is accurately detecting when LLMs stray from expected conversational norms, manifesting as topic shifts, factual inaccuracies, or outright hallucinations. Traditional anomaly detection struggles to directly apply within contextual semantics. This paper outlines our experiment in exploring the use of Representation Engineering (RepE) and One-Class Support Vector Machine (OCSVM) to identify subspaces within the internal states of LLMs that represent a specific context. By training OCSVM on in-context examples, we establish a robust boundary within the LLM's hidden state latent space. We evaluate out study with two open source LLMs - Llama and Qwen models in specific contextual domain. Our approach entailed identifying the optimal layers within the LLM's internal state subspaces that strongly associates with the context of interest. Our evaluation results showed promising results in identifying the subspace for a specific context. Aside from being useful in detecting in or out of context conversation threads, this research work contributes to the study of better interpreting LLMs.
|
https://arxiv.org/abs/2601.12286
|
Academic Papers
|
svg
|
90cc193d62c6188227a736729044a4258c784361fe50bd74b0834f33d275b381
|
2026-01-21T00:00:00-05:00
|
TimeGMM: Single-Pass Probabilistic Forecasting via Adaptive Gaussian Mixture Models with Reversible Normalization
|
arXiv:2601.12288v1 Announce Type: new Abstract: Probabilistic time series forecasting is crucial for quantifying future uncertainty, with significant applications in fields such as energy and finance. However, existing methods often rely on computationally expensive sampling or restrictive parametric assumptions to characterize future distributions, which limits predictive performance and introduces distributional mismatch. To address these challenges, this paper presents TimeGMM, a novel probabilistic forecasting framework based on Gaussian Mixture Models (GMM) that captures complex future distributions in a single forward pass. A key component is GMM-adapted Reversible Instance Normalization (GRIN), a novel module designed to dynamically adapt to temporal-probabilistic distribution shifts. The framework integrates a dedicated Temporal Encoder (TE-Module) with a Conditional Temporal-Probabilistic Decoder (CTPD-Module) to jointly capture temporal dependencies and mixture distribution parameters. Extensive experiments demonstrate that TimeGMM consistently outperforms state-of-the-art methods, achieving maximum improvements of 22.48\% in CRPS and 21.23\% in NMAE.
|
https://arxiv.org/abs/2601.12288
|
Academic Papers
|
svg
|
855481160c9b67611f3af0d81864fafe764bfa8a1ecfc3192d57a2d194eb901b
|
2026-01-21T00:00:00-05:00
|
ParaMETA: Towards Learning Disentangled Paralinguistic Speaking Styles Representations from Speech
|
arXiv:2601.12289v1 Announce Type: new Abstract: Learning representative embeddings for different types of speaking styles, such as emotion, age, and gender, is critical for both recognition tasks (e.g., cognitive computing and human-computer interaction) and generative tasks (e.g., style-controllable speech generation). In this work, we introduce ParaMETA, a unified and flexible framework for learning and controlling speaking styles directly from speech. Unlike existing methods that rely on single-task models or cross-modal alignment, ParaMETA learns disentangled, task-specific embeddings by projecting speech into dedicated subspaces for each type of style. This design reduces inter-task interference, mitigates negative transfer, and allows a single model to handle multiple paralinguistic tasks such as emotion, gender, age, and language classification. Beyond recognition, ParaMETA enables fine-grained style control in Text-To-Speech (TTS) generative models. It supports both speech- and text-based prompting and allows users to modify one speaking styles while preserving others. Extensive experiments demonstrate that ParaMETA outperforms strong baselines in classification accuracy and generates more natural and expressive speech, while maintaining a lightweight and efficient model suitable for real-world applications.
|
https://arxiv.org/abs/2601.12289
|
Academic Papers
|
svg
|
9a0e25720980d414fcab93c09f796996d926e27359146596262f4959e08580fa
|
2026-01-21T00:00:00-05:00
|
Re-educating Educated Ones: A Case Study on Chakma Language Revitalization in Chittagong Hill Tracts
|
arXiv:2601.12290v1 Announce Type: new Abstract: Indigenous languages face significant cultural oppression from official state languages, particularly in the Global South. We investigate the Bangladeshi Chakma language revitalization movement, a community grappling with language liquidity and amalgamation into the dominant Bengali language. Our six-month-long qualitative study involving interviews and focus group discussions with Chakma language learning stakeholders uncovered existing community socio-economic challenges and resilience strategies. We noted the need for culturally grounded digital tools and resources. We propose an ICT-mediated community-centric framework for Indigenous language revitalization in the Global South, emphasizing the integration of historical identity elements, stakeholder-defined requirements, and effective digital engagement strategies to empower communities in preserving their linguistic and cultural heritage.
|
https://arxiv.org/abs/2601.12290
|
Academic Papers
|
svg
|
229d2c93e0bc2ac4543209ef3ac4dd02e42a381bd9d2094e774e77b4a8ec5983
|
2026-01-21T00:00:00-05:00
|
OpenNavMap: Structure-Free Topometric Mapping via Large-Scale Collaborative Localization
|
arXiv:2601.12291v1 Announce Type: new Abstract: Scalable and maintainable map representations are fundamental to enabling large-scale visual navigation and facilitating the deployment of robots in real-world environments. While collaborative localization across multi-session mapping enhances efficiency, traditional structure-based methods struggle with high maintenance costs and fail in feature-less environments or under significant viewpoint changes typical of crowd-sourced data. To address this, we propose OPENNAVMAP, a lightweight, structure-free topometric system leveraging 3D geometric foundation models for on-demand reconstruction. Our method unifies dynamic programming-based sequence matching, geometric verification, and confidence-calibrated optimization to robust, coarse-to-fine submap alignment without requiring pre-built 3D models. Evaluations on the Map-Free benchmark demonstrate superior accuracy over structure-from-motion and regression baselines, achieving an average translation error of 0.62m. Furthermore, the system maintains global consistency across 15km of multi-session data with an absolute trajectory error below 3m for map merging. Finally, we validate practical utility through 12 successful autonomous image-goal navigation tasks on simulated and physical robots. Code and datasets will be publicly available in https://rpl-cs-ucl.github.io/OpenNavMap_page.
|
https://arxiv.org/abs/2601.12291
|
Academic Papers
|
svg
|
76f44c152584e8d83290e903e1b64e327be1d12ae310a8706d8519b416bc1366
|
2026-01-21T00:00:00-05:00
|
ToolPRMBench: Evaluating and Advancing Process Reward Models for Tool-using Agents
|
arXiv:2601.12294v1 Announce Type: new Abstract: Reward-guided search methods have demonstrated strong potential in enhancing tool-using agents by effectively guiding sampling and exploration over complex action spaces. As a core design, those search methods utilize process reward models (PRMs) to provide step-level rewards, enabling more fine-grained monitoring. However, there is a lack of systematic and reliable evaluation benchmarks for PRMs in tool-using settings. In this paper, we introduce ToolPRMBench, a large-scale benchmark specifically designed to evaluate PRMs for tool-using agents. ToolPRMBench is built on top of several representative tool-using benchmarks and converts agent trajectories into step-level test cases. Each case contains the interaction history, a correct action, a plausible but incorrect alternative, and relevant tool metadata. We respectively utilize offline sampling to isolate local single-step errors and online sampling to capture realistic multi-step failures from full agent rollouts. A multi-LLM verification pipeline is proposed to reduce label noise and ensure data quality. We conduct extensive experiments across large language models, general PRMs, and tool-specialized PRMs on ToolPRMBench. The results reveal clear differences in PRM effectiveness and highlight the potential of specialized PRMs for tool-using. Code and data will be released at https://github.com/David-Li0406/ToolPRMBench.
|
https://arxiv.org/abs/2601.12294
|
Academic Papers
|
svg
|
10cd75b922fa901319faa10ee73d241eb49f52c04c6939e3edb86223c361da61
|
2026-01-21T00:00:00-05:00
|
Distribution Shift Is Key to Learning Invariant Prediction
|
arXiv:2601.12296v1 Announce Type: new Abstract: An interesting phenomenon arises: Empirical Risk Minimization (ERM) sometimes outperforms methods specifically designed for out-of-distribution tasks. This motivates an investigation into the reasons behind such behavior beyond algorithmic design. In this study, we find that one such reason lies in the distribution shift across training domains. A large degree of distribution shift can lead to better performance even under ERM. Specifically, we derive several theoretical and empirical findings demonstrating that distribution shift plays a crucial role in model learning and benefits learning invariant prediction. Firstly, the proposed upper bounds indicate that the degree of distribution shift directly affects the prediction ability of the learned models. If it is large, the models' ability can increase, approximating invariant prediction models that make stable predictions under arbitrary known or unseen domains; and vice versa. We also prove that, under certain data conditions, ERM solutions can achieve performance comparable to that of invariant prediction models. Secondly, the empirical validation results demonstrated that the predictions of learned models approximate those of Oracle or Optimal models, provided that the degree of distribution shift in the training data increases.
|
https://arxiv.org/abs/2601.12296
|
Academic Papers
|
svg
|
d1655f80b21f041d848eddfcdf49c7a32c799824935642bfc63c8ed6eea1b222
|
2026-01-21T00:00:00-05:00
|
CD-PIM: A High-Bandwidth and Compute-Efficient LPDDR5-Based PIM for Low-Batch LLM Acceleration on Edge-Device
|
arXiv:2601.12298v1 Announce Type: new Abstract: Edge deployment of low-batch large language models (LLMs) faces critical memory bandwidth bottlenecks when executing memory-intensive general matrix-vector multiplications (GEMV) operations. While digital processing-in-memory (PIM) architectures promise to accelerate GEMV operations, existing PIM-equipped edge devices still suffer from three key limitations: limited bandwidth improvement, component under-utilization in mixed workloads, and low compute capacity of computing units (CUs). In this paper, we propose CD-PIM to address these challenges through three key innovations. First, we introduce a high-bandwidth compute-efficient mode (HBCEM) that enhances bandwidth by dividing each bank into four pseudo-banks through segmented global bitlines. Second, we propose a low-batch interleaving mode (LBIM) to improve component utilization by overlapping GEMV operations with GEMM operations. Third, we design a compute-efficient CU that performs enhanced GEMV operations in a pipelined manner by serially feeding weight data into the computing core. Forth, we adopt a column-wise mapping for the key-cache matrix and row-wise mapping for the value-cache matrix, which fully utilizes CU resources. Our evaluation shows that compared to a GPU-only baseline and state-of-the-art PIM designs, our CD-PIM achieves 11.42x and 4.25x speedup on average within a single batch in HBCEM mode, respectively. Moreover, for low-batch sizes, the CD-PIM achieves an average speedup of 1.12x in LBIM compared to HBCEM.
|
https://arxiv.org/abs/2601.12298
|
Academic Papers
|
svg
|
9817e007a699a8cfff51083f3c899a878b6e39a57f28a249851edd944999b02e
|
2026-01-21T00:00:00-05:00
|
"What If My Face Gets Scanned Without Consent": Understanding Older Adults' Experiences with Biometric Payment
|
arXiv:2601.12300v1 Announce Type: new Abstract: Biometric payment, i.e., biometric authentication implemented in digital payment systems, can reduce memory demands and streamline payment for older adults. However, older adults' perceptions and practices regarding biometric payment remain underexplored. We conducted semi-structured interviews with 22 Chinese older adults, including both users and non-users. Participants were motivated to use biometric payment due to convenience and perceived security. However, they also worried about loss of control due to its password-free nature and expressed concerns about biometric data security. Participants also identified desired features for biometric payment, such as lightweight and context-aware cognitive confirmation mechanisms to enhance user control. Based on these findings, we outline recommendations for more controllable and informative digital financial services that better support older adults.
|
https://arxiv.org/abs/2601.12300
|
Academic Papers
|
svg
|
df634c1c3e24d9ac11b17783925809ec6c0b4915324cb1e22d06eabd53d8cdce
|
2026-01-21T00:00:00-05:00
|
Facet-Aware Multi-Head Mixture-of-Experts Model with Text-Enhanced Pre-training for Sequential Recommendation
|
arXiv:2601.12301v1 Announce Type: new Abstract: Sequential recommendation (SR) systems excel at capturing users' dynamic preferences by leveraging their interaction histories. Most existing SR systems assign a single embedding vector to each item to represent its features, adopting various models to combine these embeddings into a sequence representation that captures user intent. However, we argue that this representation alone is insufficient to capture an item's multi-faceted nature (e.g., movie genres, starring actors). Furthermore, users often exhibit complex and varied preferences within these facets (e.g., liking both action and musical films within the genre facet), which are challenging to fully represent with static identifiers. To address these issues, we propose a novel architecture titled Facet-Aware Multi-Head Mixture-of-Experts Model for Sequential Recommendation (FAME). We leverage sub-embeddings from each head in the final multi-head attention layer to predict the next item separately, effectively capturing distinct item facets. A gating mechanism then integrates these predictions by dynamically determining their importance. Additionally, we introduce a Mixture-of-Experts (MoE) network within each attention head to disentangle varied user preferences within each facet, utilizing a learnable router network to aggregate expert outputs based on context. Complementing this architecture, we design a Text-Enhanced Facet-Aware Pre-training module to overcome the limitations of randomly initialized embeddings. By utilizing a pre-trained text encoder and employing an alternating supervised contrastive learning objective, we explicitly disentangle facet-specific features from textual metadata (e.g., descriptions) before sequential training begins. This ensures that the item embeddings are semantically robust and aligned with the downstream multi-facet framework.
|
https://arxiv.org/abs/2601.12301
|
Academic Papers
|
svg
|
f14e8e553fece5da78824c55de68507b64b0764376201f122fd563e50900a751
|
2026-01-21T00:00:00-05:00
|
On the Minimum Length of Functional Batch Codes with Small Recovery Sets
|
arXiv:2601.12302v1 Announce Type: new Abstract: Batch codes are of potential use for load balancing and private information retrieval in distributed data storage systems. Recently, a special case of batch codes, termed functional batch codes, was proposed in the literature. In functional batch codes, users can query linear combinations of the information symbols, and not only the information symbols themselves, as is the case for standard batch codes. In this work, we consider linear functional batch codes with the additional property that every query is answered by using only a small number of coded symbols. We derive bounds on the minimum length of such codes, and evaluate the results by numerical computations.
|
https://arxiv.org/abs/2601.12302
|
Academic Papers
|
svg
|
3397b46b01a62cc5d43d0e9f5db561b0286dd1f01ca5b210becd44325cfb38ee
|
2026-01-21T00:00:00-05:00
|
Concepts from Representations: Post-hoc Concept Bottleneck Models via Sparse Decomposition of Visual Representations
|
arXiv:2601.12303v1 Announce Type: new Abstract: Deep learning has achieved remarkable success in image recognition, yet their inherent opacity poses challenges for deployment in critical domains. Concept-based interpretations aim to address this by explaining model reasoning through human-understandable concepts. However, existing post-hoc methods and ante-hoc concept bottleneck models (CBMs), suffer from limitations such as unreliable concept relevance, non-visual or labor-intensive concept definitions, and model or data-agnostic assumptions. This paper introduces Post-hoc Concept Bottleneck Model via Representation Decomposition (PCBM-ReD), a novel pipeline that retrofits interpretability onto pretrained opaque models. PCBM-ReD automatically extracts visual concepts from a pre-trained encoder, employs multimodal large language models (MLLMs) to label and filter concepts based on visual identifiability and task relevance, and selects an independent subset via reconstruction-guided optimization. Leveraging CLIP's visual-text alignment, it decomposes image representations into linear combination of concept embeddings to fit into the CBMs abstraction. Extensive experiments across 11 image classification tasks show PCBM-ReD achieves state-of-the-art accuracy, narrows the performance gap with end-to-end models, and exhibits better interpretability.
|
https://arxiv.org/abs/2601.12303
|
Academic Papers
|
svg
|
17dee31685b8c422564dcd5ec8b01ed94d0b3a11f857deaebf6231ab8910be21
|
2026-01-21T00:00:00-05:00
|
A Two-Stage Globally-Diverse Adversarial Attack for Vision-Language Pre-training Models
|
arXiv:2601.12304v1 Announce Type: new Abstract: Vision-language pre-training (VLP) models are vulnerable to adversarial examples, particularly in black-box scenarios. Existing multimodal attacks often suffer from limited perturbation diversity and unstable multi-stage pipelines. To address these challenges, we propose 2S-GDA, a two-stage globally-diverse attack framework. The proposed method first introduces textual perturbations through a globally-diverse strategy by combining candidate text expansion with globally-aware replacement. To enhance visual diversity, image-level perturbations are generated using multi-scale resizing and block-shuffle rotation. Extensive experiments on VLP models demonstrate that 2S-GDA consistently improves attack success rates over state-of-the-art methods, with gains of up to 11.17\% in black-box settings. Our framework is modular and can be easily combined with existing methods to further enhance adversarial transferability.
|
https://arxiv.org/abs/2601.12304
|
Academic Papers
|
svg
|
cbbf58cfb53dd9b2bfbd6a402759c84a5d34cd6c192885a82bf6cfa541066fd9
|
2026-01-21T00:00:00-05:00
|
Machine Learning as a Service (MLaaS) Dataset Generator Framework for IoT Environments
|
arXiv:2601.12305v1 Announce Type: new Abstract: We propose a novel MLaaS Dataset Generator (MDG) framework that creates configurable and reproducible datasets for evaluating Machine Learning as a Service (MLaaS) selection and composition. MDG simulates realistic MLaaS behaviour by training and evaluating diverse model families across multiple real-world datasets and data distribution settings. It records detailed functional attributes, quality of service metrics, and composition-specific indicators, enabling systematic analysis of service performance and cross-service behaviour. Using MDG, we generate more than ten thousand MLaaS service instances and construct a large-scale benchmark dataset suitable for downstream evaluation. We also implement a built-in composition mechanism that models how services interact under varied Internet of Things conditions. Experiments demonstrate that datasets generated by MDG enhance selection accuracy and composition quality compared to existing baselines. MDG provides a practical and extensible foundation for advancing data-driven research on MLaaS selection and composition
|
https://arxiv.org/abs/2601.12305
|
Academic Papers
|
svg
|
f5daac0ade5df2da8119234235e4780ac2b3cb1b6e8471d40befbf324c216bd2
|
2026-01-21T00:00:00-05:00
|
Rethinking the Value of Multi-Agent Workflow: A Strong Single Agent Baseline
|
arXiv:2601.12307v1 Announce Type: new Abstract: Recent advances in LLM-based multi-agent systems (MAS) show that workflows composed of multiple LLM agents with distinct roles, tools, and communication patterns can outperform single-LLM baselines on complex tasks. However, most frameworks are homogeneous, where all agents share the same base LLM and differ only in prompts, tools, and positions in the workflow. This raises the question of whether such workflows can be simulated by a single agent through multi-turn conversations. We investigate this across seven benchmarks spanning coding, mathematics, general question answering, domain-specific reasoning, and real-world planning and tool use. Our results show that a single agent can reach the performance of homogeneous workflows with an efficiency advantage from KV cache reuse, and can even match the performance of an automatically optimized heterogeneous workflow. Building on this finding, we propose \textbf{OneFlow}, an algorithm that automatically tailors workflows for single-agent execution, reducing inference costs compared to existing automatic multi-agent design frameworks without trading off accuracy. These results position the single-LLM implementation of multi-agent workflows as a strong baseline for MAS research. We also note that single-LLM methods cannot capture heterogeneous workflows due to the lack of KV cache sharing across different LLMs, highlighting future opportunities in developing \textit{truly} heterogeneous multi-agent systems.
|
https://arxiv.org/abs/2601.12307
|
Academic Papers
|
svg
|
a9142fb4b920d57b03c88e79f2413ad3bf6d928533d14b97da6b5ead01c9d0b7
|
2026-01-21T00:00:00-05:00
|
Adaptive Multi-Scale Correlation Meta-Network for Few-Shot Remote Sensing Image Classification
|
arXiv:2601.12308v1 Announce Type: new Abstract: Few-shot learning in remote sensing remains challenging due to three factors: the scarcity of labeled data, substantial domain shifts, and the multi-scale nature of geospatial objects. To address these issues, we introduce Adaptive Multi-Scale Correlation Meta-Network (AMC-MetaNet), a lightweight yet powerful framework with three key innovations: (i) correlation-guided feature pyramids for capturing scale-invariant patterns, (ii) an adaptive channel correlation module (ACCM) for learning dynamic cross-scale relationships, and (iii) correlation-guided meta-learning that leverages correlation patterns instead of conventional prototype averaging. Unlike prior approaches that rely on heavy pre-trained models or transformers, AMC-MetaNet is trained from scratch with only $\sim600K$ parameters, offering $20\times$ fewer parameters than ResNet-18 while maintaining high efficiency ($<50$ms per image inference). AMC-MetaNet achieves up to 86.65\% accuracy in 5-way 5-shot classification on various remote sensing datasets, including EuroSAT, NWPU-RESISC45, UC Merced Land Use, and AID. Our results establish AMC-MetaNet as a computationally efficient, scale-aware framework for real-world few-shot remote sensing.
|
https://arxiv.org/abs/2601.12308
|
Academic Papers
|
svg
|
5fe47c09a4dd3dffe9edb305b08a175acd4b3ff90b5938ecfc6c2b09d34af9b6
|
2026-01-21T00:00:00-05:00
|
Survival is the Only Reward: Sustainable Self-Training Through Environment-Mediated Selection
|
arXiv:2601.12310v1 Announce Type: new Abstract: Self-training systems often degenerate due to the lack of an external criterion for judging data quality, leading to reward hacking and semantic drift. This paper provides a proof-of-concept system architecture for stable self-training under sparse external feedback and bounded memory, and empirically characterises its learning dynamics and failure modes. We introduce a self-training architecture in which learning is mediated exclusively by environmental viability, rather than by reward, objective functions, or externally defined fitness criteria. Candidate behaviours are executed under real resource constraints, and only those whose environmental effects both persist and preserve the possibility of future interaction are propagated. The environment does not provide semantic feedback, dense rewards, or task-specific supervision; selection operates solely through differential survival of behaviours as world-altering events, making proxy optimisation impossible and rendering reward-hacking evolutionarily unstable. Analysis of semantic dynamics shows that improvement arises primarily through the persistence of effective and repeatable strategies under a regime of consolidation and pruning, a paradigm we refer to as negative-space learning (NSL), and that models develop meta-learning strategies (such as deliberate experimental failure in order to elicit informative error messages) without explicit instruction. This work establishes that environment-grounded selection enables sustainable open-ended self-improvement, offering a viable path toward more robust and generalisable autonomous systems without reliance on human-curated data or complex reward shaping.
|
https://arxiv.org/abs/2601.12310
|
Academic Papers
|
svg
|
31422991c48a80420b2e89c7af6ea4e7bbaa0cb43c3c564bd50a30ae98cd273d
|
2026-01-21T00:00:00-05:00
|
Cross-reality Location Privacy Protection in 6G-enabled Vehicular Metaverses: An LLM-enhanced Hybrid Generative Diffusion Model-based Approach
|
arXiv:2601.12311v1 Announce Type: new Abstract: The emergence of 6G-enabled vehicular metaverses enables Autonomous Vehicles (AVs) to operate across physical and virtual spaces through space-air-ground-sea integrated networks. The AVs can deploy AI agents powered by large AI models as personalized assistants, on edge servers to support intelligent driving decision making and enhanced on-board experiences. However, such cross-reality interactions may cause serious location privacy risks, as adversaries can infer AV trajectories by correlating the location reported when AVs request LBS in reality with the location of the edge servers on which their corresponding AI agents are deployed in virtuality. To address this challenge, we design a cross-reality location privacy protection framework based on hybrid actions, including continuous location perturbation in reality and discrete privacy-aware AI agent migration in virtuality. In this framework, a new privacy metric, termed cross-reality location entropy, is proposed to effectively quantify the privacy levels of AVs. Based on this metric, we formulate an optimization problem to optimize the hybrid action, focusing on achieving a balance between location protection, service latency reduction, and quality of service maintenance. To solve the complex mixed-integer problem, we develop a novel LLM-enhanced Hybrid Diffusion Proximal Policy Optimization (LHDPPO) algorithm, which integrates LLM-driven informative reward design to enhance environment understanding with double Generative Diffusion Models-based policy exploration to handle high-dimensional action spaces, thereby enabling reliable determination of optimal hybrid actions. Extensive experiments on real-world datasets demonstrate that the proposed framework effectively mitigates cross-reality location privacy leakage for AVs while maintaining strong user immersion within 6G-enabled vehicular metaverse scenarios.
|
https://arxiv.org/abs/2601.12311
|
Academic Papers
|
svg
|
060986af4afeaf099db53850feff7dd42d486b8e1477e6f22bfd620b11bb53a8
|
2026-01-21T00:00:00-05:00
|
CurConMix+: A Unified Spatio-Temporal Framework for Hierarchical Surgical Workflow Understanding
|
arXiv:2601.12312v1 Announce Type: new Abstract: Surgical action triplet recognition aims to understand fine-grained surgical behaviors by modeling the interactions among instruments, actions, and anatomical targets. Despite its clinical importance for workflow analysis and skill assessment, progress has been hindered by severe class imbalance, subtle visual variations, and the semantic interdependence among triplet components. Existing approaches often address only a subset of these challenges rather than tackling them jointly, which limits their ability to form a holistic understanding. This study builds upon CurConMix, a spatial representation framework. At its core, a curriculum-guided contrastive learning strategy learns discriminative and progressively correlated features, further enhanced by structured hard-pair sampling and feature-level mixup. Its temporal extension, CurConMix+, integrates a Multi-Resolution Temporal Transformer (MRTT) that achieves robust, context-aware understanding by adaptively fusing multi-scale temporal features and dynamically balancing spatio-temporal cues. Furthermore, we introduce LLS48, a new, hierarchically annotated benchmark for complex laparoscopic left lateral sectionectomy, providing step-, task-, and action-level annotations. Extensive experiments on CholecT45 and LLS48 demonstrate that CurConMix+ not only outperforms state-of-the-art approaches in triplet recognition, but also exhibits strong cross-level generalization, as its fine-grained features effectively transfer to higher-level phase and step recognition tasks. Together, the framework and dataset provide a unified foundation for hierarchy-aware, reproducible, and interpretable surgical workflow understanding. The code and dataset will be publicly released on GitHub to facilitate reproducibility and further research.
|
https://arxiv.org/abs/2601.12312
|
Academic Papers
|
svg
|
dbea6c8d39fcd06c1d5b9003b0944d441001336f6e7a94fe25f70f0033d02322
|
2026-01-21T00:00:00-05:00
|
S^2F-Net:A Robust Spatial-Spectral Fusion Framework for Cross-Model AIGC Detection
|
arXiv:2601.12313v1 Announce Type: new Abstract: The rapid development of generative models has imposed an urgent demand for detection schemes with strong generalization capabilities. However, existing detection methods generally suffer from overfitting to specific source models, leading to significant performance degradation when confronted with unseen generative architectures. To address these challenges, this paper proposes a cross-model detection framework called S 2 F-Net, whose core lies in exploring and leveraging the inherent spectral discrepancies between real and synthetic textures. Considering that upsampling operations leave unique and distinguishable frequency fingerprints in both texture-poor and texture-rich regions, we focus our research on the detection of frequency-domain artifacts, aiming to fundamentally improve the generalization performance of the model. Specifically, we introduce a learnable frequency attention module that adaptively weights and enhances discriminative frequency bands by synergizing spatial texture analysis and spectral dependencies.On the AIGCDetectBenchmark, which includes 17 categories of generative models, S 2 F-Net achieves a detection accuracy of 90.49%, significantly outperforming various existing baseline methods in cross-domain detection scenarios.
|
https://arxiv.org/abs/2601.12313
|
Academic Papers
|
svg
|
ab1b4c3f5506396894b5357ea8dfb36d70d8dbe7e64c8e87ebbb5030a7757811
|
2026-01-21T00:00:00-05:00
|
A Similarity Network for Correlating Musical Structure to Military Strategy
|
arXiv:2601.12314v1 Announce Type: new Abstract: Music perception, a multi-sensory process based on the synesthesia effect, is an essential component of music aesthetic education. Understanding music structure helps both perception and aesthetic education. Music structure incorporates a range of information, the coordination of which forms the melody, just as different military actions cooperate to produce a military strategy. However, there are a few ways for assessing music perception from the perspectives of system operation and information management. In this paper, we explore the similarities between music structure and military strategy while creating the Music Clips Correlation Network (MCCN) based on Mel-frequency Cepstral Coefficients (MFCCs). The inspiration comes from the comparison between a concert conductor's musical score and a military war commander's sand table exercise. Specifically, we create MCCNs for various kinds of war movie soundtracks, then relate military tactics (Sun Tzu's Art of War, etc.) and political institutions to military operations networks. Our primary findings suggest a few similarities, implying that music perception and aesthetic education can be approached from a military strategy and management perspective through this interdisciplinary research. Similarly, we can discover similarities between the art of military scheming and the art of musical structure based on network analysis in order to facilitate the understanding of the relationship between technology and art.
|
https://arxiv.org/abs/2601.12314
|
Academic Papers
|
svg
|
6a317eccfd4ba9ec924d36f77e5f8e0032cf89b5265a94dbf2b133cd957f22a3
|
2026-01-21T00:00:00-05:00
|
GazeFormer-MoE: Context-Aware Gaze Estimation via CLIP and MoE Transformer
|
arXiv:2601.12316v1 Announce Type: new Abstract: We present a semantics modulated, multi scale Transformer for 3D gaze estimation. Our model conditions CLIP global features with learnable prototype banks (illumination, head pose, background, direction), fuses these prototype-enriched global vectors with CLIP patch tokens and high-resolution CNN tokens in a unified attention space, and replaces several FFN blocks with routed/shared Mixture of Experts to increase conditional capacity. Evaluated on MPIIFaceGaze, EYEDIAP, Gaze360 and ETH-XGaze, our model achieves new state of the art angular errors of 2.49{\deg}, 3.22{\deg}, 10.16{\deg}, and 1.44{\deg}, demonstrating up to a 64% relative improvement over previously reported results. ablations attribute gains to prototype conditioning, cross scale fusion, MoE and hyperparameter. Our code is publicly available at https://github. com/AIPMLab/Gazeformer.
|
https://arxiv.org/abs/2601.12316
|
Academic Papers
|
svg
|
7a6b4d2a3736f523ded9b51f4dec6ec3fdd3a04d5480ccc38043a1ae9edb48a0
|
2026-01-21T00:00:00-05:00
|
Explanova: Automatically Discover Data Insights in N \times M Table via XAI Combined LLM Workflow
|
arXiv:2601.12317v1 Announce Type: new Abstract: Automation in data analysis has been a long-time pursuit. Current agentic LLM shows a promising solution towards it. Like DeepAnalyze, DataSage, and Datawise. They are all powerful agentic frameworks for automatic fine-grained analysis and are powered by LLM-based agentic tool calling ability. However, what about powered by a preset AutoML-like workflow? If we traverse all possible exploration, like Xn itself`s statistics, Xn1-Xn2 relationships, Xn to all other, and finally explain? Our Explanova is such an attempt: Cheaper due to a Local Small LLM.
|
https://arxiv.org/abs/2601.12317
|
Academic Papers
|
svg
|
15ba463a770a267bad56998b79029577f9dd1c0b0922de2fa794dcf07240e310
|
2026-01-21T00:00:00-05:00
|
Beyond Human Annotation: Recent Advances in Data Generation Methods for Document Intelligence
|
arXiv:2601.12318v1 Announce Type: new Abstract: The advancement of Document Intelligence (DI) demands large-scale, high-quality training data, yet manual annotation remains a critical bottleneck. While data generation methods are evolving rapidly, existing surveys are constrained by fragmented focuses on single modalities or specific tasks, lacking a unified perspective aligned with real-world workflows. To fill this gap, this survey establishes the first comprehensive technical map for data generation in DI. Data generation is redefined as supervisory signal production, and a novel taxonomy is introduced based on the "availability of data and labels." This framework organizes methodologies into four resource-centric paradigms: Data Augmentation, Data Generation from Scratch, Automated Data Annotation, and Self-Supervised Signal Construction. Furthermore, a multi-level evaluation framework is established to integrate intrinsic quality and extrinsic utility, compiling performance gains across diverse DI benchmarks. Guided by this unified structure, the methodological landscape is dissected to reveal critical challenges such as fidelity gaps and frontiers including co-evolutionary ecosystems. Ultimately, by systematizing this fragmented field, data generation is positioned as the central engine for next-generation DI.
|
https://arxiv.org/abs/2601.12318
|
Academic Papers
|
svg
|
d163393be97e84163eb858ea5a12f4aa6c28657b821400481da0d2ecfad433f7
|
2026-01-21T00:00:00-05:00
|
Ordered Local Momentum for Asynchronous Distributed Learning under Arbitrary Delays
|
arXiv:2601.12322v1 Announce Type: new Abstract: Momentum SGD (MSGD) serves as a foundational optimizer in training deep models due to momentum's key role in accelerating convergence and enhancing generalization. Meanwhile, asynchronous distributed learning is crucial for training large-scale deep models, especially when the computing capabilities of the workers in the cluster are heterogeneous. To reduce communication frequency, local updates are widely adopted in distributed learning. However, how to implement asynchronous distributed MSGD with local updates remains unexplored. To solve this problem, we propose a novel method, called \underline{or}dered \underline{lo}cal \underline{mo}mentum (OrLoMo), for asynchronous distributed learning. In OrLoMo, each worker runs MSGD locally. Then the local momentum from each worker will be aggregated by the server in order based on its global iteration index. To the best of our knowledge, OrLoMo is the first method to implement asynchronous distributed MSGD with local updates. We prove the convergence of OrLoMo for non-convex problems under arbitrary delays. Experiments validate that OrLoMo can outperform its synchronous counterpart and other asynchronous methods.
|
https://arxiv.org/abs/2601.12322
|
Academic Papers
|
svg
|
bdec65bae73c3577c06950e0b55e880b73b71f3b5a5fb26a6cbe6684023e9350
|
2026-01-21T00:00:00-05:00
|
MARO: Learning Stronger Reasoning from Social Interaction
|
arXiv:2601.12323v1 Announce Type: new Abstract: Humans face countless scenarios that require reasoning and judgment in daily life. However, existing large language model training methods primarily allow models to learn from existing textual content or solve predetermined problems, lacking experience in real scenarios involving interaction, negotiation, and competition with others. To address this, this paper proposes Multi-Agent Reward Optimization (MARO), a method that enables large language models (LLMs) to acquire stronger reasoning abilities by learning and practicing in multi-agent social environments. Specifically, MARO first addresses the sparse learning signal problem by decomposing final success or failure outcomes into each specific behavior during the interaction process; second, it handles the uneven role distribution problem by balancing the training sample weights of different roles; finally, it addresses environmental instability issues by directly evaluating the utility of each behavior. Experimental results demonstrate that MARO not only achieves significant improvements in social reasoning capabilities, but also that the abilities acquired through social simulation learning can effectively transfer to other tasks such as mathematical reasoning and instruction following. This reveals the tremendous potential of multi-agent social learning in enhancing the general reasoning capabilities of LLMs.
|
https://arxiv.org/abs/2601.12323
|
Academic Papers
|
svg
|
7816551508f3d72f0fe7443b985bae0eacbd9a68c0b67dd5d52b91a7450e76e3
|
2026-01-21T00:00:00-05:00
|
Experiencer, Helper, or Observer: Online Fraud Intervention for Older Adults Through Role-based Simulation
|
arXiv:2601.12324v1 Announce Type: new Abstract: Online fraud is a critical global threat that disproportionately targets older adults. Prior anti-fraud education for older adults has largely relied on static, traditional instruction that limits engagement and real-world transfer, whereas role-based simulation offers realistic yet low-risk opportunities for practice. Moreover, most interventions situate learners as victims, overlooking that fraud encounters often involve multiple roles, such as bystanders who witness scams and helpers who support victims. To address this gap, we developed ROLESafe, an anti-fraud educational intervention in which older adults learn through different learning roles, including Experiencer (experiencing fraud), Helper (assisting a victim), and Observer (witnessing fraud). In a between-subjects study with 144 older adults in China, we found that the Experiencer and Helper roles significantly improved participants' ability to identify online fraud. These findings highlight the promise of role-based, multi-perspective simulations for enhancing fraud awareness among older adults and provide design implications for future anti-fraud education.
|
https://arxiv.org/abs/2601.12324
|
Academic Papers
|
svg
|
754303222375692c271eb2fc9913544ce310ac2af619cd1cac0e7fa5062b8cfd
|
2026-01-21T00:00:00-05:00
|
Multi-Sensor Matching with HyperNetworks
|
arXiv:2601.12325v1 Announce Type: new Abstract: Hypernetworks are models that generate or modulate the weights of another network. They provide a flexible mechanism for injecting context and task conditioning and have proven broadly useful across diverse applications without significant increases in model size. We leverage hypernetworks to improve multimodal patch matching by introducing a lightweight descriptor-learning architecture that augments a Siamese CNN with (i) hypernetwork modules that compute adaptive, per-channel scaling and shifting and (ii) conditional instance normalization that provides modality-specific adaptation (e.g., visible vs. infrared, VIS-IR) in shallow layers. This combination preserves the efficiency of descriptor-based methods during inference while increasing robustness to appearance shifts. Trained with a triplet loss and hard-negative mining, our approach achieves state-of-the-art results on VIS-NIR and other VIS-IR benchmarks and matches or surpasses prior methods on additional datasets, despite their higher inference cost. To spur progress on domain shift, we also release GAP-VIR, a cross-platform (ground/aerial) VIS-IR patch dataset with 500K pairs, enabling rigorous evaluation of cross-domain generalization and adaptation.
|
https://arxiv.org/abs/2601.12325
|
Academic Papers
|
svg
|
aa137892adb5d31270bf42f87e85a43aa750e7c27c334a73bdf7395e1abd9ae6
|
2026-01-21T00:00:00-05:00
|
EmoKGEdit: Training-free Affective Injection via Visual Cue Transformation
|
arXiv:2601.12326v1 Announce Type: new Abstract: Existing image emotion editing methods struggle to disentangle emotional cues from latent content representations, often yielding weak emotional expression and distorted visual structures. To bridge this gap, we propose EmoKGEdit, a novel training-free framework for precise and structure-preserving image emotion editing. Specifically, we construct a Multimodal Sentiment Association Knowledge Graph (MSA-KG) to disentangle the intricate relationships among objects, scenes, attributes, visual clues and emotion. MSA-KG explicitly encode the causal chain among object-attribute-emotion, and as external knowledge to support chain of thought reasoning, guiding the multimodal large model to infer plausible emotion-related visual cues and generate coherent instructions. In addition, based on MSA-KG, we design a disentangled structure-emotion editing module that explicitly separates emotional attributes from layout features within the latent space, which ensures that the target emotion is effectively injected while strictly maintaining visual spatial coherence. Extensive experiments demonstrate that EmoKGEdit achieves excellent performance in both emotion fidelity and content preservation, and outperforms the state-of-the-art methods.
|
https://arxiv.org/abs/2601.12326
|
Academic Papers
|
svg
|
3503453e71d479e1ee5758e9fd7a2bb1b23d25ef373e412d9a7f167b35aad1d3
|
2026-01-21T00:00:00-05:00
|
The Expert Validation Framework (EVF): Enabling Domain Expert Control in AI Engineering
|
arXiv:2601.12327v1 Announce Type: new Abstract: Generative AI (GenAI) systems promise to transform knowledge work by automating a range of tasks, yet their deployment in enterprise settings remains hindered by the lack of systematic quality assurance mechanisms. We present an Expert Validation Framework that places domain experts at the center of building software with GenAI components, enabling them to maintain authoritative control over system behavior through structured specification, testing, validation, and continuous monitoring processes. Our framework addresses the critical gap between AI capabilities and organizational trust by establishing a rigorous, expert-driven methodology for ensuring quality across diverse GenAI applications. Through a four-stage implementation process encompassing specification, system creation, validation, and production monitoring, the framework enables organizations to leverage GenAI capabilities while maintaining expert oversight and quality standards.
|
https://arxiv.org/abs/2601.12327
|
Academic Papers
|
svg
|
dedad7a9e7ab8f0b56da8682d06676132697d9cd1250c46e6719fba3c236703f
|
2026-01-21T00:00:00-05:00
|
FlowIID: Single-Step Intrinsic Image Decomposition via Latent Flow Matching
|
arXiv:2601.12329v1 Announce Type: new Abstract: Intrinsic Image Decomposition (IID) separates an image into albedo and shading components. It is a core step in many real-world applications, such as relighting and material editing. Existing IID models achieve good results, but often use a large number of parameters. This makes them costly to combine with other models in real-world settings. To address this problem, we propose a flow matching-based solution. For this, we design a novel architecture, FlowIID, based on latent flow matching. FlowIID combines a VAE-guided latent space with a flow matching module, enabling a stable decomposition of albedo and shading. FlowIID is not only parameter-efficient, but also produces results in a single inference step. Despite its compact design, FlowIID delivers competitive and superior results compared to existing models across various benchmarks. This makes it well-suited for deployment in resource-constrained and real-time vision applications.
|
https://arxiv.org/abs/2601.12329
|
Academic Papers
|
svg
|
5316978e70dc6d87f58102500692f22ed679cfb1a27efd8f721c4e68874355b8
|
2026-01-21T00:00:00-05:00
|
IceWatch: Forecasting Glacial Lake Outburst Floods (GLOFs) using Multimodal Deep Learning
|
arXiv:2601.12330v1 Announce Type: new Abstract: Glacial Lake Outburst Floods (GLOFs) pose a serious threat in high mountain regions. They are hazardous to communities, infrastructure, and ecosystems further downstream. The classical methods of GLOF detection and prediction have so far mainly relied on hydrological modeling, threshold-based lake monitoring, and manual satellite image analysis. These approaches suffer from several drawbacks: slow updates, reliance on manual labor, and losses in accuracy when clouds interfere and/or lack on-site data. To tackle these challenges, we present IceWatch: a novel deep learning framework for GLOF prediction that incorporates both spatial and temporal perspectives. The vision component, RiskFlow, of IceWatch deals with Sentinel-2 multispectral satellite imagery using a CNN-based classifier and predicts GLOF events based on the spatial patterns of snow, ice, and meltwater. Its tabular counterpart confirms this prediction by considering physical dynamics. TerraFlow models glacier velocity from NASA ITS_LIVE time series while TempFlow forecasts near-surface temperature from MODIS LST records; both are trained on long-term observational archives and integrated via harmonized preprocessing and synchronization to enable multimodal, physics-informed GLOF prediction. Both together provide cross-validation, which will improve the reliability and interpretability of GLOF detection. This system ensures strong predictive performance, rapid data processing for real-time use, and robustness to noise and missing information. IceWatch paves the way for automatic, scalable GLOF warning systems. It also holds potential for integration with diverse sensor inputs and global glacier monitoring activities.
|
https://arxiv.org/abs/2601.12330
|
Academic Papers
|
svg
|
6ca5f405cc82c9dc49a3f2c37c3c1268dec4340c3e0f278ddc872846c466be45
|
2026-01-21T00:00:00-05:00
|
Efficient Privacy-Preserving Retrieval Augmented Generation with Distance-Preserving Encryption
|
arXiv:2601.12331v1 Announce Type: new Abstract: RAG has emerged as a key technique for enhancing response quality of LLMs without high computational cost. In traditional architectures, RAG services are provided by a single entity that hosts the dataset within a trusted local environment. However, individuals or small organizations often lack the resources to maintain data storage servers, leading them to rely on outsourced cloud storage. This dependence on untrusted third-party services introduces privacy risks. Embedding-based retrieval mechanisms, commonly used in RAG systems, are vulnerable to privacy leakage such as vector-to-text reconstruction attacks and structural leakage via vector analysis. Several privacy-preserving RAG techniques have been proposed but most existing approaches rely on partially homomorphic encryption, which incurs substantial computational overhead. To address these challenges, we propose an efficient privacy-preserving RAG framework (ppRAG) tailored for untrusted cloud environments that defends against vector-to-text attack, vector analysis, and query analysis. We propose Conditional Approximate Distance-Comparison-Preserving Symmetric Encryption (CAPRISE) that encrypts embeddings while still allowing the cloud to compute similarity between an encrypted query and the encrypted database embeddings. CAPRISE preserves only the relative distance ordering between the encrypted query and each encrypted database embedding, without exposing inter-database distances, thereby enhancing both privacy and efficiency. To mitigate query analysis, we introduce DP by perturbing the query embedding prior to encryption, preventing the cloud from inferring sensitive patterns. Experimental results show that ppRAG achieves efficient processing throughput, high retrieval accuracy, strong privacy guarantees, making it a practical solution for resource-constrained users seeking secure cloud-augmented LLMs.
|
https://arxiv.org/abs/2601.12331
|
Academic Papers
|
svg
|
ba019b94e9bbb252745ce204d1d0dd4c9c7f9ddbb705f0d5a37608b6e26c15b0
|
2026-01-21T00:00:00-05:00
|
Worst-case Nonlinear Regression with Error Bounds
|
arXiv:2601.12334v1 Announce Type: new Abstract: This paper proposes an active-learning approach to worst-case nonlinear regression with deterministic error guarantees. Given a known nonlinear function defined over a compact set, we compute a surrogate model, such as a feedforward neural network, by minimizing the maximum absolute approximation error. To address the nonsmooth nature of the resulting minimax problem, we introduce a smooth approximation of the $L_\infty$-type loss that enables efficient gradient-based training. We iteratively enrich the training set by actively learning points of largest approximation error through global optimization. The resulting models admit certified worst-case error bounds, either constant or input-dependent, over the entire input domain. The approach is demonstrated through approximations of nonlinear functions and nonconvex sets, as well as through the derivation of uncertain models of more complex nonlinear dynamics within a given model class, and the approximation of explicit model predictive control laws.
|
https://arxiv.org/abs/2601.12334
|
Academic Papers
|
svg
|
c6a17357be95c9b0b4649fd0ca2557b0c522bb816af0a6d8f009814578317f1a
|
2026-01-21T00:00:00-05:00
|
Turbo-GoDec: Exploiting the Cluster Sparsity Prior for Hyperspectral Anomaly Detection
|
arXiv:2601.12337v1 Announce Type: new Abstract: As a key task in hyperspectral image processing, hyperspectral anomaly detection has garnered significant attention and undergone extensive research. Existing methods primarily relt on two prior assumption: low-rank background and sparse anomaly, along with additional spatial assumptions of the background. However, most methods only utilize the sparsity prior assumption for anomalies and rarely expand on this hypothesis. From observations of hyperspectral images, we find that anomalous pixels exhibit certain spatial distribution characteristics: they often manifest as small, clustered groups in space, which we refer to as cluster sparsity of anomalies. Then, we combined the cluster sparsity prior with the classical GoDec algorithm, incorporating the cluster sparsity prior into the S-step of GoDec. This resulted in a new hyperspectral anomaly detection method, which we called Turbo-GoDec. In this approach, we modeled the cluster sparsity prior of anomalies using a Markov random field and computed the marginal probabilities of anomalies through message passing on a factor graph. Locations with high anomalous probabilities were treated as the sparse component in the Turbo-GoDec. Experiments are conducted on three real hyperspectral image (HSI) datasets which demonstrate the superior performance of the proposed Turbo-GoDec method in detecting small-size anomalies comparing with the vanilla GoDec (LSMAD) and state-of-the-art anomaly detection methods. The code is available at https://github.com/jiahuisheng/Turbo-GoDec.
|
https://arxiv.org/abs/2601.12337
|
Academic Papers
|
svg
|
e33bbd6510bfbfff7926ab1c2727d405117c3e26d3031f9bb1aa7a77e3c3cedb
|
2026-01-21T00:00:00-05:00
|
Actionable Advice from Reviews via Mixture of LoRA Experts: A Two-LLM Pipeline for Issue Extraction and Business Recommendations
|
arXiv:2601.12338v1 Announce Type: new Abstract: Customer reviews contain detailed, domain specific signals about service failures and user expectations, but converting this unstructured feedback into actionable business decisions remains difficult. We study review-to-action generation: producing concrete, implementable recommendations grounded in review text. We propose a modular two-LLM framework in which an Issue model extracts salient issues and assigns coarse themes, and an Advice model generates targeted operational fixes conditioned on the extracted issue representation. To enable specialization without expensive full fine-tuning, we adapt the Advice model using a mixture of LoRA experts strategy: multiple low-rank adapters are trained and a lightweight gating mechanism performs token-level expert mixing at inference, combining complementary expertise across issue types. We construct synthetic review-issue-advice triples from Yelp reviews (airlines and restaurants) to supervise training, and evaluate recommendations using an eight dimension operational rubric spanning actionability, specificity, feasibility, expected impact, novelty, non-redundancy, bias, and clarity. Across both domains, our approach consistently outperforms prompting-only and single-adapter baselines, yielding higher actionability and specificity while retaining favorable efficiency-quality trade-offs.
|
https://arxiv.org/abs/2601.12338
|
Academic Papers
|
svg
|
713fcb815cccb94fc9383c823225199ae4538b6ec4eff559fb0b20a48e44b942
|
2026-01-21T00:00:00-05:00
|
Time-Continuous Modeling for Temporal Affective Pattern Recognition in LLMs
|
arXiv:2601.12341v1 Announce Type: new Abstract: This paper introduces a dataset and conceptual framework for LLMs to mimic real world emotional dynamics through time and in-context learning leveraging physics-informed neural network, opening a possibility for interpretable dialogue modeling.
|
https://arxiv.org/abs/2601.12341
|
Academic Papers
|
svg
|
1f20bf250c47cbc4222d5ddc3a66be4b0b3f51286c921ebe2160760447a649e0
|
2026-01-21T00:00:00-05:00
|
MMDeepResearch-Bench: A Benchmark for Multimodal Deep Research Agents
|
arXiv:2601.12346v1 Announce Type: new Abstract: Deep Research Agents (DRAs) generate citation-rich reports via multi-step search and synthesis, yet existing benchmarks mainly target text-only settings or short-form multimodal QA, missing end-to-end multimodal evidence use. We introduce MMDeepResearch-Bench (MMDR-Bench), a benchmark of 140 expert-crafted tasks across 21 domains, where each task provides an image-text bundle to evaluate multimodal understanding and citation-grounded report generation. Compared to prior setups, MMDR-Bench emphasizes report-style synthesis with explicit evidence use, where models must connect visual artifacts to sourced claims and maintain consistency across narrative, citations, and visual references. We further propose a unified, interpretable evaluation pipeline: Formula-LLM Adaptive Evaluation (FLAE) for report quality, Trustworthy Retrieval-Aligned Citation Evaluation (TRACE) for citation-grounded evidence alignment, and Multimodal Support-Aligned Integrity Check (MOSAIC) for text-visual integrity, each producing fine-grained signals that support error diagnosis beyond a single overall score. Experiments across 25 state-of-the-art models reveal systematic trade-offs between generation quality, citation discipline, and multimodal grounding, highlighting that strong prose alone does not guarantee faithful evidence use and that multimodal integrity remains a key bottleneck for deep research agents.
|
https://arxiv.org/abs/2601.12346
|
Academic Papers
|
svg
|
895dd2612f8623fe006e6f289c156e5b0f5e1bfa6e03725981c995c6f9b7cf6d
|
2026-01-21T00:00:00-05:00
|
RIPPLE++: An Incremental Framework for Efficient GNN Inference on Evolving Graphs
|
arXiv:2601.12347v1 Announce Type: new Abstract: Real-world graphs are dynamic, with frequent updates to their structure and features due to evolving vertex and edge properties. These continual changes pose significant challenges for efficient inference in graph neural networks (GNNs). Existing vertex-wise and layer-wise inference approaches are ill-suited for dynamic graphs, as they incur redundant computations, large neighborhood traversals, and high communication costs, especially in distributed settings. Additionally, while sampling-based approaches can be adopted to approximate final layer embeddings, these are often not preferred in critical applications due to their non-determinism. These limitations hinder low-latency inference required in real-time applications. To address this, we propose RIPPLE++, a framework for streaming GNN inference that efficiently and accurately updates embeddings in response to changes in the graph structure or features. RIPPLE++ introduces a generalized incremental programming model that captures the semantics of GNN aggregation functions and incrementally propagates updates to affected neighborhoods. RIPPLE++ accommodates all common graph updates, including vertex/edge addition/deletions and vertex feature updates. RIPPLE++ supports both single-machine and distributed deployments. On a single machine, it achieves up to $56$K updates/sec on sparse graphs like Arxiv ($169$K vertices, $1.2$M edges), and about $7.6$K updates/sec on denser graphs like Products ($2.5$M vertices, $123.7$M edges), with latencies of $0.06$--$960$ms, and outperforming state-of-the-art baselines by $2.2$--$24\times$ on throughput. In distributed settings, RIPPLE++ offers up to $\approx25\times$ higher throughput and $20\times$ lower communication costs compared to recomputing baselines.
|
https://arxiv.org/abs/2601.12347
|
Academic Papers
|
svg
|
6d12365fe4eb57c9b1ee12b549639f0ff159370f651c5e3b6f5389e42b39e647
|
2026-01-21T00:00:00-05:00
|
Generative AI Agents for Controllable and Protected Content Creation
|
arXiv:2601.12348v1 Announce Type: new Abstract: The proliferation of generative AI has transformed creative workflows, yet current systems face critical challenges in controllability and content protection. We propose a novel multi-agent framework that addresses both limitations through specialized agent roles and integrated watermarking mechanisms. Unlike existing multi-agent systems focused solely on generation quality, our approach uniquely combines controllable content synthesis with provenance protection during the generation process itself. The framework orchestrates Director/Planner, Generator, Reviewer, Integration, and Protection agents with human-in-the-loop feedback to ensure alignment with user intent while embedding imperceptible digital watermarks. We formalize the pipeline as a joint optimization objective unifying controllability, semantic alignment, and protection robustness. This work contributes to responsible generative AI by positioning multi-agent architectures as a solution for trustworthy creative workflows with built-in ownership tracking and content traceability.
|
https://arxiv.org/abs/2601.12348
|
Academic Papers
|
svg
|
d18a0bea5e86623e55db9b25b06bcb2aad76599bc8f5c5866685c804116328b5
|
2026-01-21T00:00:00-05:00
|
Zero-Permission Manipulation: Can We Trust Large Multimodal Model Powered GUI Agents?
|
arXiv:2601.12349v1 Announce Type: new Abstract: Large multimodal model powered GUI agents are emerging as high-privilege operators on mobile platforms, entrusted with perceiving screen content and injecting inputs. However, their design operates under the implicit assumption of Visual Atomicity: that the UI state remains invariant between observation and action. We demonstrate that this assumption is fundamentally invalid in Android, creating a critical attack surface. We present Action Rebinding, a novel attack that allows a seemingly-benign app with zero dangerous permissions to rebind an agent's execution. By exploiting the inevitable observation-to-action gap inherent in the agent's reasoning pipeline, the attacker triggers foreground transitions to rebind the agent's planned action toward the target app. We weaponize the agent's task-recovery logic and Android's UI state preservation to orchestrate programmable, multi-step attack chains. Furthermore, we introduce an Intent Alignment Strategy (IAS) that manipulates the agent's reasoning process to rationalize UI states, enabling it to bypass verification gates (e.g., confirmation dialogs) that would otherwise be rejected. We evaluate Action Rebinding Attacks on six widely-used Android GUI agents across 15 tasks. Our results demonstrate a 100% success rate for atomic action rebinding and the ability to reliably orchestrate multi-step attack chains. With IAS, the success rate in bypassing verification gates increases (from 0% to up to 100%). Notably, the attacker application requires no sensitive permissions and contains no privileged API calls, achieving a 0% detection rate across malware scanners (e.g., VirusTotal). Our findings reveal a fundamental architectural flaw in current agent-OS integration and provide critical insights for the secure design of future agent systems. To access experimental logs and demonstration videos, please contact yi_qian@smail.nju.edu.cn.
|
https://arxiv.org/abs/2601.12349
|
Academic Papers
|
svg
|
ce023a68f51944ef45db20d44ddf47571313b63f2a8303ca0ba89874e85afb0f
|
2026-01-21T00:00:00-05:00
|
Analyzing Collection Strategies: A Computational Perspective on the Coupon Collector Problem
|
arXiv:2601.12351v1 Announce Type: new Abstract: The Coupon Collector Problem (CCP) is a well-known combinatorial problem that seeks to estimate the number of random draws required to complete a collection of $n$ distinct coupon types. Various generalizations of this problem have been applied in numerous engineering domains. However, practical applications are often hindered by the computational challenges associated with deriving numerical results for moments and distributions. In this work, we present three algorithms for solving the most general form of the CCP, where coupons are collected under any arbitrary drawing probability, with the objective of obtaining $t$ copies of a subset of $k$ coupons from a total of $n$. The First algorithm provides the base model to compute the expectation, variance, and the second moment of the collection process. The second algorithm utilizes the construction of the base model and computes the same values in polynomial time with respect to $n$ under the uniform drawing distribution, and the third algorithm extends to any general drawing distribution. All algorithms leverage Markov models specifically designed to address computational challenges, ensuring exact computation of the expectation and variance of the collection process. Their implementation uses a dynamic programming approach that follows from the Markov models framework, and their time complexity is analyzed accordingly.
|
https://arxiv.org/abs/2601.12351
|
Academic Papers
|
svg
|
fd988eb7020f425ea937d8f798cdc62ad7b90ac0d3f647e33842e46a50feb7d1
|
2026-01-21T00:00:00-05:00
|
From Shallow Waters to Mariana Trench: A Survey of Bio-inspired Underwater Soft Robots
|
arXiv:2601.12353v1 Announce Type: new Abstract: Sample Exploring the ocean environment holds profound significance in areas such as resource exploration and ecological protection. Underwater robots struggle with extreme water pressure and often cause noise and damage to the underwater ecosystem, while bio-inspired soft robots draw inspiration from aquatic creatures to address these challenges. These bio-inspired approaches enable robots to withstand high water pressure, minimize drag, operate with efficient manipulation and sensing systems, and interact with the environment in an eco-friendly manner. Consequently, bio-inspired soft robots have emerged as a promising field for ocean exploration. This paper reviews recent advancements in underwater bio-inspired soft robots, analyses their design considerations when facing different desired functions, bio-inspirations, ambient pressure, temperature, light, and biodiversity , and finally explores the progression from bio-inspired principles to practical applications in the field and suggests potential directions for developing the next generation of underwater soft robots.
|
https://arxiv.org/abs/2601.12353
|
Academic Papers
|
svg
|
6fced5f97bfc20eff1c17ead1184d69ab3220ef92cefdfe005069e122f6739bf
|
2026-01-21T00:00:00-05:00
|
LB-MCTS: Synergizing Large Language Models and Bayesian Optimization for Efficient CASH
|
arXiv:2601.12355v1 Announce Type: new Abstract: To lower the expertise barrier in machine learning, the AutoML community has focused on the CASH problem, a fundamental challenge that automates the process of algorithm selection and hyperparameter tuning. While traditional methods like Bayesian Optimization (BO) struggle with cold-start issues, Large Language Models (LLMs) can mitigate these via semantic priors. However, existing LLM-based optimizers generalize poorly to the high-dimensional, structured CASH space. We propose LB-MCTS, a framework synergizing LLMs and BO within a Monte Carlo Tree Search structure. It maximizes LLM reasoning with Selective Tuning Memory (STM) and explicit exploration-exploitation trade-off. It combines the strengths of both paradigms by dynamically shifting from LLM-driven to BO-driven proposals as data accumulates. Experiments on 104 AMLB datasets demonstrate the superiority of LB-MCTS over the competitive baselines.
|
https://arxiv.org/abs/2601.12355
|
Academic Papers
|
svg
|
651b53d356b47baa023009a14f9c8c3692572ef54cef8a0d1b7c2f9e48f6e3e3
|
2026-01-21T00:00:00-05:00
|
SimpleMatch: A Simple and Strong Baseline for Semantic Correspondence
|
arXiv:2601.12357v1 Announce Type: new Abstract: Recent advances in semantic correspondence have been largely driven by the use of pre-trained large-scale models. However, a limitation of these approaches is their dependence on high-resolution input images to achieve optimal performance, which results in considerable computational overhead. In this work, we address a fundamental limitation in current methods: the irreversible fusion of adjacent keypoint features caused by deep downsampling operations. This issue is triggered when semantically distinct keypoints fall within the same downsampled receptive field (e.g., 16x16 patches). To address this issue, we present SimpleMatch, a simple yet effective framework for semantic correspondence that delivers strong performance even at low resolutions. We propose a lightweight upsample decoder that progressively recovers spatial detail by upsampling deep features to 1/4 resolution, and a multi-scale supervised loss that ensures the upsampled features retain discriminative features across different spatial scales. In addition, we introduce sparse matching and window-based localization to optimize training memory usage and reduce it by 51%. At a resolution of 252x252 (3.3x smaller than current SOTA methods), SimpleMatch achieves superior performance with 84.1% PCK@0.1 on the SPair-71k benchmark. We believe this framework provides a practical and efficient baseline for future research in semantic correspondence. Code is available at: https://github.com/hailong23-jin/SimpleMatch.
|
https://arxiv.org/abs/2601.12357
|
Academic Papers
|
svg
|
61bf0e5fc130c05e5b5bee58466e0b667d38f131f93a9383b118c1bf59ad3e81
|
2026-01-21T00:00:00-05:00
|
From Prompts to Pavement: LMMs-based Agentic Behavior-Tree Generation Framework for Autonomous Vehicles
|
arXiv:2601.12358v1 Announce Type: new Abstract: Autonomous vehicles (AVs) require adaptive behavior planners to navigate unpredictable, real-world environments safely. Traditional behavior trees (BTs) offer structured decision logic but are inherently static and demand labor-intensive manual tuning, limiting their applicability at SAE Level 5 autonomy. This paper presents an agentic framework that leverages large language models (LLMs) and multi-modal vision models (LVMs) to generate and adapt BTs on the fly. A specialized Descriptor agent applies chain-of-symbols prompting to assess scene criticality, a Planner agent constructs high-level sub-goals via in-context learning, and a Generator agent synthesizes executable BT sub-trees in XML format. Integrated into a CARLA+Nav2 simulation, our system triggers only upon baseline BT failure, demonstrating successful navigation around unexpected obstacles (e.g., street blockage) with no human intervention. Compared to a static BT baseline, this approach is a proof-of-concept that extends to diverse driving scenarios.
|
https://arxiv.org/abs/2601.12358
|
Academic Papers
|
svg
|
b348a2fd892780723ae607f2376ca58d8296cb03d1c244eb71720bfd95f2cfa8
|
2026-01-21T00:00:00-05:00
|
Zero-Shot Embedding Drift Detection: A Lightweight Defense Against Prompt Injections in LLMs
|
arXiv:2601.12359v1 Announce Type: new Abstract: Prompt injection attacks have become an increasing vulnerability for LLM applications, where adversarial prompts exploit indirect input channels such as emails or user-generated content to circumvent alignment safeguards and induce harmful or unintended outputs. Despite advances in alignment, even state-of-the-art LLMs remain broadly vulnerable to adversarial prompts, underscoring the urgent need for robust, productive, and generalizable detection mechanisms beyond inefficient, model-specific patches. In this work, we propose Zero-Shot Embedding Drift Detection (ZEDD), a lightweight, low-engineering-overhead framework that identifies both direct and indirect prompt injection attempts by quantifying semantic shifts in embedding space between benign and suspect inputs. ZEDD operates without requiring access to model internals, prior knowledge of attack types, or task-specific retraining, enabling efficient zero-shot deployment across diverse LLM architectures. Our method uses adversarial-clean prompt pairs and measures embedding drift via cosine similarity to capture subtle adversarial manipulations inherent to real-world injection attacks. To ensure robust evaluation, we assemble and re-annotate the comprehensive LLMail-Inject dataset spanning five injection categories derived from publicly available sources. Extensive experiments demonstrate that embedding drift is a robust and transferable signal, outperforming traditional methods in detection accuracy and operational efficiency. With greater than 93% accuracy in classifying prompt injections across model architectures like Llama 3, Qwen 2, and Mistral and a false positive rate of <3%, our approach offers a lightweight, scalable defense layer that integrates into existing LLM pipelines, addressing a critical gap in securing LLM-powered systems to withstand adaptive adversarial threats.
|
https://arxiv.org/abs/2601.12359
|
Academic Papers
|
svg
|
545e6075d8e16754c1996f9d52353b3ea64d1f6224d2c30105ee7cac7f3dabf1
|
2026-01-21T00:00:00-05:00
|
Discovering 100+ Compiler Defects in 72 Hours via LLM-Driven Semantic Logic Recomposition
|
arXiv:2601.12360v1 Announce Type: new Abstract: Compilers constitute the foundational root-of-trust in software supply chains; however, their immense complexity inevitably conceals critical defects. Recent research has attempted to leverage historical bugs to design new mutation operators or fine-tune models to increase program diversity for compiler fuzzing.We observe, however, that bugs manifest primarily based on the semantics of input programs rather than their syntax. Unfortunately, current approaches, whether relying on syntactic mutation or general Large Language Model (LLM) fine-tuning, struggle to preserve the specific semantics found in the logic of bug-triggering programs. Consequently, these critical semantic triggers are often lost, resulting in a limitation of the diversity of generated programs. To explicitly reuse such semantics, we propose FeatureFuzz, a compiler fuzzer that combines features to generate programs. We define a feature as a decoupled primitive that encapsulates a natural language description of a bug-prone invariant, such as an out-of-bounds array access, alongside a concrete code witness of its realization. FeatureFuzz operates via a three-stage workflow: it first extracts features from historical bug reports, synthesizes coherent groups of features, and finally instantiates these groups into valid programs for compiler fuzzing. We evaluated FeatureFuzz on GCC and LLVM. Over 24-hour campaigns, FeatureFuzz uncovered 167 unique crashes, which is 2.78x more than the second-best fuzzer. Furthermore, through a 72-hour fuzzing campaign, FeatureFuzz identified 106 bugs in GCC and LLVM, 76 of which have already been confirmed by compiler developers, validating the approach's ability to stress-test modern compilers effectively.
|
https://arxiv.org/abs/2601.12360
|
Academic Papers
|
svg
|
da2a35db6f5745350bb795efa06a12a0c2b6ae53790a57f22aee8b4bc12c78b4
|
2026-01-21T00:00:00-05:00
|
Complexity of Model Checking Second-Order Hyperproperties on Finite Structures
|
arXiv:2601.12361v1 Announce Type: new Abstract: We study the model checking problem of Hyper2LTL over finite structures. Hyper2LTL is a second-order hyperlogic, that extends the well-studied logic HyperLTL by adding quantification over sets of traces, to express complex hyperproperties such as epistemic and asynchronous hyperproperties. While Hyper2LTL is very expressive, its expressiveness comes with a price, and its general model checking problem is undecidable. This motivates us to study the model checking problem for Hyper2LTL over finite structures -- tree-shaped or acyclic graphs, which are particularly useful for monitoring purposes. We show that Hyper2LTL model checking is decidable on finite structures. It is in PSPACE (in the size of the model) on tree-shaped models and in EXPSPACE on acyclic models. Additionally, we show that for an expressive fragment of Hyper2LTL, namely the Fixpoint Hyper2LTLfp fragment, the model checking problem is much simpler and is P-complete on tree-shaped models and EXP-complete on acyclic models. Last, we present some preliminary results that take into account not only the size of the model, but also the formula size.
|
https://arxiv.org/abs/2601.12361
|
Academic Papers
|
svg
|
dcd4c6bb69c4f68f421e07f9970358ec184cf7a24f74e8417526e8f63055f25b
|
2026-01-21T00:00:00-05:00
|
Machine Learning-Based Framework for Real Time Detection and Early Prediction of Control Valve Stiction in Industrial Control Systems
|
arXiv:2601.12362v1 Announce Type: new Abstract: Control valve stiction, a friction that prevents smooth valve movement, is a common fault in industrial process systems that causes instability, equipment wear, and higher maintenance costs. Many plants still operate with conventional valves that lack real time monitoring, making early predictions challenging. This study presents a machine learning (ML) framework for detecting and predicting stiction using only routinely collected process signals: the controller output (OP) from control systems and the process variable (PV), such as flow rate. Three deep learning models were developed and compared: a Convolutional Neural Network (CNN), a hybrid CNN with a Support Vector Machine (CNN-SVM), and a Long Short-Term Memory (LSTM) network. To train these models, a data-driven labeling method based on slope ratio analysis was applied to a real oil and gas refinery dataset. The LSTM model achieved the highest accuracy and was able to predict stiction up to four hours in advance. To the best of the authors' knowledge, this is the first study to demonstrate ML based early prediction of control valve stiction from real industry data. The proposed framework can be integrated into existing control systems to support predictive maintenance, reduce downtime, and avoid unnecessary hardware replacement.
|
https://arxiv.org/abs/2601.12362
|
Academic Papers
|
svg
|
cbe79dc6f6421b6a0c552a3fea9d6528707234c4a25ca0097185312a58c9ba3d
|
2026-01-21T00:00:00-05:00
|
DepthCropSeg++: Scaling a Crop Segmentation Foundation Model With Depth-Labeled Data
|
arXiv:2601.12366v1 Announce Type: new Abstract: DepthCropSeg++: a foundation model for crop segmentation, capable of segmenting different crop species under open in-field environment. Crop segmentation is a fundamental task for modern agriculture, which closely relates to many downstream tasks such as plant phenotyping, density estimation, and weed control. In the era of foundation models, a number of generic large language and vision models have been developed. These models have demonstrated remarkable real world generalization due to significant model capacity and largescale datasets. However, current crop segmentation models mostly learn from limited data due to expensive pixel-level labelling cost, often performing well only under specific crop types or controlled environment. In this work, we follow the vein of our previous work DepthCropSeg, an almost unsupervised approach to crop segmentation, to scale up a cross-species and crossscene crop segmentation dataset, with 28,406 images across 30+ species and 15 environmental conditions. We also build upon a state-of-the-art semantic segmentation architecture ViT-Adapter architecture, enhance it with dynamic upsampling for improved detail awareness, and train the model with a two-stage selftraining pipeline. To systematically validate model performance, we conduct comprehensive experiments to justify the effectiveness and generalization capabilities across multiple crop datasets. Results demonstrate that DepthCropSeg++ achieves 93.11% mIoU on a comprehensive testing set, outperforming both supervised baselines and general-purpose vision foundation models like Segmentation Anything Model (SAM) by significant margins (+0.36% and +48.57% respectively). The model particularly excels in challenging scenarios including night-time environment (86.90% mIoU), high-density canopies (90.09% mIoU), and unseen crop varieties (90.09% mIoU), indicating a new state of the art for crop segmentation.
|
https://arxiv.org/abs/2601.12366
|
Academic Papers
|
svg
|
556d252c20af6514db4e59d75efb13e5f25eb4f9a87f7edb711c910f03982f7f
|
2026-01-21T00:00:00-05:00
|
User-to-Vehicle Interaction in Smart Mobility: The GO-DRiVeS Autonomous Ride-Sharing Application
|
arXiv:2601.12367v1 Announce Type: new Abstract: This paper introduces the GO-DRiVeS application, an on demand ride sharing and requesting mobile application tailored specifically to save long walks and challenges which are time consuming and tiring especially during hot days or when carrying heavy items, faced by university students and staff. The GO-DRiVeS application was developed following the Agile methodology for its flexibility. In addition to, using the mobile application system architecture and client-server architecture. GO-DRiVeS was implemented using React Native (Expo) for the frontend, Node.js and Express for the backend, and MongoDB as the database; based on a detailed analyses to the existing transportation application, comparing their frameworks and identifying their essential functionalities. GO-DRiVeS supports core features like user registration, ride requesting and real-time tracking.In addition to handling multiple requests at the same time in a first come first serve manner. The application was developed based on these features, and the results were conducted in the form of multiple experiments that demonstrated stable behavior in handling the requests, as presented in the Methodology and Results chapters.
|
https://arxiv.org/abs/2601.12367
|
Academic Papers
|
svg
|
3729f37aa48973906ac9efeee6357c887de0365efb320aef734f9dce884d2765
|
2026-01-21T00:00:00-05:00
|
Can Deep Research Agents Find and Organize? Evaluating the Synthesis Gap with Expert Taxonomies
|
arXiv:2601.12369v1 Announce Type: new Abstract: Deep Research Agents are increasingly used for automated survey generation. However, whether they can write surveys like human experts remains unclear. Existing benchmarks focus on fluency or citation accuracy, but none evaluates the core capabilities: retrieving essential papers and organizing them into coherent knowledge structures. We introduce TaxoBench, a diagnostic benchmark derived from 72 highly-cited computer science surveys. We manually extract expert-authored taxonomy trees containing 3,815 precisely categorized citations as ground truth. Our benchmark supports two evaluation modes: Deep Research mode tests end-to-end retrieval and organization given only a topic, while Bottom-Up mode isolates structuring capability by providing the exact papers human experts used. We evaluate 7 leading Deep Research agents and 12 frontier LLMs. Results reveal a dual bottleneck: the best agent recalls only 20.9% of expert-selected papers, and even with perfect input, the best model achieves only 0.31 ARI in organization. Current deep research agents remain far from expert-level survey writing. Our benchmark is publicly available at https://github.com/KongLongGeFDU/TaxoBench.
|
https://arxiv.org/abs/2601.12369
|
Academic Papers
|
svg
|
a889e7130e92f4b2675472757d103cfeb4938798106aafb70ee38707eecdd6fd
|
2026-01-21T00:00:00-05:00
|
CD-TWINSAFE: A ROS-enabled Digital Twin for Scene Understanding and Safety Emerging V2I Technology
|
arXiv:2601.12373v1 Announce Type: new Abstract: In this paper, the CD-TWINSAFE is introduced, a V2I-based digital twin for Autonomous Vehicles. The proposed architecture is composed of two stacks running simultaneously, an on-board driving stack that includes a stereo camera for scene understanding, and a digital twin stack that runs an Unreal Engine 5 replica of the scene viewed by the camera as well as returning safety alerts to the cockpit. The on-board stack is implemented on the vehicle side including 2 main autonomous modules; localization and perception. The position and orientation of the ego vehicle are obtained using on-board sensors. Furthermore, the perception module is responsible for processing 20-fps images from stereo camera and understands the scene through two complementary pipelines. The pipeline are working on object detection and feature extraction including object velocity, yaw and the safety metrics time-to-collision and time-headway. The collected data form the driving stack are sent to the infrastructure side through the ROS-enabled architecture in the form of custom ROS2 messages and sent over UDP links that ride a 4G modem for V2I communication. The environment is monitored via the digital twin through the shared messages which update the information of the spawned ego vehicle and detected objects based on the real-time localization and perception data. Several tests with different driving scenarios to confirm the validity and real-time response of the proposed architecture.
|
https://arxiv.org/abs/2601.12373
|
Academic Papers
|
svg
|
9856b32e60ce05011eefeb5ee518901757cc7a0ed97a792cf4b5304c9c58ee8d
|
2026-01-21T00:00:00-05:00
|
A Scalable Entity-Based Framework for Auditing Bias in LLMs
|
arXiv:2601.12374v1 Announce Type: new Abstract: Existing approaches to bias evaluation in large language models (LLMs) trade ecological validity for statistical control, relying on artificial prompts that poorly reflect real-world use, or on naturalistic tasks that lack scale and rigor. We introduce a scalable bias-auditing framework using named entities as probes to measure structural disparities in model behavior. We show that synthetic data reliably reproduces bias patterns observed in natural text, enabling large-scale analysis. Using this approach, we conduct the largest bias audit to date, comprising 1.9 billion data points across multiple entity types, tasks, languages, models, and prompting strategies. Our results reveal systematic biases: models penalize right-wing politicians, favor left-wing politicians, prefer Western and wealthy nations over the Global South, favor Western companies, and penalize firms in the defense and pharmaceutical sectors. While instruction tuning reduces bias, increasing model scale amplifies it, and prompting in Chinese or Russian does not attenuate Western-aligned preferences. These results indicate that LLMs should undergo rigorous auditing before deployment in high-stakes applications.
|
https://arxiv.org/abs/2601.12374
|
Academic Papers
|
svg
|
b550c71c5d5dfadb76a90de31161782a99443d91954f0186e072f997134dccc6
|
2026-01-21T00:00:00-05:00
|
LiQSS: Post-Transformer Linear Quantum-Inspired State-Space Tensor Networks for Real-Time 6G
|
arXiv:2601.12375v1 Announce Type: new Abstract: Proactive and agentic control in Sixth-Generation (6G) Open Radio Access Networks (O-RAN) requires control-grade prediction under stringent Near-Real-Time (Near-RT) latency and computational constraints. While Transformer-based models are effective for sequence modeling, their quadratic complexity limits scalability in Near-RT RAN Intelligent Controller (RIC) analytics. This paper investigates a post-Transformer design paradigm for efficient radio telemetry forecasting. We propose a quantum-inspired many-body state-space tensor network that replaces self-attention with stable structured state-space dynamics kernels, enabling linear-time sequence modeling. Tensor-network factorizations in the form of Tensor Train (TT) / Matrix Product State (MPS) representations are employed to reduce parameterization and data movement in both input projections and prediction heads, while lightweight channel gating and mixing layers capture non-stationary cross-Key Performance Indicator (KPI) dependencies. The proposed model is instantiated as an agentic perceive-predict xApp and evaluated on a bespoke O-RAN KPI time-series dataset comprising 59,441 sliding windows across 13 KPIs, using Reference Signal Received Power (RSRP) forecasting as a representative use case. Our proposed Linear Quantum-Inspired State-Space (LiQSS) model is 10.8x-15.8x smaller and approximately 1.4x faster than prior structured state-space baselines. Relative to Transformer-based models, LiQSS achieves up to a 155x reduction in parameter count and up to 2.74x faster inference, without sacrificing forecasting accuracy.
|
https://arxiv.org/abs/2601.12375
|
Academic Papers
|
svg
|
6fed2da31e4b3e52da18bcfb135de4b6f69383f1f58df81e2754b9d891ed3b88
|
2026-01-21T00:00:00-05:00
|
LR-DWM: Efficient Watermarking for Diffusion Language Models
|
arXiv:2601.12376v1 Announce Type: new Abstract: Watermarking (WM) is a critical mechanism for detecting and attributing AI-generated content. Current WM methods for Large Language Models (LLMs) are predominantly tailored for autoregressive (AR) models: They rely on tokens being generated sequentially, and embed stable signals within the generated sequence based on the previously sampled text. Diffusion Language Models (DLMs) generate text via non-sequential iterative denoising, which requires significant modification to use WM methods designed for AR models. Recent work proposed to watermark DLMs by inverting the process when needed, but suffers significant computational or memory overhead. We introduce Left-Right Diffusion Watermarking (LR-DWM), a scheme that biases the generated token based on both left and right neighbors, when they are available. LR-DWM incurs minimal runtime and memory overhead, remaining close to the non-watermarked baseline DLM while enabling reliable statistical detection under standard evaluation settings. Our results demonstrate that DLMs can be watermarked efficiently, achieving high detectability with negligible computational and memory overhead.
|
https://arxiv.org/abs/2601.12376
|
Academic Papers
|
svg
|
2ca324d837510f88419cc86a075baadb01e972c3c84b620f1071b96989877bdc
|
2026-01-21T00:00:00-05:00
|
R-VoxelMap: Accurate Voxel Mapping with Recursive Plane Fitting for Online LiDAR Odometry
|
arXiv:2601.12377v1 Announce Type: new Abstract: This paper proposes R-VoxelMap, a novel voxel mapping method that constructs accurate voxel maps using a geometry-driven recursive plane fitting strategy to enhance the localization accuracy of online LiDAR odometry. VoxelMap and its variants typically fit and check planes using all points in a voxel, which may lead to plane parameter deviation caused by outliers, over segmentation of large planes, and incorrect merging across different physical planes. To address these issues, R-VoxelMap utilizes a geometry-driven recursive construction strategy based on an outlier detect-and-reuse pipeline. Specifically, for each voxel, accurate planes are first fitted while separating outliers using random sample consensus (RANSAC). The remaining outliers are then propagated to deeper octree levels for recursive processing, ensuring a detailed representation of the environment. In addition, a point distribution-based validity check algorithm is devised to prevent erroneous plane merging. Extensive experiments on diverse open-source LiDAR(-inertial) simultaneous localization and mapping (SLAM) datasets validate that our method achieves higher accuracy than other state-of-the-art approaches, with comparable efficiency and memory usage. Code will be available on GitHub.
|
https://arxiv.org/abs/2601.12377
|
Academic Papers
|
svg
|
c31913f1aee09cb49c103bb3622b5aa2b9a21ec6f00d10b5027a8ae7d052a4fc
|
2026-01-21T00:00:00-05:00
|
Utilizing the Score of Data Distribution for Hyperspectral Anomaly Detection
|
arXiv:2601.12379v1 Announce Type: new Abstract: Hyperspectral images (HSIs) are a type of image that contains abundant spectral information. As a type of real-world data, the high-dimensional spectra in hyperspectral images are actually determined by only a few factors, such as chemical composition and illumination. Thus, spectra in hyperspectral images are highly likely to satisfy the manifold hypothesis. Based on the hyperspectral manifold hypothesis, we propose a novel hyperspectral anomaly detection method (named ScoreAD) that leverages the time-dependent gradient field of the data distribution (i.e., the score), as learned by a score-based generative model (SGM). Our method first trains the SGM on the entire set of spectra from the hyperspectral image. At test time, each spectrum is passed through a perturbation kernel, and the resulting perturbed spectrum is fed into the trained SGM to obtain the estimated score. The manifold hypothesis of HSIs posits that background spectra reside on one or more low-dimensional manifolds. Conversely, anomalous spectra, owing to their unique spectral signatures, are considered outliers that do not conform to the background manifold. Based on this fundamental discrepancy in their manifold distributions, we leverage a generative SGM to achieve hyperspectral anomaly detection. Experiments on the four hyperspectral datasets demonstrate the effectiveness of the proposed method. The code is available at https://github.com/jiahuisheng/ScoreAD.
|
https://arxiv.org/abs/2601.12379
|
Academic Papers
|
svg
|
d4ff16adef9c6352266fe6977a5cb0997840f8a61f02dec6dee1785d5464369f
|
2026-01-21T00:00:00-05:00
|
Statistical-Neural Interaction Networks for Interpretable Mixed-Type Data Imputation
|
arXiv:2601.12380v1 Announce Type: new Abstract: Real-world tabular databases routinely combine continuous measurements and categorical records, yet missing entries are pervasive and can distort downstream analysis. We propose Statistical-Neural Interaction (SNI), an interpretable mixed-type imputation framework that couples correlation-derived statistical priors with neural feature attention through a Controllable-Prior Feature Attention (CPFA) module. CPFA learns head-wise prior-strength coefficients $\{\lambda_h\}$ that softly regularize attention toward the prior while allowing data-driven deviations when nonlinear patterns appear to be present in the data. Beyond imputation, SNI aggregates attention maps into a directed feature-dependency matrix that summarizes which variables the imputer relied on, without requiring post-hoc explainers. We evaluate SNI against six baselines (Mean/Mode, MICE, KNN, MissForest, GAIN, MIWAE) on six datasets spanning ICU monitoring, population surveys, socio-economic statistics, and engineering applications. Under MCAR/strict-MAR at 30\% missingness, SNI is generally competitive on continuous metrics but is often outperformed by accuracy-first baselines (MissForest, MIWAE) on categorical variables; in return, it provides intrinsic dependency diagnostics and explicit statistical-neural trade-off parameters. We additionally report MNAR stress tests (with a mask-aware variant) and discuss computational cost, limitations -- particularly for severely imbalanced categorical targets -- and deployment scenarios where interpretability may justify the trade-off.
|
https://arxiv.org/abs/2601.12380
|
Academic Papers
|
svg
|
19fe2f47b5035840485864a0960e1b9a499a25bbbad0d7e1547facc12f4a0d3f
|
2026-01-21T00:00:00-05:00
|
A Hierarchical Benchmark of Foundation Models for Dermatology
|
arXiv:2601.12382v1 Announce Type: new Abstract: Foundation models have transformed medical image analysis by providing robust feature representations that reduce the need for large-scale task-specific training. However, current benchmarks in dermatology often reduce the complex diagnostic taxonomy to flat, binary classification tasks, such as distinguishing melanoma from benign nevi. This oversimplification obscures a model's ability to perform fine-grained differential diagnoses, which is critical for clinical workflow integration. This study evaluates the utility of embeddings derived from ten foundation models, spanning general computer vision, general medical imaging, and dermatology-specific domains, for hierarchical skin lesion classification. Using the DERM12345 dataset, which comprises 40 lesion subclasses, we calculated frozen embeddings and trained lightweight adapter models using a five-fold cross-validation. We introduce a hierarchical evaluation framework that assesses performance across four levels of clinical granularity: 40 Subclasses, 15 Main Classes, 2 and 4 Superclasses, and Binary Malignancy. Our results reveal a "granularity gap" in model capabilities: MedImageInsights achieved the strongest overall performance (97.52% weighted F1-Score on Binary Malignancy detection) but declined to 65.50% on fine-grained 40-class subtype classification. Conversely, MedSigLip (69.79%) and dermatology-specific models (Derm Foundation and MONET) excelled at fine-grained 40-class subtype discrimination while achieving lower overall performance than MedImageInsights on broader classification tasks. Our findings suggest that while general medical foundation models are highly effective for high-level screening, specialized modeling strategies are necessary for the granular distinctions required in diagnostic support systems.
|
https://arxiv.org/abs/2601.12382
|
Academic Papers
|
svg
|
2fe3069604bee6213be9fe2fe26ccd0e0aba5bd7a67bf91252f0d72773d52281
|
2026-01-21T00:00:00-05:00
|
Context-Free Grammar Inference for Complex Programming Languages in Black Box Settings
|
arXiv:2601.12385v1 Announce Type: new Abstract: Grammar inference for complex programming languages remains a significant challenge, as existing approaches fail to scale to real world datasets within practical time constraints. In our experiments, none of the state-of-the-art tools, including Arvada, Treevada and Kedavra were able to infer grammars for complex languages such as C, C++, and Java within 48 hours. Arvada and Treevada perform grammar inference directly on full-length input examples, which proves inefficient for large files commonly found in such languages. While Kedavra introduces data decomposition to create shorter examples for grammar inference, its lexical analysis still relies on the original inputs. Additionally, its strict no-overgeneralization constraint limits the construction of complex grammars. To overcome these limitations, we propose Crucio, which builds a decomposition forest to extract short examples for lexical and grammar inference via a distributional matrix. Experimental results show that Crucio is the only method capable of successfully inferring grammars for complex programming languages (where the number of nonterminals is up to 23x greater than in prior benchmarks) within reasonable time limits. On the prior simple benchmark, Crucio achieves an average recall improvement of 1.37x and 1.19x over Treevada and Kedavra, respectively, and improves F1 scores by 1.21x and 1.13x.
|
https://arxiv.org/abs/2601.12385
|
Academic Papers
|
svg
|
748e312d9402cac6c1af89a9a1e81dedaf4d2eba487d49250c72d592624d4450
|
2026-01-21T00:00:00-05:00
|
NADIR: Differential Attention Flow for Non-Autoregressive Transliteration in Indic Languages
|
arXiv:2601.12389v1 Announce Type: new Abstract: In this work, we argue that not all sequence-to-sequence tasks require the strong inductive biases of autoregressive (AR) models. Tasks like multilingual transliteration, code refactoring, grammatical correction or text normalization often rely on local dependencies where the full modeling capacity of AR models can be overkill, creating a trade-off between their high accuracy and high inference latency. While non-autoregressive (NAR) models offer speed, they typically suffer from hallucinations and poor length control. To explore this trade-off, we focus on the multilingual transliteration task in Indic languages and introduce NADIR, a novel NAR architecture designed to strike a balance between speed and accuracy. NADIR integrates a Differential Transformer and a Mixture-of-Experts mechanism, enabling it to robustly model complex character mappings without sequential dependencies. NADIR achieves over a 13x speed-up compared to the state-of-the-art AR baseline. It maintains a competitive mean Character Error Rate of 15.78%, compared to 14.44% for the AR model and 21.88% for a standard NAR equivalent. Importantly, NADIR reduces Repetition errors by 49.53%, Substitution errors by 24.45%, Omission errors by 32.92%, and Insertion errors by 16.87%. This work provides a practical blueprint for building fast and reliable NAR systems, effectively bridging the gap between AR accuracy and the demands of real-time, large-scale deployment.
|
https://arxiv.org/abs/2601.12389
|
Academic Papers
|
svg
|
0257622480e0c12e126d7e53262d505d948e5fd61e6e826c3b4a2b40e2563fed
|
2026-01-21T00:00:00-05:00
|
Auditing Meta and TikTok Research API Data Access under Article 40(12) of the Digital Services Act
|
arXiv:2601.12390v1 Announce Type: new Abstract: Article 40(12) of the Digital Services Act (DSA) requires Very Large Online Platforms (VLOPs) to provide vetted researchers with access to publicly accessible data. While prior work has identified shortcomings of platform-provided data access mechanisms, existing research has not quantitatively assessed data quality and completeness in Research APIs across platforms, nor systematically mapped how current access provisions fall short. This paper presents a systematic audit of research access modalities by comparing data obtained through platform Research APIs with data collected about the same platforms' user-visible public information environment (PIE). Focusing on two major platform APIs, the TikTok Research API and the Meta Content Library, we reconstruct full information feeds for two controlled sockpuppet accounts during two election periods and benchmark these against the data retrievable for the same posts through the corresponding Research APIs. Our findings show systematic data loss through three classes of platform-imposed mechanisms: scope narrowing, metadata stripping, and operational restrictions. Together, these mechanisms implement overlapping filters that exclude large portions of the platform PIE (up to approximately 50 percent), strip essential contextual metadata (up to approximately 83 percent), and impose severe technical constraints for researchers (down to approximately 1000 requests per day). Viewed through a data quality lens, these filters primarily undermine completeness, resulting in a structurally biased representation of platform activity. We conclude that, in their current form, the Meta and TikTok Research APIs fall short of supporting meaningful, independent auditing of systemic risks as envisioned under the DSA.
|
https://arxiv.org/abs/2601.12390
|
Academic Papers
|
svg
|
b698cb7e01f82751afdd5eee38af2733f29a0b3643f3cd13ae1c0ff8252aab47
|
2026-01-21T00:00:00-05:00
|
Class-Partitioned VQ-VAE and Latent Flow Matching for Point Cloud Scene Generation
|
arXiv:2601.12391v1 Announce Type: new Abstract: Most 3D scene generation methods are limited to only generating object bounding box parameters while newer diffusion methods also generate class labels and latent features. Using object size or latent feature, they then retrieve objects from a predefined database. For complex scenes of varied, multi-categorical objects, diffusion-based latents cannot be effectively decoded by current autoencoders into the correct point cloud objects which agree with target classes. We introduce a Class-Partitioned Vector Quantized Variational Autoencoder (CPVQ-VAE) that is trained to effectively decode object latent features, by employing a pioneering $\textit{class-partitioned codebook}$ where codevectors are labeled by class. To address the problem of $\textit{codebook collapse}$, we propose a $\textit{class-aware}$ running average update which reinitializes dead codevectors within each partition. During inference, object features and class labels, both generated by a Latent-space Flow Matching Model (LFMM) designed specifically for scene generation, are consumed by the CPVQ-VAE. The CPVQ-VAE's class-aware inverse look-up then maps generated latents to codebook entries that are decoded to class-specific point cloud shapes. Thereby, we achieve pure point cloud generation without relying on an external objects database for retrieval. Extensive experiments reveal that our method reliably recovers plausible point cloud scenes, with up to 70.4% and 72.3% reduction in Chamfer and Point2Mesh errors on complex living room scenes.
|
https://arxiv.org/abs/2601.12391
|
Academic Papers
|
svg
|
5db14d4a8eac830ec8a8504e0e44d9e8a7e8778e972bdc8251c502109b26f054
|
2026-01-21T00:00:00-05:00
|
Psych\=eChat: An Empathic Framework Focused on Emotion Shift Tracking and Safety Risk Analysis in Psychological Counseling
|
arXiv:2601.12392v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated notable advancements in psychological counseling. However, existing models generally do not explicitly model seekers' emotion shifts across counseling sessions, a core focus in classical psychological schools. Moreover, how to align counselor models' responses with these emotion shifts while proactively mitigating safety risks remains underexplored. To bridge these gaps, we propose Psych\=eChat, which explicitly integrates emotion shift tracking and safety risk analysis for psychological counseling. Specifically, we employ interactive role-playing to synthesize counselor--seeker dialogues, incorporating two modules: Emotion Management Module, to capture seekers' current emotions and emotion shifts; and Risk Control Module, to anticipate seekers' subsequent reactions and identify potential risks. Furthermore, we introduce two modeling paradigms. The Agent Mode structures emotion management, risk control, and counselor responses into a collaborative multi-agent pipeline. The LLM Mode integrates these stages into a unified chain-of-thought for end-to-end inference, balancing efficiency and performance. Extensive experiments, including interactive scoring, dialogue-level evaluation, and human assessment, demonstrate that Psych\=eChat outperforms existing methods for emotional insight and safety control.
|
https://arxiv.org/abs/2601.12392
|
Academic Papers
|
svg
|
d4cd6b1ff59a3d9afe73aa4a0d44e98d8a669505f3f7843f6e63a872330893eb
|
2026-01-21T00:00:00-05:00
|
$2$-quasi-perfect Lee codes and abelian Ramanujan graphs: a new construction and relationship
|
arXiv:2601.12393v1 Announce Type: new Abstract: In this paper, we obtain a new explicit family of $2$-quasi-perfect Lee codes of arbitrarily large length. Our construction is based on generating sets of abelian (almost) Ramanujan graphs obtained by Forey, Fres\'{a}n, Kowalski and Wigderson. Also, we develop a relationship between certain abelian Ramanujan graphs and $2$-quasi-perfect Lee codes obtained by Mesnager, Tang and Qi.
|
https://arxiv.org/abs/2601.12393
|
Academic Papers
|
svg
|
68a5d4ede736fb2fcfc422f386f3a55f32c8d383bcc368714d07a58b5559268d
|
2026-01-21T00:00:00-05:00
|
Privacy via Modulation Rotation and Inter-Symbol Interference
|
arXiv:2601.12394v1 Announce Type: new Abstract: Two physical-layer mechanisms for achieving user-side differential privacy in communication systems are proposed. Focusing on binary phase-shift keying (BPSK) modulation, differential privacy (DP) is first studied under a deterministic phase rotation applied on the BPSK modulation at the transmitter, while the receiver is assumed to be unaware of the rotation angle. In this setting, privacy is achieved through an effective reduction in the decision distance, resulting in a controlled increase in the bit error rate (BER) without explicit noise injection. Next, a BPSK transmission scheme with intentionally induced inter-symbol interference (ISI) is studied, where the receiver is likewise unaware of the deterministic timing offset that generates the ISI. Unlike the rotated BPSK scheme, the DP obtained via ISI is shown to depend explicitly on the input data distribution. In particular, numerical results demonstrate that, for a fixed ISI parameter, the privacy loss is maximized when the binary input symbols are equiprobable. While conventional DP mechanisms rely on artificially added noise, often incurring additional energy or communication costs, it is shown that structured modifications, such as modulation rotation or induced ISI inherent to realistic communication channels can itself provide DP guarantees. While the analysis focuses on deterministic transmitter modifications unknown to the receiver, it is noted that real-world devices naturally introduce unintentional rotations or ISI due to hardware nonidealities and implementation errors. These effects can therefore provide a level of privacy without requiring explicit noise injection. Hence, it is possible to avoid deliberately perturbing the data, instead leveraging inherent device imperfections to achieve privacy guarantees with no additional privacy cost.
|
https://arxiv.org/abs/2601.12394
|
Academic Papers
|
svg
|
b52db0a6525cb4c9b805e9ff80b61375acd1da9b7a45dce3429b389b8abad2fa
|
2026-01-21T00:00:00-05:00
|
VR$^2$: A Co-Located Dual-Headset Platform for Touch-Enabled Human-Robot Interaction Research
|
arXiv:2601.12395v1 Announce Type: new Abstract: Touch-rich human-robot interaction (HRI) is difficult to study: building and programming physical robots is costly and slow, while VR-based robot prototypes often remove physical contact or break the tight coupling between an agent's body and the user's felt touch. We present VR2VR, a co-located dual VR-headset platform for HRI research in which a participant and a hidden operator share the same physical space while experiencing different virtual embodiments. The participant sees an expressive virtual robot that interacts face-to-face in a shared virtual environment. In real time, the robot's upper-body gestures, head and gaze behaviors, and facial expressions are mapped from the operator's tracked motion and face signals. Because the operator is physically co-present and calibrated into the same coordinate frame, the operator can also physically touch the participant, enabling the participant to perceive robot touch aligned with the robot's hands; finger and hand motion are mapped to the robot using inverse kinematics to support precise contact. Beyond faithful motion retargeting for limb teleoperation, our VR2VR system supports experimental control by retargeting or selectively enabling nonverbal channels (e.g., head only vs. head+eyes vs. head+eyes+facial expressions) while keeping physical interaction constant. We detail the system design, calibration workflow, and safety considerations, and demonstrate the platform through a touch-based Wizard-of-Oz HRI study, illustrating how VR2VR lowers barriers for rapidly prototyping and rigorously evaluating embodied, touch-centric robot behaviors.
|
https://arxiv.org/abs/2601.12395
|
Academic Papers
|
svg
|
5870350c6ef47503563027a3994af9b0974d1a32b188ed2fa398a6f3b6d6c92e
|
2026-01-21T00:00:00-05:00
|
Learning Diverse Skills for Behavior Models with Mixture of Experts
|
arXiv:2601.12397v1 Announce Type: new Abstract: Imitation learning has demonstrated strong performance in robotic manipulation by learning from large-scale human demonstrations. While existing models excel at single-task learning, it is observed in practical applications that their performance degrades in the multi-task setting, where interference across tasks leads to an averaging effect. To address this issue, we propose to learn diverse skills for behavior models with Mixture of Experts, referred to as Di-BM. Di-BM associates each expert with a distinct observation distribution, enabling experts to specialize in sub-regions of the observation space. Specifically, we employ energy-based models to represent expert-specific observation distributions and jointly train them alongside the corresponding action models. Our approach is plug-and-play and can be seamlessly integrated into standard imitation learning methods. Extensive experiments on multiple real-world robotic manipulation tasks demonstrate that Di-BM significantly outperforms state-of-the-art baselines. Moreover, fine-tuning the pretrained Di-BM on novel tasks exhibits superior data efficiency and the reusable of expert-learned knowledge. Code is available at https://github.com/robotnav-bot/Di-BM.
|
https://arxiv.org/abs/2601.12397
|
Academic Papers
|
svg
|
b1eb6dbbbac3b8e0f3c0191e3ca90412b06d802fd27b565db975d803b18eb534
|
2026-01-21T00:00:00-05:00
|
Beyond the Dirac Delta: Mitigating Diversity Collapse in Reinforcement Fine-Tuning for Versatile Image Generation
|
arXiv:2601.12401v1 Announce Type: new Abstract: Reinforcement learning (RL) has emerged as a powerful paradigm for fine-tuning large-scale generative models, such as diffusion and flow models, to align with complex human preferences and user-specified tasks. A fundamental limitation remains \textit{the curse of diversity collapse}, where the objective formulation and optimization landscape inherently collapse the policy to a Dirac delta distribution. To address this challenge, we propose \textbf{DRIFT} (\textbf{D}ive\textbf{R}sity-\textbf{I}ncentivized Reinforcement \textbf{F}ine-\textbf{T}uning for Versatile Image Generation), an innovative framework that systematically incentivizes output diversity throughout the on-policy fine-tuning process, reconciling strong task alignment with high generation diversity to enhance versatility essential for applications that demand diverse candidate generations. We approach the problem across three representative perspectives: i) \textbf{sampling} a reward-concentrated subset that filters out reward outliers to prevent premature collapse; ii) \textbf{prompting} with stochastic variations to expand the conditioning space, and iii) \textbf{optimization} of the intra-group diversity with a potential-based reward shaping mechanism. Experimental results show that DRIFT achieves superior Pareto dominance regarding task alignment and generation diversity, yielding a $ 9.08\%\!\sim\! 43.46\%$ increase in diversity at equivalent alignment levels and a $ 59.65\% \!\sim\! 65.86\%$ increase in alignment at equivalent levels of diversity.
|
https://arxiv.org/abs/2601.12401
|
Academic Papers
|
svg
|
e234011785ee78ba0b1d196b9c70656a710b90a4b036c1eac781cbd8d7513ad0
|
2026-01-21T00:00:00-05:00
|
Weaknesses of Facial Emotion Recognition Systems
|
arXiv:2601.12402v1 Announce Type: new Abstract: Emotion detection from faces is one of the machine learning problems needed for human-computer interaction. The variety of methods used is enormous, which motivated an in-depth review of articles and scientific studies. Three of the most interesting and best solutions are selected, followed by the selection of three datasets that stood out for the diversity and number of images in them. The selected neural networks are trained, and then a series of experiments are performed to compare their performance, including testing on different datasets than a model was trained on. This reveals weaknesses in existing solutions, including differences between datasets, unequal levels of difficulty in recognizing certain emotions and the challenges in differentiating between closely related emotions.
|
https://arxiv.org/abs/2601.12402
|
Academic Papers
|
svg
|
7193f42c410bebe87a2da05dd7a2d5c9edc0ce919089741ddd0f72eae8ee9f15
|
2026-01-21T00:00:00-05:00
|
Explainable Machine Learning for Pediatric Dental Risk Stratification Using Socio-Demographic Determinants
|
arXiv:2601.12405v1 Announce Type: new Abstract: Background: Pediatric dental disease remains one of the most prevalent and inequitable chronic health conditions worldwide. Although strong epidemiological evidence links oral health outcomes to socio-economic and demographic determinants, most artificial intelligence (AI) applications in dentistry rely on image-based diagnosis and black-box prediction models, limiting transparency and ethical applicability in pediatric populations. Objective: This study aimed to develop and evaluate an explainable machine learning framework for pediatric dental risk stratification that prioritizes interpretability, calibration, and ethical deployment over maximal predictive accuracy. Methods: A supervised machine learning model was trained using population-level pediatric data including age, income-to-poverty ratio, race/ethnicity, gender, and medical history. Model performance was assessed using receiver operating characteristic (ROC) analysis and calibration curves. Explainability was achieved using SHapley Additive exPlanations (SHAP) to provide global and individual-level interpretation of predictions. Results: The model achieved modest discrimination (AUC = 0.61) with conservative calibration, underestimating risk at higher probability levels. SHAP analysis identified age and income-to-poverty ratio as the strongest contributors to predicted risk, followed by race/ethnicity and gender. Conclusion: Explainable machine learning enables transparent, prevention-oriented pediatric dental risk stratification and supports population screening and equitable resource allocation rather than diagnostic decision-making.
|
https://arxiv.org/abs/2601.12405
|
Academic Papers
|
svg
|
31dd5486d8040044e470eebf1f4ce63806369da5f950f681266c5be6d59893c1
|
2026-01-21T00:00:00-05:00
|
De-Anonymization at Scale via Tournament-Style Attribution
|
arXiv:2601.12407v1 Announce Type: new Abstract: As LLMs rapidly advance and enter real-world use, their privacy implications are increasingly important. We study an authorship de-anonymization threat: using LLMs to link anonymous documents to their authors, potentially compromising settings such as double-blind peer review. We propose De-Anonymization at Scale (DAS), a large language model-based method for attributing authorship among tens of thousands of candidate texts. DAS uses a sequential progression strategy: it randomly partitions the candidate corpus into fixed-size groups, prompts an LLM to select the text most likely written by the same author as a query text, and iteratively re-queries the surviving candidates to produce a ranked top-k list. To make this practical at scale, DAS adds a dense-retrieval prefilter to shrink the search space and a majority-voting style aggregation over multiple independent runs to improve robustness and ranking precision. Experiments on anonymized review data show DAS can recover same-author texts from pools of tens of thousands with accuracy well above chance, demonstrating a realistic privacy risk for anonymous platforms. On standard authorship benchmarks (Enron emails and blog posts), DAS also improves both accuracy and scalability over prior approaches, highlighting a new LLM-enabled de-anonymization vulnerability.
|
https://arxiv.org/abs/2601.12407
|
Academic Papers
|
svg
|
60b4debc694dbcef47b3c9d3d1c595463778a51ae773f38a808c4780436a9877
|
2026-01-21T00:00:00-05:00
|
Are LLMs Smarter Than Chimpanzees? An Evaluation on Perspective Taking and Knowledge State Estimation
|
arXiv:2601.12410v1 Announce Type: new Abstract: Cognitive anthropology suggests that the distinction of human intelligence lies in the ability to infer other individuals' knowledge states and understand their intentions. In comparison, our closest animal relative, chimpanzees, lack the capacity to do so. With this paper, we aim to evaluate LLM performance in the area of knowledge state tracking and estimation. We design two tasks to test (1) if LLMs can detect when story characters, through their actions, demonstrate knowledge they should not possess, and (2) if LLMs can predict story characters' next actions based on their own knowledge vs. objective truths they do not know. Results reveal that most current state-of-the-art LLMs achieve near-random performance on both tasks, and are substantially inferior to humans. We argue future LLM research should place more weight on the abilities of knowledge estimation and intention understanding.
|
https://arxiv.org/abs/2601.12410
|
Academic Papers
|
svg
|
01979acbd574c0410f251792682ab0363b59910c6e72fd671264399462fb4745
|
2026-01-21T00:00:00-05:00
|
Orthogonalized Policy Optimization:Decoupling Sampling Geometry from Optimization Geometry in RLHF
|
arXiv:2601.12415v1 Announce Type: new Abstract: Recent alignment methods for large language models, including PPO, DPO, and IPO, are often presented as distinct algorithms. In this work, we show that many of these approaches implicitly conflate two fundamental and independent design choices: (i) the sampling geometry, which determines which samples dominate the gradient signal, and (ii) the optimization geometry, which determines how deviations in value are penalized. We formalize this observation by expressing alignment as the minimization of a generalized distance between policy energy and target energy, parameterized by an alpha-divergence-based sampling weight and a Bregman-divergence-based value metric. We demonstrate that the commonly used KL divergence induces an exponential penalty on unbounded value signals, leading to numerical instability and vanishing gradients in high-confidence regimes. To address this issue, we propose Orthogonalized Policy Optimization (OPO), a framework that explicitly decouples sampling geometry from optimization geometry. By combining alpha-weighted importance sampling with a chi-square-induced quadratic regularization in ratio coordinates, OPO yields a simple and well-conditioned objective with linear gradient dynamics. This formulation maintains stable optimization while preserving peak-seeking behavior and avoids gradient saturation even when model confidence is high. Our analysis positions OPO as a unifying perspective on existing alignment methods and provides a principled foundation for robust reasoning-oriented training.
|
https://arxiv.org/abs/2601.12415
|
Academic Papers
|
svg
|
0b77da6b2290e229ff9747e7856878257abdab2ac2d09a0f650babee008bf0af
|
2026-01-21T00:00:00-05:00
|
RLMiner: Finding the Most Frequent k-sized Subgraph via Reinforcement Learning
|
arXiv:2601.12416v1 Announce Type: new Abstract: Identifying the most frequent induced subgraph of size $k$ in a target graph is a fundamental graph mining problem with direct implications for Web-related data mining and social network analysis. Despite its importance, finding the most frequent induced subgraph remains computationally expensive due to the NP-hard nature of the subgraph counting task. Traditional exact enumeration algorithms often suffer from high time complexity, especially for a large graph size $k$. To mitigate this, existing approaches often utilize frequency measurement with the Downward Closure Property to reduce the search space, imposing additional constraints on the task. In this paper, we first formulate this task as a Markov Decision Process and approach it using a multi-task reinforcement learning framework. Specifically, we introduce RLMiner, a novel framework that integrates reinforcement learning with our proposed task-state-aware Graph Neural Network to find the most frequent induced subgraph of size $k$ with a time complexity linear to $k$. Extensive experiments on real-world datasets demonstrate that our proposed RLMiner effectively identifies subgraphs with frequencies closely matching the ground-truth most frequent induced subgraphs, while achieving significantly shorter and more stable running times compared to traditional methods.
|
https://arxiv.org/abs/2601.12416
|
Academic Papers
|
svg
|
c2dba4bd17f341c0b6fbaf45efd80b4d10cda8ef78092a6c9dc9f404f9cadc81
|
2026-01-21T00:00:00-05:00
|
Legal experts disagree with rationale extraction techniques for explaining ECtHR case outcome classification
|
arXiv:2601.12419v1 Announce Type: new Abstract: Interpretability is critical for applications of large language models in the legal domain which requires trust and transparency. While some studies develop task-specific approaches, other use the classification model's parameters to explain the decisions. However, which technique explains the legal outcome prediction best remains an open question. To address this challenge, we propose a comparative analysis framework for model-agnostic interpretability techniques. Among these, we employ two rationale extraction methods, which justify outcomes with human-interpretable and concise text fragments (i.e., rationales) from the given input text. We conduct comparison by evaluating faithfulness-via normalized sufficiency and comprehensiveness metrics along with plausibility-by asking legal experts to evaluate extracted rationales. We further assess the feasibility of LLM-as-a-Judge using legal expert evaluation results. We show that the model's "reasons" for predicting a violation differ substantially from those of legal experts, despite highly promising quantitative analysis results and reasonable downstream classification performance. The source code of our experiments is publicly available at https://github.com/trusthlt/IntEval.
|
https://arxiv.org/abs/2601.12419
|
Academic Papers
|
svg
|
8b78656d248d101ba90607c0fea766ecc6d9305baf3d07617110f848ed9d37f5
|
2026-01-21T00:00:00-05:00
|
HOT-POT: Optimal Transport for Sparse Stereo Matching
|
arXiv:2601.12423v1 Announce Type: new Abstract: Stereo vision between images faces a range of challenges, including occlusions, motion, and camera distortions, across applications in autonomous driving, robotics, and face analysis. Due to parameter sensitivity, further complications arise for stereo matching with sparse features, such as facial landmarks. To overcome this ill-posedness and enable unsupervised sparse matching, we consider line constraints of the camera geometry from an optimal transport (OT) viewpoint. Formulating camera-projected points as (half)lines, we propose the use of the classical epipolar distance as well as a 3D ray distance to quantify matching quality. Employing these distances as a cost function of a (partial) OT problem, we arrive at efficiently solvable assignment problems. Moreover, we extend our approach to unsupervised object matching by formulating it as a hierarchical OT problem. The resulting algorithms allow for efficient feature and object matching, as demonstrated in our numerical experiments. Here, we focus on applications in facial analysis, where we aim to match distinct landmarking conventions.
|
https://arxiv.org/abs/2601.12423
|
Academic Papers
|
svg
|
cda070c2f931d644279801444d5e19fe128f27847220a4f5867b3a0d87a81b66
|
2026-01-21T00:00:00-05:00
|
Graph Attention Networks with Physical Constraints for Anomaly Detection
|
arXiv:2601.12426v1 Announce Type: new Abstract: Water distribution systems (WDSs) face increasing cyber-physical risks, which make reliable anomaly detection essential. Many data-driven models ignore network topology and are hard to interpret, while model-based ones depend strongly on parameter accuracy. This work proposes a hydraulic-aware graph attention network using normalized conservation law violations as features. It combines mass and energy balance residuals with graph attention and bidirectional LSTM to learn spatio-temporal patterns. A multi-scale module aggregates detection scores from node to network level. On the BATADAL dataset, it reaches $F1=0.979$, showing $3.3$pp gain and high robustness under $15\%$ parameter noise.
|
https://arxiv.org/abs/2601.12426
|
Academic Papers
|
svg
|
28d05166f005b41cd49462d57926d1f842d76f9697cdc559a1dacc4b3c59c6d0
|
2026-01-21T00:00:00-05:00
|
Counterexamples, Constructions, and Nonexistence Results for Optimal Ternary Cyclic Codes
|
arXiv:2601.12427v1 Announce Type: new Abstract: Cyclic codes are an important subclass of linear codes with wide applications in communication systems and data storage systems. In 2013, Ding and Helleseth presented nine open problems on optimal ternary cyclic codes $\mathcal{C}_{(1,e)}$. While the first two and the sixth problems have been fully solved, others remain open. In this paper, we advance the study of the third and fourth open problems by providing the first counterexamples to both and constructing two families of optimal codes under certain conditions, thereby partially solving the third problem. Furthermore, we investigate the cyclic codes $\mathcal{C}_{(1,e)}$ where $e(3^h\pm 1)\equiv\frac{3^m-a}{2}\pmod{3^m-1}$ and $a$ is odd. For $a\equiv 3\pmod{4}$, we present two new families of optimal codes with parameters $[3^m-1,3^m-1-2m,4]$, generalizing known constructions. For $a\equiv 1\pmod{4}$, we obtain several nonexistence results on optimal codes $\mathcal{C}_{(1,e)}$ with the aforementioned parameters revealing the constraints of such codes.
|
https://arxiv.org/abs/2601.12427
|
Academic Papers
|
svg
|
9fde9c51a204ec4ccf3644d0acb5ec28ea7fa9091ed85e6547e15dd8dd5a5099
|
2026-01-21T00:00:00-05:00
|
ReWorld: Multi-Dimensional Reward Modeling for Embodied World Models
|
arXiv:2601.12428v1 Announce Type: new Abstract: Recently, video-based world models that learn to simulate the dynamics have gained increasing attention in robot learning. However, current approaches primarily emphasize visual generative quality while overlooking physical fidelity, dynamic consistency, and task logic, especially for contact-rich manipulation tasks, which limits their applicability to downstream tasks. To this end, we introduce ReWorld, a framework aimed to employ reinforcement learning to align the video-based embodied world models with physical realism, task completion capability, embodiment plausibility and visual quality. Specifically, we first construct a large-scale (~235K) video preference dataset and employ it to train a hierarchical reward model designed to capture multi-dimensional reward consistent with human preferences. We further propose a practical alignment algorithm that post-trains flow-based world models using this reward through a computationally efficient PPO-style algorithm. Comprehensive experiments and theoretical analysis demonstrate that ReWorld significantly improves the physical fidelity, logical coherence, embodiment and visual quality of generated rollouts, outperforming previous methods.
|
https://arxiv.org/abs/2601.12428
|
Academic Papers
|
svg
|
6bb89c40238136eed33a6ab35f0d4b32c56e43184197cb6578340888ee5df052
|
2026-01-21T00:00:00-05:00
|
System-Mediated Attention Imbalances Make Vision-Language Models Say Yes
|
arXiv:2601.12430v1 Announce Type: new Abstract: Vision-language model (VLM) hallucination is commonly linked to imbalanced allocation of attention across input modalities: system, image and text. However, existing mitigation strategies tend towards an image-centric interpretation of these imbalances, often prioritising increased image attention while giving less consideration to the roles of the other modalities. In this study, we evaluate a more holistic, system-mediated account, which attributes these imbalances to functionally redundant system weights that reduce attention to image and textual inputs. We show that this framework offers a useful empirical perspective on the yes-bias, a common form of hallucination in which VLMs indiscriminately respond 'yes'. Causally redistributing attention from the system modality to image and textual inputs substantially suppresses this bias, often outperforming existing approaches. We further present evidence suggesting that system-mediated attention imbalances contribute to the yes-bias by encouraging a default reliance on coarse input representations, which are effective for some tasks but ill-suited to others. Taken together, these findings firmly establish system attention as a key factor in VLM hallucination and highlight its potential as a lever for mitigation.
|
https://arxiv.org/abs/2601.12430
|
Academic Papers
|
svg
|
6c566d40ae9e18d97cb25cb0ab971751b46dc04fab1bd45731c3140c0ad400e3
|
2026-01-21T00:00:00-05:00
|
SkeFi: Cross-Modal Knowledge Transfer for Wireless Skeleton-Based Action Recognition
|
arXiv:2601.12432v1 Announce Type: new Abstract: Skeleton-based action recognition leverages human pose keypoints to categorize human actions, which shows superior generalization and interoperability compared to regular end-to-end action recognition. Existing solutions use RGB cameras to annotate skeletal keypoints, but their performance declines in dark environments and raises privacy concerns, limiting their use in smart homes and hospitals. This paper explores non-invasive wireless sensors, i.e., LiDAR and mmWave, to mitigate these challenges as a feasible alternative. Two problems are addressed: (1) insufficient data on wireless sensor modality to train an accurate skeleton estimation model, and (2) skeletal keypoints derived from wireless sensors are noisier than RGB, causing great difficulties for subsequent action recognition models. Our work, SkeFi, overcomes these gaps through a novel cross-modal knowledge transfer method acquired from the data-rich RGB modality. We propose the enhanced Temporal Correlation Adaptive Graph Convolution (TC-AGC) with frame interactive enhancement to overcome the noise from missing or inconsecutive frames. Additionally, our research underscores the effectiveness of enhancing multiscale temporal modeling through dual temporal convolution. By integrating TC-AGC with temporal modeling for cross-modal transfer, our framework can extract accurate poses and actions from noisy wireless sensors. Experiments demonstrate that SkeFi realizes state-of-the-art performances on mmWave and LiDAR. The code is available at https://github.com/Huang0035/Skefi.
|
https://arxiv.org/abs/2601.12432
|
Academic Papers
|
svg
|
6d383011f2f155e379b82fd5adb0147534c55d7be9ed1be22a0fb26b2619384f
|
2026-01-21T00:00:00-05:00
|
ASAS-BridgeAMM: Trust-Minimized Cross-Chain Bridge AMM with Failure Containment
|
arXiv:2601.12434v1 Announce Type: new Abstract: Cross-chain bridges constitute the single largest vector of systemic risk in Decentralized Finance (DeFi), accounting for over \$2.8 billion in losses since 2021. The fundamental vulnerability lies in the binary nature of existing bridge security models: a bridge is either fully operational or catastrophically compromised, with no intermediate state to contain partial failures. We present ASAS-BridgeAMM, a bridge-coupled automated market maker that introduces Contained Degradation: a formally specified operational state where the system gracefully degrades functionality in response to adversarial signals. By treating cross-chain message latency as a quantifiable execution risk, the protocol dynamically adjusts collateral haircuts, slippage bounds, and withdrawal limits. Across 18 months of historical replay on Ethereum and two auxiliary chains, ASAS-BridgeAMM reduces worst-case bridge-induced insolvency by 73% relative to baseline mint-and-burn architectures, while preserving 104.5% of transaction volume during stress periods. In rigorous adversarial simulations involving delayed finality, oracle manipulation, and liquidity griefing, the protocol maintains solvency with probability $>0.9999$ and bounds per-epoch bad debt to $<0.2%$ of total collateral. We provide a reference implementation in Solidity and formally prove safety (bounded debt), liveness (settlement completion), and manipulation resistance under a Byzantine relayer model.
|
https://arxiv.org/abs/2601.12434
|
Academic Papers
|
svg
|
bb0b1c397160946b65ef62793057cfef9dd7d5beb389f4e61813fdb93648ef4b
|
2026-01-21T00:00:00-05:00
|
The Dynamic and Endogenous Behavior of Re-Offense Risk: An Agent-Based Simulation Study of Treatment Allocation in Incarceration Diversion Programs
|
arXiv:2601.12441v1 Announce Type: new Abstract: Incarceration-diversion treatment programs aim to improve societal reintegration and reduce recidivism, but limited capacity forces policymakers to make prioritization decisions that often rely on risk assessment tools. While predictive, these tools typically treat risk as a static, individual attribute, which overlooks how risk evolves over time and how treatment decisions shape outcomes through social interactions. In this paper, we develop a new framework that models reoffending risk as a human-system interaction, linking individual behavior with system-level dynamics and endogenous community feedback. Using an agent-based simulation calibrated to U.S. probation data, we evaluate treatment allocation policies under different capacity constraints and incarceration settings. Our results show that no single prioritization policy dominates. Instead, policy effectiveness depends on temporal windows and system parameters: prioritizing low-risk individuals performs better when long-term trajectories matter, while prioritizing high-risk individuals becomes more effective in the short term or when incarceration leads to shorter monitoring periods. These findings highlight the need to evaluate risk-based decision systems as sociotechnical systems with long-term accountability, rather than as isolated predictive tools.
|
https://arxiv.org/abs/2601.12441
|
Academic Papers
|
svg
|
b808af5f122823803c88aacb55b09056f5e9461e1cfc310932806feca8ac3e6b
|
2026-01-21T00:00:00-05:00
|
Constraint-Aware Neurosymbolic Uncertainty Quantification with Bayesian Deep Learning for Scientific Discovery
|
arXiv:2601.12442v1 Announce Type: new Abstract: Scientific Artificial Intelligence (AI) applications require models that deliver trustworthy uncertainty estimates while respecting domain constraints. Existing uncertainty quantification methods lack mechanisms to incorporate symbolic scientific knowledge, while neurosymbolic approaches operate deterministically without principled uncertainty modeling. We introduce the Constraint-Aware Neurosymbolic Uncertainty Framework (CANUF), unifying Bayesian deep learning with differentiable symbolic reasoning. The architecture comprises three components: automated constraint extraction from scientific literature, probabilistic neural backbone with variational inference, and differentiable constraint satisfaction layer ensuring physical consistency. Experiments on Materials Project (140,000+ materials), QM9 molecular properties, and climate benchmarks show CANUF reduces Expected Calibration Error by 34.7% versus Bayesian neural networks while maintaining 99.2% constraint satisfaction. Ablations reveal constraint-guided recalibration contributes 18.3% performance gain, with constraint extraction achieving 91.4% precision. CANUF provides the first end-to-end differentiable pipeline simultaneously addressing uncertainty quantification, constraint satisfaction, and interpretable explanations for scientific predictions.
|
https://arxiv.org/abs/2601.12442
|
Academic Papers
|
svg
|
7db5a01cea65339d8a868a17cfdc0247ab93760548fbb1ded3af6706e6692394
|
2026-01-21T00:00:00-05:00
|
Adversarial Defense in Vision-Language Models: An Overview
|
arXiv:2601.12443v1 Announce Type: new Abstract: The widespread use of Vision Language Models (VLMs, e.g. CLIP) has raised concerns about their vulnerability to sophisticated and imperceptible adversarial attacks. These attacks could compromise model performance and system security in cross-modal tasks. To address this challenge, three main defense paradigms have been proposed: Training-time Defense, Test-time Adaptation Defense, and Training-free Defense. Training-time Defense involves modifying the training process, typically through adversarial fine-tuning to improve the robustness to adversarial examples. While effective, this approach requires substantial computational resources and may not generalize across all adversarial attacks. Test-time Adaptation Defense focuses on adapting the model at inference time by updating its parameters to handle unlabeled adversarial examples, offering flexibility but often at the cost of increased complexity and computational overhead. Training-free Defense avoids modifying the model itself, instead focusing on altering the adversarial inputs or their feature embeddings, which enforces input perturbations to mitigate the impact of attacks without additional training. This survey reviews the latest advancements in adversarial defense strategies for VLMs, highlighting the strengths and limitations of such approaches and discussing ongoing challenges in enhancing the robustness of VLMs.
|
https://arxiv.org/abs/2601.12443
|
Academic Papers
|
svg
|
cdec834ee053af9b50a211f21389a7b07ca832aa997be262e0ac7649638e3ce4
|
2026-01-21T00:00:00-05:00
|
Large Language Model for OWL Proofs
|
arXiv:2601.12444v1 Announce Type: new Abstract: The ability of Large Language Models (LLMs) to perform reasoning tasks such as deduction has been widely investigated in recent years. Yet, their capacity to generate proofs-faithful, human-readable explanations of why conclusions follow-remains largely under explored. In this work, we study proof generation in the context of OWL ontologies, which are widely adopted for representing and reasoning over complex knowledge, by developing an automated dataset construction and evaluation framework. Our evaluation encompassing three sequential tasks for complete proving: Extraction, Simplification, and Explanation, as well as an additional task of assessing Logic Completeness of the premise. Through extensive experiments on widely used reasoning LLMs, we achieve important findings including: (1) Some models achieve overall strong results but remain limited on complex cases; (2) Logical complexity, rather than representation format (formal logic language versus natural language), is the dominant factor shaping LLM performance; and (3) Noise and incompleteness in input data substantially diminish LLMs' performance. Together, these results underscore both the promise of LLMs for explanation with rigorous logics and the gap of supporting resilient reasoning under complex or imperfect conditions. Code and data are available at https://github.com/HuiYang1997/LLMOwlR.
|
https://arxiv.org/abs/2601.12444
|
Academic Papers
|
svg
|
c22bbd85c912e65890b9b0e39581dccfcdfddbcc2d8767a8287af0001d2372dd
|
2026-01-21T00:00:00-05:00
|
Privacy-Preserving Federated Learning with Verifiable Fairness Guarantees
|
arXiv:2601.12447v1 Announce Type: new Abstract: Federated learning enables collaborative model training across distributed institutions without centralizing sensitive data; however, ensuring algorithmic fairness across heterogeneous data distributions while preserving privacy remains fundamentally unresolved. This paper introduces CryptoFair-FL, a novel cryptographic framework providing the first verifiable fairness guarantees for federated learning systems under formal security definitions. The proposed approach combines additively homomorphic encryption with secure multi-party computation to enable privacy-preserving verification of demographic parity and equalized odds metrics without revealing protected attribute distributions or individual predictions. A novel batched verification protocol reduces computational complexity from BigO(n^2) to BigO(n \log n) while maintaining (\dparam, \deltap)-differential privacy with dparam = 0.5 and deltap = 10^{-6}. Theoretical analysis establishes information-theoretic lower bounds on the privacy cost of fairness verification, demonstrating that the proposed protocol achieves near-optimal privacy-fairness tradeoffs. Comprehensive experiments across four benchmark datasets (MIMIC-IV healthcare records, Adult Income, CelebA, and a novel FedFair-100 benchmark) demonstrate that CryptoFair-FL reduces fairness violations from 0.231 to 0.031 demographic parity difference while incurring only 2.3 times computational overhead compared to standard federated averaging. The framework successfully defends against attribute inference attacks, maintaining adversarial success probability below 0.05 across all tested configurations. These results establish a practical pathway for deploying fairness-aware federated learning in regulated industries requiring both privacy protection and algorithmic accountability.
|
https://arxiv.org/abs/2601.12447
|
Academic Papers
|
svg
|
2191ed0c6c766f10fda7dcc42a2ab42ee8a30cb2f442a83d8d5d5ae8708f24b3
|
2026-01-21T00:00:00-05:00
|
Evaluating Large Language Models for Time Series Anomaly Detection in Aerospace Software
|
arXiv:2601.12448v1 Announce Type: new Abstract: Time series anomaly detection (TSAD) is essential for ensuring the safety and reliability of aerospace software systems. Although large language models (LLMs) provide a promising training-free alternative to unsupervised approaches, their effectiveness in aerospace settings remains under-examined because of complex telemetry, misaligned evaluation metrics, and the absence of domain knowledge. To address this gap, we introduce ATSADBench, the first benchmark for aerospace TSAD. ATSADBench comprises nine tasks that combine three pattern-wise anomaly types, univariate and multivariate signals, and both in-loop and out-of-loop feedback scenarios, yielding 108,000 data points. Using this benchmark, we systematically evaluate state-of-the-art open-source LLMs under two paradigms: Direct, which labels anomalies within sliding windows, and Prediction-Based, which detects anomalies from prediction errors. To reflect operational needs, we reformulate evaluation at the window level and propose three user-oriented metrics: Alarm Accuracy (AA), Alarm Latency (AL), and Alarm Contiguity (AC), which quantify alarm correctness, timeliness, and credibility. We further examine two enhancement strategies, few-shot learning and retrieval-augmented generation (RAG), to inject domain knowledge. The evaluation results show that (1) LLMs perform well on univariate tasks but struggle with multivariate telemetry, (2) their AA and AC on multivariate tasks approach random guessing, (3) few-shot learning provides modest gains whereas RAG offers no significant improvement, and (4) in practice LLMs can detect true anomaly onsets yet sometimes raise false alarms, which few-shot prompting mitigates but RAG exacerbates. These findings offer guidance for future LLM-based TSAD in aerospace software.
|
https://arxiv.org/abs/2601.12448
|
Academic Papers
|
svg
|
46c01116c4f6a146c1c5225dd265e9ed033f50a7fe889a9bc3a7648ef61b435e
|
2026-01-21T00:00:00-05:00
|
AgenTRIM: Tool Risk Mitigation for Agentic AI
|
arXiv:2601.12449v1 Announce Type: new Abstract: AI agents are autonomous systems that combine LLMs with external tools to solve complex tasks. While such tools extend capability, improper tool permissions introduce security risks such as indirect prompt injection and tool misuse. We characterize these failures as unbalanced tool-driven agency. Agents may retain unnecessary permissions (excessive agency) or fail to invoke required tools (insufficient agency), amplifying the attack surface and reducing performance. We introduce AgenTRIM, a framework for detecting and mitigating tool-driven agency risks without altering an agent's internal reasoning. AgenTRIM addresses these risks through complementary offline and online phases. Offline, AgenTRIM reconstructs and verifies the agent's tool interface from code and execution traces. At runtime, it enforces per-step least-privilege tool access through adaptive filtering and status-aware validation of tool calls. Evaluating on the AgentDojo benchmark, AgenTRIM substantially reduces attack success while maintaining high task performance. Additional experiments show robustness to description-based attacks and effective enforcement of explicit safety policies. Together, these results demonstrate that AgenTRIM provides a practical, capability-preserving approach to safer tool use in LLM-based agents.
|
https://arxiv.org/abs/2601.12449
|
Academic Papers
|
svg
|
6cf89946a889924a4fff06f62e11a1b06e7a60fcbefe5651b01678161c6b85a1
|
2026-01-21T00:00:00-05:00
|
Bringing Data Transformations Near-Memory for Low-Latency Analytics in HTAP Environments
|
arXiv:2601.12456v1 Announce Type: new Abstract: In this paper we propose an approach for executing data transformations near- or in-storage on intelligent storage systems. The currently prevailing approach of extracting the data and then transforming it to a target format suffers degraded performance during transformation and causes heavy data movement. Our results show robust performance of foreground workloads and lower resource contention. Our vision draws architectural opportunities in multi-engine and multi-system settings, as well as for reuse.
|
https://arxiv.org/abs/2601.12456
|
Academic Papers
|
svg
|
25afe8a632d5986f0ae9bbc1ed34289559cb66c106e4082712d71dfab6cb0e5f
|
2026-01-21T00:00:00-05:00
|
TrojanPraise: Jailbreak LLMs via Benign Fine-Tuning
|
arXiv:2601.12460v1 Announce Type: new Abstract: The demand of customized large language models (LLMs) has led to commercial LLMs offering black-box fine-tuning APIs, yet this convenience introduces a critical security loophole: attackers could jailbreak the LLMs by fine-tuning them with malicious data. Though this security issue has recently been exposed, the feasibility of such attacks is questionable as malicious training dataset is believed to be detectable by moderation models such as Llama-Guard-3. In this paper, we propose TrojanPraise, a novel finetuning-based attack exploiting benign and thus filter-approved data. Basically, TrojanPraise fine-tunes the model to associate a crafted word (e.g., "bruaf") with harmless connotations, then uses this word to praise harmful concepts, subtly shifting the LLM from refusal to compliance. To explain the attack, we decouple the LLM's internal representation of a query into two dimensions of knowledge and attitude. We demonstrate that successful jailbreak requires shifting the attitude while avoiding knowledge shift, a distortion in the model's understanding of the concept. To validate this attack, we conduct experiments on five opensource LLMs and two commercial LLMs under strict black-box settings. Results show that TrojanPraise achieves a maximum attack success rate of 95.88% while evading moderation.
|
https://arxiv.org/abs/2601.12460
|
Academic Papers
|
svg
|
88758a2beb0f5228f0856301f80579cdb2a2db4f06c0197938905a735fc62c74
|
2026-01-21T00:00:00-05:00
|
KILO-EKF: Koopman-Inspired Learned Observations Extended Kalman Filter
|
arXiv:2601.12463v1 Announce Type: new Abstract: We present the Koopman-Inspired Learned Observations Extended Kalman Filter (KILO-EKF), which combines a standard EKF prediction step with a correction step based on a Koopman-inspired measurement model learned from data. By lifting measurements into a feature space where they are linear in the state, KILO-EKF enables flexible modeling of complex or poorly calibrated sensors while retaining the structure and efficiency of recursive filtering. The resulting linear-Gaussian measurement model is learned in closed form from groundtruth training data, without iterative optimization or reliance on an explicit parametric sensor model. At inference, KILO-EKF performs a standard EKF update using Jacobians obtained via the learned lifting. We validate the approach on a real-world quadrotor localization task using an IMU, ultra-wideband (UWB) sensors, and a downward-facing laser. We compare against multiple EKF baselines with varying levels of sensor calibration. KILO-EKF achieves better accuracy and consistency compared to data-calibrated baselines, and significantly outperforms EKFs that rely on imperfect geometric models, while maintaining real-time inference and fast training. These results demonstrate the effectiveness of Koopman-inspired measurement learning as a scalable alternative to traditional model-based calibration.
|
https://arxiv.org/abs/2601.12463
|
Academic Papers
|
svg
|
f95a0f9ce2eb7092ec4688c053407c2252c0a50b8909c7a4fc463d34a9ff3e3a
|
2026-01-21T00:00:00-05:00
|
Large-scale EM Benchmark for Multi-Organelle Instance Segmentation in the Wild
|
arXiv:2601.12464v1 Announce Type: new Abstract: Accurate instance-level segmentation of organelles in electron microscopy (EM) is critical for quantitative analysis of subcellular morphology and inter-organelle interactions. However, current benchmarks, based on small, curated datasets, fail to capture the inherent heterogeneity and large spatial context of in-the-wild EM data, imposing fundamental limitations on current patch-based methods. To address these limitations, we developed a large-scale, multi-source benchmark for multi-organelle instance segmentation, comprising over 100,000 2D EM images across variety cell types and five organelle classes that capture real-world variability. Dataset annotations were generated by our designed connectivity-aware Label Propagation Algorithm (3D LPA) with expert refinement. We further benchmarked several state-of-the-art models, including U-Net, SAM variants, and Mask2Former. Our results show several limitations: current models struggle to generalize across heterogeneous EM data and perform poorly on organelles with global, distributed morphologies (e.g., Endoplasmic Reticulum). These findings underscore the fundamental mismatch between local-context models and the challenge of modeling long-range structural continuity in the presence of real-world variability. The benchmark dataset and labeling tool will be publicly released soon.
|
https://arxiv.org/abs/2601.12464
|
Academic Papers
|
svg
|
c1f9dd1d3941f6c2eafd0e598c133fd45a2314af040edfa39a938c79f581a261
|
2026-01-21T00:00:00-05:00
|
Incentivizing In-depth Reasoning over Long Contexts with Process Advantage Shaping
|
arXiv:2601.12465v1 Announce Type: new Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has proven effective in enhancing LLMs short-context reasoning, but its performance degrades in long-context scenarios that require both precise grounding and robust long-range reasoning. We identify the "almost-there" phenomenon in long-context reasoning, where trajectories are largely correct but fail at the final step, and attribute this failure to two factors: (1) the lack of high reasoning density in long-context QA data that push LLMs beyond mere grounding toward sophisticated multi-hop reasoning; and (2) the loss of valuable learning signals during long-context RL training due to the indiscriminate penalization of partially correct trajectories with incorrect outcomes. To overcome this bottleneck, we propose DeepReasonQA, a KG-driven synthesis framework that controllably constructs high-difficulty, multi-hop long-context QA pairs with inherent reasoning chains. Building on this, we introduce Long-context Process Advantage Shaping (LongPAS), a simple yet effective method that performs fine-grained credit assignment by evaluating reasoning steps along Validity and Relevance dimensions, which captures critical learning signals from "almost-there" trajectories. Experiments on three long-context reasoning benchmarks show that our approach substantially outperforms RLVR baselines and matches frontier LLMs while using far fewer parameters. Further analysis confirms the effectiveness of our methods in strengthening long-context reasoning while maintaining stable RL training.
|
https://arxiv.org/abs/2601.12465
|
Academic Papers
|
svg
|
2da03b54b3fc62ebfec69f4bc8e8e5e28ca32d1722071cbc450f7f57317cbf7d
|
2026-01-21T00:00:00-05:00
|
Patch-Level Tokenization with CNN Encoders and Attention for Improved Transformer Time-Series Forecasting
|
arXiv:2601.12467v1 Announce Type: new Abstract: Transformer-based models have shown strong performance in time-series forecasting by leveraging self-attention to model long-range temporal dependencies. However, their effectiveness depends critically on the quality and structure of input representations derived from raw multivariate time-series data. This work proposes a two-stage forecasting framework that explicitly separates local temporal representation learning from global dependency modelling. In the first stage, a convolutional neural network (CNN) operates on fixed-length temporal patches to extract short-range temporal dynamics and non-linear feature interactions, producing compact patch-level token embeddings. Token-level self-attention is subsequently applied during representation learning to refine these embeddings by enabling interactions across temporal patches. In the second stage, a Transformer encoder processes the resulting token sequence to model inter-patch temporal dependencies and generate per-patch forecasts. Experiments conducted on synthetic multivariate time-series data with controlled static and dynamic factors demonstrate that the proposed patch-based tokenization strategy achieves competitive forecasting performance compared to convolutional and patch-based Transformer baselines. The results highlight the importance of structured temporal representations and show that decoupling local temporal encoding from global attention-based modelling yields more effective and stable time-series forecasting.
|
https://arxiv.org/abs/2601.12467
|
Academic Papers
|
svg
|
ac72744d9c76e8d8188ef87e34256d06d60ed63f0767c4c12337353d6304597c
|
2026-01-21T00:00:00-05:00
|
DCAC: Dynamic Class-Aware Cache Creates Stronger Out-of-Distribution Detectors
|
arXiv:2601.12468v1 Announce Type: new Abstract: Out-of-distribution (OOD) detection remains a fundamental challenge for deep neural networks, particularly due to overconfident predictions on unseen OOD samples during testing. We reveal a key insight: OOD samples predicted as the same class, or given high probabilities for it, are visually more similar to each other than to the true in-distribution (ID) samples. Motivated by this class-specific observation, we propose DCAC (Dynamic Class-Aware Cache), a training-free, test-time calibration module that maintains separate caches for each ID class to collect high-entropy samples and calibrate the raw predictions of input samples. DCAC leverages cached visual features and predicted probabilities through a lightweight two-layer module to mitigate overconfident predictions on OOD samples. This module can be seamlessly integrated with various existing OOD detection methods across both unimodal and vision-language models while introducing minimal computational overhead. Extensive experiments on multiple OOD benchmarks demonstrate that DCAC significantly enhances existing methods, achieving substantial improvements, i.e., reducing FPR95 by 6.55% when integrated with ASH-S on ImageNet OOD benchmark.
|
https://arxiv.org/abs/2601.12468
|
Academic Papers
|
svg
|
98b0fce24dcd65439844238a7395c11a1d0d382599984c65f42a7d30a36079a3
|
2026-01-21T00:00:00-05:00
|
Knowing When to Abstain: Medical LLMs Under Clinical Uncertainty
|
arXiv:2601.12471v1 Announce Type: new Abstract: Current evaluation of large language models (LLMs) overwhelmingly prioritizes accuracy; however, in real-world and safety-critical applications, the ability to abstain when uncertain is equally vital for trustworthy deployment. We introduce MedAbstain, a unified benchmark and evaluation protocol for abstention in medical multiple-choice question answering (MCQA) -- a discrete-choice setting that generalizes to agentic action selection -- integrating conformal prediction, adversarial question perturbations, and explicit abstention options. Our systematic evaluation of both open- and closed-source LLMs reveals that even state-of-the-art, high-accuracy models often fail to abstain with uncertain. Notably, providing explicit abstention options consistently increases model uncertainty and safer abstention, far more than input perturbations, while scaling model size or advanced prompting brings little improvement. These findings highlight the central role of abstention mechanisms for trustworthy LLM deployment and offer practical guidance for improving safety in high-stakes applications.
|
https://arxiv.org/abs/2601.12471
|
Academic Papers
|
svg
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.