跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+816新增论文
+642其中 AI 相关
+95已有论文修订
+95新增 AI 导读

arXiv 最近完成扫描:2026-11-21 12:19 · 全站最近入库:2026-11-21 12:19各信源进度 →

统计口径与数据新鲜度

统计 2026-11-20 13:10 — 2026-11-21 13:10(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-21 12:19最近发现新增版本:2026-11-21 12:19库内论文最新更新时间:2026-11-20 02:59

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

29,158 已收录论文24,181 AI 相关8,992 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 21,490 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-21 12:17(北京时间)
版本更新 · v2arXiv:2611.15360

CARE-Drive: Coupled Alternating Refinement for Ego-trajectories

Xiao Ma, Yijun Chen, Zhiquan Hu, Jicheng Zhao, Ran Ji, Yuming Li, Chuang Dai

原文摘要

Vision-language-action (VLA) models bring multimodal understanding and world knowledge to end-to-end autonomous driving. To further improve the quality of predicted trajectories, existing studies introduce candidate refinement, residual denoising, and token correction. However, these methods typically focus on unified refinement of the entire trajectory, paying limited attention to the distinct physical quantities and their interactions. We further investigate physical specialization and coordination in trajectory refinement, centering on position and heading for simpler and more intuitive correction modeling. To this end, we propose CARE-Drive (Coupled Alternating Refinement for Ego-trajectories), which decomposes trajectory refinement into waypoint residual correction and position-conditioned heading generation. Through imitation learning and a future-token auxiliary task, the backbone provides predictive representations and a base trajectory. The position expert predicts bounded metric waypoint residuals, while the heading expert reconstructs vehicle orientation from adjacent angular increments conditioned on the corrected path. With the backbone frozen, the two experts are optimized alternately through supervised initialization and reinforcement fine-tuning. This design simplifies refinement through clear physical specialization while preserving conditional coupling during prediction and optimization. On NAVSIM v1, CARE-Drive achieves 88.8 PDMS, improving by 2.4 points over the unrefined base trajectories generated by the same backbone. Ablation studies further support the effectiveness of specialized experts and staged training.

cs.CV · cs.AI

版本更新 · v4arXiv:2609.18078

Beyond Embedding Transfer: Architectural Interfaces, Iso-Parameter Scaling, and Preconditioner-Governed Consolidation in Grokking

Zeyu Jia

原文摘要

Grokking, or delayed generalization after memorization, incurs substantial training steps across algebraic tasks. Cross-operator transfer from a grokked donor accelerates target acquisition, yet post-grokking AdamW continuation triggers test-accuracy relapse as unannealed updates drift across residual interfaces and decaying second moments amplify late gradients. Across depths one to three on modular arithmetic ($\mathbb{Z}_{113}$), non-Abelian $S_5$ conjugation, and real-sensory MNIST modular arithmetic, we separate representational transfer from preconditioner-governed consolidation across thirteen experimental batteries. First, co-transferring internal blocks accelerates one-layer and two-layer modular takeoff beyond boundary interfaces, while graded prefix pinning bounds depth-three drift under unannealed continuation. Second, an iso-parameter $S_5$ scaling grid separates attention head dimension from MLP width: widening two-head attention to dimension $64$ reaches $98.54\%$ cold accuracy at $427\mathrm{k}$ parameters and $85.95\%$ boundary-only transfer, while final-slot representations refine conjugation directly across blocks without an intermediate binary-product bottleneck. Third, Plateau-Gated Leaky Consolidation pairs validation-envelope step damping with leaky second-moment decay, reaching $84.85\%$ zero-leakage validation-gated risk-constrained accuracy across five pre-saturation two-layer $S_5$ suites while cutting step budget relative to hard anchoring. Fourth, transferring a grokked visual trunk and blocks to image-disjoint MNIST targets reaches $92.27\%$ and $90.76\%$ accuracy, prunes odd centroid harmonics, and reaches $90.42\%$ risk-constrained accuracy across $60/60$ branches without hyperparameter retuning.

cs.LG

新论文 · v1arXiv:2611.17905

Cracks in Embeddings: Robustness and Stability of JEPA-Based Chest X-Ray Representations

Zahid Mansur, Kalpana Yadav, Deepak Mishra

原文摘要

Self-supervised Joint Embedding Predictive Architecture (JEPA) models are increasingly used for medical imaging, however there is a limited understanding about robustness of these models toward real-world image changes. We test four JEPA-based chest X-ray encoders, trained on 78,468 NIH ChestX-ray14 images, under six tests: two black-box adversarial attacks (SPSA, Boundary Attack), three image corruptions (Gaussian noise, brightness shift, contrast shift), and a label-scarcity test, measuring embedding change and classification AUROC. These perturbations cause severe, model-dependent damage: encoder-level Cosine Similarity drops as low as 0.33, and AUROC falls by as much as 0.12 in the worst case (I-JEPA under the Boundary Test). We test two ways of training the small predictor on top of each frozen encoder to address this. Adversarial fine-tuning, trained on clean and perturbed images together, raises embedding consistency above 0.97 in nearly every case and improves AUROC, relative to the same perturbed baseline, for most models, with the largest gain of $+0.10$ on I-JEPA. A Vector-Quantized codebook method, trained only on clean images and evaluated on perturbed test images for the first time, instead degrades under distortion for all four models, most noticeably for I-JEPA ($-0.15$ AUROC on average), showing that its discrete bottleneck alone does not provide the robustness that comes from training on perturbed data directly.

cs.CV

版本更新 · v2arXiv:2606.13880

A Horizon-Specific Attribute-Conditioned Neural Network for Estimating Health-State Transition Probabilities in Irregular Longitudinal Data: The LANTERN Framework

Bright Kwaku Manu, Beckett Sterner, Petar Jevtic

原文摘要

Accurate estimation of health-state transition probabilities is essential for long-term care insurance modeling. Conventional actuarial multi-state models provide an interpretable framework but may be restrictive in capturing nonlinear relationships and accumulated health histories from irregular longitudinal observations. This paper proposes LANTERN, a history-dependent neural framework that integrates recurrent health representations, elapsed observation time, and adaptive demographic and socioeconomic attribute conditioning to estimate horizon-specific transition probabilities among Healthy, Mild disability, Severe disability, and Death states. Using longitudinal Health and Retirement Study data, we evaluate our model against a recurrent neural network, multinomial logistic regression, gradient-boosted trees, and an actuarial continuous-time Markov chain across one-, two-, and three-year forecast horizons. At the primary two-year horizon, LANTERN achieves competitive predictive accuracy and calibration, with improvements over selected benchmarks. Its aggregated transition-matrix estimates outperform multinomial logistic regression and are comparable to those of the strongest remaining benchmarks. These findings suggest that history-conditioned neural estimation provides a flexible and competitive approach to estimating horizon-specific health-state transition probabilities, with potential applications in actuarial long-term care modeling.

cs.LG · q-fin.RM

版本更新 · v3arXiv:2604.06822

Non-RS type cyclic MDS codes over finite fields via cyclotomic field reduction

Can Xiang, Chunming Tang

原文摘要

Cyclic maximum distance separable (MDS for short) codes are a special subclass of linear codes and have received a lot of attention, as these codes have very important applications in many areas including quantum codes, designs and finite geometry. However, the existing construction methods for cyclic MDS codes mainly rely on strict restrictions on certain parameters or involve relatively complex procedures. In this paper, we investigate this problem further via norm reduction in cyclotomic fields. By converting the verification of the MDS property over a finite field into checking non-zero minors in characteristic zero, we propose a construction method of cyclic MDS codes over finite fields via cyclotomic field reduction. Based on this method, we obtain several cyclic MDS codes over finite fields and many non-RS type cyclic MDS codes are produced. Compared with the existing construction method in [Non-Reed-Solomon Type Cyclic MDS Codes, IEEE Trans. Inf. Theory, 71(5): 3489--3496, 2025], we give a new and relatively direct characterization method for constructing cyclic MDS codes over finite fields and the obtained non-RS cyclic MDS codes have flexible parameters in terms of the code length $n$, the defining set $J$, and the underlying finite field $\mathbb{F}_q$.

cs.IT

版本更新 · v2arXiv:2602.24195

Uncertainty Quantification for Multimodal Large Language Models with Incoherence-adjusted Semantic Volume

Gregory Kang Ruey Lau, Hieu Dao, Nicole Kan Hui Lin, Bryan Kian Hsiang Low

原文摘要

Despite their capabilities, Multimodal Large Language Models (MLLMs) may produce plausible but erroneous outputs, hindering reliable deployment. Accurate uncertainty metrics could enable escalation of unreliable queries to human experts or larger models for improved performance. However, existing uncertainty metrics have practical constraints, such as being designed only for specific modalities, reliant on external tools, or computationally expensive. We introduce UMPIRE, a training-free uncertainty quantification framework for MLLMs that works efficiently across various input and output modalities without external tools, relying only on the models' own internal modality features. Inspired by Determinantal Point Processes (DPPs), UMPIRE computes a novel incoherence-adjusted semantic volume of sampled MLLM responses for a given task instance, which couples semantic embedding dispersion with the quadratic entropy of the responses. This approach effectively captures both the global semantic diversity of samples and the local incoherence of responses based on internal model confidence. We provide theoretical analysis motivating UMPIRE's design, and extensive experiments show that UMPIRE consistently outperforms baseline metrics in error detection and uncertainty calibration across image, audio, and video-text benchmarks, even with small sample budgets. We also demonstrate UMPIRE's generalization to non-text output tasks, including image and audio generation. Our code is available at https://github.com/daohieu17ctt/UMPIRE.

cs.AI · cs.CL · cs.CV · cs.LG

新论文 · v1arXiv:2611.17900

AmbiSteer: Selectively Eliciting Ambiguity-Aware Responses in Vision-Language Models

Monica Riedler, Samuele Antonelli, Barbara Plank, Raffaella Bernardi

原文摘要

Vision-language models (VLMs) often exhibit overconfidence, providing firm answers even when visual evidence is insufficient. A prominent case is referential ambiguity in Visual Question Answering (VQA), where multiple entities in an image are plausible referents of a question. Addressing such cases requires recognizing when ambiguity is present: ambiguous questions can be handled by asking for clarification or enumerating plausible referents, while unambiguous questions should receive a direct answer. Current VLMs struggle with this distinction, often defaulting to a single interpretation. To investigate whether activation steering can serve as an inference-time mechanism to elicit such behaviors, we introduce AmbiSteer, a lightweight framework to selectively elicit ambiguity-aware behavior. Across three VLMs, we find that steering alone has limited and model-dependent effects, while prompting is generally more effective. Combining prompting and steering substantially increases clarification and enumeration behaviors, but also elicits them on unambiguous cases. AmbiSteer addresses this using a linear classifier to predict input ambiguity and condition the steering direction accordingly, increasing the proportion of inputs handled appropriately from 59.0% to 77.0% on average across models. Our work sets the ground for lightweight approaches to reduce model overconfidence in uncertain multimodal settings.

cs.CV · cs.CL

版本更新 · v2arXiv:2608.09654

Hallucination-Free GUI Grounding via Regression-Free Layout-Aware Matching

Yuke Li, Xuehan Hou, Jie Chen

原文摘要

GUI agents are shifting from metadata-dependent large language models to purely visual multimodal large language models (MLLMs) that operate directly on screenshots. The core task, GUI grounding, requires translating abstract user instructions into precise element coordinates. This task faces a persistent dual obstacle: conventional grounding models lack the semantic richness to interpret abstract instructions, while end-to-end MLLMs suffer from coordinate hallucinations caused by deficient fine-grained perception. We propose a regression-free framework where a frozen MLLM performs instruction parsing and a dedicated grounding model handles precise localization without learning any coordinate regression. A frozen MLLM first elaborates the abstract instruction into a structured visual description rich in layout cues. These descriptions are then fed to a novel Layout-Aware GUI Grounding Model, which performs regression-free localization by matching against layout-prior candidates, inherently suppressing hallucinations and avoiding expensive fine-tuning. The grounding model is trained with only Text/Icon binary labels, requiring no coordinate regression parameters. On ScreenSpot-Pro, our method achieves over 20% improvement in grounding accuracy over end-to-end systems; on Mind2Web, it raises success rate and element selection rate by more than 15%. These results demonstrate that decoupling instruction understanding from layout-aware localization effectively resolves the core challenges of GUI interaction.

cs.AI

新论文 · v1arXiv:2611.17899

More Genes, Faster Inference: Efficient Large-Panel Spatial Transcriptomics Prediction from Histology

Songming Zhang, Yuxiao Luo, Shifu Luo, Siran Chen, Shuquan Su, Jinyan Li

原文摘要

Spatial transcriptomics (ST) prediction from histology aims to infer spatially resolved gene expression profiles directly from tissue images. This is a challenging problem that is currently often addressed with complicated model architectures, additional inputs, and complicated training procedures, while remaining limited to fixed, small gene panels. This raises a natural question: Is there an efficient method for large-panel ST prediction? Here, we present GeneFlow, a simpler yet effective framework for large-scale spatial gene expression prediction that unifies flow-matching generative modeling with structured multimodal fusion. Specifically, GeneFlow first encodes both histological morphology and cross-gene interactions in spatial spots through histology and gene tokenizers. The model directly predicts spatial gene expression, rather than an intermediate velocity field, while being trained with a velocity-based flow-matching objective. We show that, even without relying on heavy generative pipelines, GeneFlow achieves state-of-the-art performance across multiple ST benchmarks, with up to 56.8% gains in 1,000-gene prediction. Extensive experiments further show that GeneFlow produces biologically meaningful spot profiles while reducing computational cost by orders of magnitude, thereby improving efficiency and scalability, particularly in clinical settings. We believe GeneFlow provides an opportunity for leveraging widely available histology images for downstream biological discovery. Code is available at: https://github.com/ZZhangsm/GeneFlow

cs.CV · cs.AI · q-bio.QM

新论文 · v1arXiv:2611.17898

Evaluating Polling Schemes for OFDMA Uplink Scheduling in Time-Sensitive Wi-Fi Networks

Douglas Dziedzorm Agbeve, Ferhat Van Herck, Andrey Belogaev, Jeroen Famaey

原文摘要

IEEE 802.11ax introduced orthogonal frequency division multiple access (OFDMA) to improve channel utilization through concurrent multi-user transmissions. For effective uplink (UL) scheduling, the access point (AP) requires timely knowledge of buffered traffic at the stations (STAs). Explicit buffer-status polling provides this, yet each polling round can query only as many STAs as there are resource units (RUs). Scheduled-access, random-access, activity-based, and learning-based schemes address this buffer-status acquisition problem differently and have generally been evaluated in isolation under different conditions, so it remains unclear how they compare across time-sensitive workloads. We evaluate common schemes from the Wi-Fi standard and literature under three time-sensitive environments: a large-scale teleconferencing session, an industrial IoT deployment, and an extended reality (XR) service. The relative performance of the schemes depends strongly on traffic characteristics and offered load. In conferencing, activity-based and learning-based polling perform best, while conventional contention-based access remains competitive under light load. In the industrial scenario, scheduled-access most effectively delivers time-sensitive packets within their deadlines under heavy load, whereas contention-based access performs best under lighter load. In XR, OFDMA-based schemes achieve high video throughput that contention-based access cannot sustain; among them, activity-based polling achieves the lowest deadline-miss ratios for uplink head-tracking traffic, at the expense of longer video delay tails. These findings highlight the importance of selecting channel access and scheduling schemes according to traffic requirements and operating conditions.

cs.NI

新论文 · v1arXiv:2611.17897

See and Detect: Image-Driven Open-Vocabulary Multi-View 3D Object Detection

Yating Xu, Na Zhao, Chenhan Jiang, Qingwen Zhang, Jiankang Deng, Jifei Song

原文摘要

Recent open-vocabulary multi-view indoor 3D detection methods typically follow a point-cloud-driven pipeline, where pseudo 3D boxes are generated from reconstructed or estimated geometry and semantic labels are inferred from projected regions. However, such a design makes box generation dependent on the completeness of point clouds and restricts semantic learning to local appearance, causing inferior 3D object localization and open-vocabulary classification. In this paper, we present See and Detect, an image-driven framework that uses multi-view images as the primary evidence for 3D box and semantic label generation. For box generation, we first discover object candidates from individual images, enabling objects missing in point clouds to be recovered from visual evidence. However, single-view predictions are often noisy and redundant. To address this, we propose multi-view consensus box consolidation to fuse spatially consistent predictions across views, and box quality assessment to reduce the impact of boxes with partial object coverage. For semantic label generation, we introduce context-aware semantic learning, where semantic labels are queried from full image context, enabling more accurate semantic prediction. Extensive experiments on ScanNet and ARKitScenes demonstrate that our method achieves state-of-the-art open-vocabulary multi-view 3D detection performance while running over 500$\times$ faster than the previous state-of-the-art method.

cs.CV

版本更新 · v2arXiv:2611.12762

Sage-Taiyi: A Metacognitive Architecture for Scientific Discovery

Weida Wang, Benteng Chen, Yigeng Jiang, Shuchen Pu, Weiping Shen, Zichao Liu, Yulin Su, Zitong Wang 等 90 位作者

原文摘要

We introduce Sage-Taiyi, a metacognition-inspired architecture that uses accumulated scientific understanding to guide what to investigate next and how to investigate it. Through explicit monitoring and control, the system revises its understanding of evidence, assumptions, and unresolved questions, then uses these assessments to adapt research targets and select reusable reasoning methods. On Taiyi-Bench, an AI co-scientist ideation benchmark comprising 265 tasks from 50 papers across ten disciplines, Sage-Taiyi ranks first on all eight metrics of historical rediscovery and open-ended ideation, achieving an Overall score of 54.0 compared with 47.9 for the strongest baseline, AI Scientist-v2. It also performs well on two external scientific ideation benchmarks. Five expert-guided collaborations further connect its proposals to experimental and computational validation across chemistry, materials science, and engineering. In chemistry, reaction-condition optimization raises the yield of cobalt-catalyzed carbonylation from trace amounts to 84% and photocatalytic allylic amination from 13% to 69%. These results support metacognitive regulation as an approach to scientific ideation and its progression toward practical validation.

cs.AI

版本更新 · v2arXiv:2603.26378

Generative machine learning for protein design

Senura Hansaja Wanasekara, Manoj M Wagle, Minh-Duong Nguyen, Xiaochen Liu, Nguyen H. Tran, Manolis Kellis, Ken-Tye Yong

原文摘要

Designing proteins with specific biological functions is of considerable interest for both the development of new therapeutics and the engineering of enzymes, yet achieving the intended activity remains challenging. Generative artificial intelligence (AI) has emerged as a powerful tool to explore the vast space of possible designs by learning from protein sequence and structural data. Here, we review 59 generative approaches across six protein design tasks, including sequence generation, backbone generation, structure-conditioned sequence design, joint sequence-structure design, motif scaffolding, and binder and antibody design. We outline common generative approaches and protein representations and discuss how structural constraints and functional information are utilized during generation. This is followed by a discussion of the computational and experimental approaches used to evaluate designed proteins and what such assessments reveal about their properties. Finally, we discuss current limitations in designing proteins beyond well-represented families and incorporating experimental feedback into the design process. We expect that rapid advances in generative modeling will enable more reliable design of a wider range of protein structures and functions over the coming years.

cs.LG · cs.AI

新论文 · v1arXiv:2611.17892

Hierarchical Bayesian physics-augmented neural networks for populations of hyperelastic laws: specimen variability and uncertainty from scarce, noisy data

Vincent C. Scholz, Phaedon-Stelios Koutsourelakis

原文摘要

The hyperelastic law of a material such as soft tissue differs between specimens, and often only a few specimens can be tested. Physics-augmented neural networks (PANNs) learn such laws from data without a prescribed functional form while keeping them physically admissible, and Bayesian versions of related models quantify the uncertainty of the law of one specimen or of a pooled group of specimens. A pooled fit, however, attributes the differences between specimens to measurement noise or to uncertainty about a mean law, and neither describes a new specimen. We first develop a Bayesian PANN for one specimen, whose priors keep the learned law plausible for few, noisy data. We then extend it to a hierarchical Bayesian PANN, a nonlinear mixed-effects model in which all specimens share learned polyconvex features and each specimen is a draw from a population of laws. Features, specimens, and population are inferred jointly by variational inference, with a separate model of the measurement noise, so that the variability between specimens is not mistaken for noise, the population carries its own uncertainty, and every specimen drawn from it is polyconvex. The population predicts the response of new specimens with credible intervals, without data or from a partial test, and assigns them to cohorts with quantified uncertainty, all without a prescribed form of the law. On synthetic populations, it recovers the population from few specimens and predicts new specimens more reliably than three published methods. On Treloar's rubber and on healthy and aneurysmal aortas, it reaches the accuracy of a deterministic PANN and assigns no aorta specimen to the wrong cohort with confidence, although it underestimates the uncertainty of untested curves.

cs.CE · cs.LG · stat.ML

版本更新 · v2arXiv:2610.32664

Zoom Where the Answer Looks: Cross-Resolution Self-Distillation for MLLMs under Visual-Token Budgets

Zhensheng Hu, Guanglong Xu, Guanqiang Hu

原文摘要

Under the small per-view visual-token budgets of efficient deployment, the fine text of documents, charts and infographics becomes illegible to multimodal large language models (MLLMs). Zooming in recovers it, but the region must be chosen before the model answers. Existing methods decode an answer or a box first, at the cost of generation, or read the attention available before answering, which localizes the evidence only in top layers whose depth varies with model size. Learned region proposers instead distill the attention of the model's answers, but on the same budgeted view. Yet this attention is precise only on a view the model can read: on Qwen3-VL-4B, a window around its peak covers 83% of the evidence on a high-resolution view but 54% on the budgeted view. Since answers and high-resolution views are available during training, we introduce LookZoom, a zooming method that trains a dedicated look token by cross-resolution self-distillation of the answer's attention: with the MLLM frozen, the token learns to attend on the budgeted view, before any answer, where the model's own correct answers attend on a high-resolution view. The region around its attention peak is the window to zoom into, and reinforcement learning then trains the MLLM to answer from both views. With 98 visual tokens per view on five text-rich benchmarks, LookZoom surpasses attention-guided zooming by 20-24 points and grounding-based zooming and learned crop routing by 12-15 points on Qwen3-VL at three scales, and it outperforms learned region proposers at every budget from 98 to 392 tokens. On Qwen3-VL, the look token alone, with the MLLM frozen, already surpasses every baseline that leaves the language model unchanged, and the recipe carries over to InternVL3.5 and Qwen2.5-VL.

cs.CV

新论文 · v1arXiv:2611.17887

RuralShield: Super Intelligence-Assisted Security Triage and Serverless Infrastructure for Resource-Constrained Healthcare

Mohammad Arman Ullah, Asif Ahmed

原文摘要

Resource-constrained healthcare providers must connect cloud security controls to an operational response process. Introducing language-model assessments adds an authorization problem: a recommendation must be converted into a permitted action against a specific resource. RuralShield combines a configurable serverless application foundation with structured model assessment, deterministic routing, and human-approved containment. Its design separates workload access, assessment validation, and action authorization so that each boundary produces inspectable records. We evaluate the implementation through a frozen 120-hour non-AI workload and separate desktop and AWS AI cohorts. The workload recorded 999,861 authenticated requests and 999,860 successful responses, with one HTTP 404 and one missing minute representing 139 scheduled requests. Median, 95th-percentile, and 99th-percentile Ohio client latencies were 28.57, 53.01, and 110.77 ms. Eight successful model requests exercised assessment and reporting in two four-case cohorts. The AWS cohort and five scripted controls produced nine completed workflows, including two human-approved Lambda containment cases with verified restoration of the original concurrency setting. Source-linked, checksum-verified archives preserved the records before scoped cleanup. The study demonstrates an inspectable integration of application controls and human-mediated AI response under the evaluated synthetic conditions.

cs.CR

新论文 · v1arXiv:2611.17881

Photogrammetric 3D Model-Based Animal Behavior Classification

Olaf Hellwich, Moritz Heine, Patrik Reiske, Sole Traverso, Katharina Hohlbaum, Henning Sprekeler, Christa Thöne-Reineke, Lars Lewejohann 等 9 位作者

原文摘要

Video-based behavior estimation is often implemented end-to-end from image sequences. We advocate a two-stage alternative: reconstruct 3D body and object geometry before classifying behavior. This makes ambiguities of the ill-posed 2D-to-3D inverse problem explicit and potentially correctable. We demonstrate the strategy with simulated interacting mice and extend it to real video.

cs.CV · cs.AI

新论文 · v1arXiv:2611.17880

Soteria: Formally Verified Planning with Runtime Enforcement for Safe LLM Agents

Mike He, Ankush Desai, Sharad Malik, Aarti Gupta

原文摘要

LLM agents that execute multi-step tool calls must satisfy two objectives simultaneously: completing the user's task (utility) and conforming to domain policies and correctness constraints (safety). These objectives are in tension -- blocking an unsafe action prevents a violation but can leave the agent stranded with no principled way to recover. Existing approaches sacrifice one objective for the other. We introduce Soteria, a framework that reconciles safety and utility through verified hierarchical planning and runtime enforcement of specifications. Before execution, the agent generates a structured plan that is formally verified against the specifications prior to any tool invocation. During execution, the verified plan serves dual roles: it ensures that the agent takes specification-conformant trajectories and provides guidance when unsafe actions are blocked. Across multiple benchmarks covering a diverse range of tool-use tasks, Soteria achieves 1.7 times the average task success rate of the best baselines, with no violations of policy specifications, demonstrating that safety and utility need not be traded off.

cs.AI · cs.SE

新论文 · v1arXiv:2611.17879

From Local Prediction Error to Final Outputs: A Theory of Expert Sharing in Multi-Step Reasoning

Tiantong Wang, Zhiren Gong, Yixin Wang, Tiantong Wu, Wei Yang Bryan Lim

原文摘要

Mixture-of-experts (MoE) models can share experts across layers or computational steps, pooling training information from different contexts. This can reduce local prediction error, but multi-step systems are ultimately evaluated by their final outputs. We study when these local gains survive composition. In specified linear regression models, we derive exact sharing and merging criteria by balancing variance savings against bias from sharing across roles with different target behavior. Under a common fixed linear propagation map, we characterize expected squared final-output risk, retaining error directions and correlations. We construct cases where sharing lowers average local error yet increases final-output error. Within this setting, we also identify when orthogonal full pooling with isotropic noise preserves every local improvement at the final output. For general routed computations, we derive conditional end-to-end error bounds that account for distribution shifts, errors in surrounding components, and changes in expert selection. Separately, we show that correct predictions and route labels alone do not establish that individual experts implement the attributed operations. Controlled experiments realize a strict local-to-final risk reversal. In low-data symbolic settings, trained Transformers exhibit statistically supported teacher-forced one-step gains without statistically resolved free-running final-output gains. Separate intervention tests distinguish successful multi-layer actions from single-expert attribution and generalization to unseen transitions. Together, these results distinguish when sharing improves local estimation, when those gains survive to the final output, and what evidence is needed to attribute reusable operations to individual experts.

cs.LG · cs.AI

版本更新 · v5arXiv:2507.14632

BusterX++: Towards Unified Cross-Modal AI-Generated Content Detection and Explanation with MLLM

Haiquan Wen, Tianxiao Li, Zhenglin Huang, Yiwei He, Guangliang Cheng

原文摘要

The rapid advancement of generative AI has substantially improved image and video synthesis, amplifying the risk of multimodal visual misinformation. Recent MLLMs have shown promise for transparent AI-generated content detection through reasoning and explanation, yet existing approaches largely treat image and video forensics as isolated tasks, leaving cross-modal synergies underexplored. To address this, we present \textbf{BusterX++}, a unified MLLM for joint image and video detection with interpretable reasoning. We also introduce \textbf{GenBuster-Bench++}, a curated benchmark containing balanced image and video samples spanning recent generation models and diverse real-world scenarios, with synthetic images and videos screened using the same human difficulty criterion. Using this controlled setting, we revisit the widely adopted SFT+RL post-training paradigm. Notably, in the unified image-video setting, our results demonstrate that a single-stage pure RL strategy driven primarily by sparse outcome rewards outperforms a strong SFT+RL baseline. We observe that single-stage pure RL maintains higher policy entropy than SFT-initialized RL throughout training. This observation suggests a possible role for exploration in the performance gains observed under unified image-video training. Extensive experiments demonstrate that BusterX++ achieves state-of-the-art performance, highlighting the powerful potential of RL for unified cross-modal visual reasoning.

cs.CV

新论文 · v1arXiv:2611.17876

Delayed Effects of Temporary Weight Decay in Muon Pretraining

Md Rifat Arefin, Ravid Shwartz-Ziv, Ernie Chang, Chinnadhurai Sankar, Rylan Conway, Aristide Baratin, Irina Rish, Adithya Sagar 等 9 位作者

原文摘要

A temporary optimization intervention can increase validation loss when it ends yet reduce it at the final training budget. We demonstrate this with temporary weight decay in Muon-trained decoder-only Transformers with 162M, 405M, 1.4B, and 3.4B parameters. For models up to 1.4B parameters, moving a fixed-duration decay window from early to late training reverses its final effect. At 162M, changing window duration alters final loss even at approximately matched total shrinkage. Most strikingly, a schedule selected at 162M and transferred without retuning to 1.4B is 0.26 nats worse than no decay when removed, overtakes it only after 73% of training, and ultimately finishes 0.028 nats better while also outperforming the best sampled constant-decay baseline. Early decay leaves a persistent weight-norm deficit, which keeps relative steps larger after the intervention ends. An approximate norm-dynamics recurrence captures the slow convergence of these relative steps, and fixed-norm replays of the measured trajectories reproduce the performance ordering and a similar loss difference. At 162M and 405M, norm-reset experiments show that smaller weight norms at the end of the decay window explain most of the final loss improvement. A separate Gram intervention exhibits related post-removal norm and relative-step dynamics. These results show that temporary optimization schedules can have delayed effects long after removal and should therefore be evaluated at the intended final training budget, not only at removal.

cs.LG · cs.CL

新论文 · v1arXiv:2611.17872

Protection-Oriented Assessment of Machine Learning-Based Fault Detection in Power Systems

Julian Oelhaf, Georg Kordowich, Changhun Kim, Christian Bergler, Andreas Maier, Johann Jäger, Siming Bayer

原文摘要

Machine learning (ML)-based fault detectors are increasingly proposed for power system protection, yet aggregate metrics can obscure the consequences of false and missed trips. We present a protection-oriented assessment of three ML detectors and a sequence-current reference on PROTECT-90, comprising 9,022 simulated fault episodes on a 90 kV double-circuit line. The assessment combines five window conditions, causal window-end labelling, episode-grouped cross-validation, and operating-point selection under a fixed security budget. It reports security, dependability, detection timing, decision margins, and episode-level nuisance assertions. Gradient-boosted trees, a convolutional neural network, and logistic regression achieve AUC-ROC values above 0.999, yet at a security budget of $10^{-3}$ their developed-fault missed-trip rates range from 0.007% to 1.99%. Post-clearing false-trip rates of approximately 0.12-0.16% per window coexist with nuisance assertions in 11.3-19.1% of episodes. Among the analysed physical descriptors, missed trips are most consistently associated with higher fault resistance within the sampled range. An analytical current proxy suggests a fault-current gradient only partly reproduced by waveform-derived current (rank correlation 0.62). No single topology state or faulted phase consistently dominates within the sampled domain. Because PROTECT-90 contains fault episodes only, security is assessed on pre-fault and post-clearing windows rather than independent non-fault disturbances.

cs.LG · eess.SP

新论文 · v1arXiv:2611.17870

OmniScore-Bench: Evaluating Vision-Language Models on Musical Score Reading

Congren Dai, Zhancheng Guo, Danni Zhao, Wenchu Wang, Xiaobing Li, Maosong Sun

原文摘要

Vision-language models (VLMs) can answer questions about musical scores, but their ability to recognise notation, recover its structure and interpret its musical meaning requires separate evaluation. Existing optical music recognition (OMR) resources establish detection and transcription measurements; a transcription score alone leaves the strengths and bottlenecks of a VLM score reader unresolved. We introduce OmniScore-Bench, an evaluation of OMR capabilities in VLMs through 3,000 questions spanning six aspects and 24 task subtypes. Each question specifies a visible scope, reference annotations and an answer schema, enabling automatic evaluation from symbol localisation to musical consistency and evidence-linked question answering. Our assessment of twelve VLMs reveals operation-specific capability profiles and different transcription leaders across excerpt, single-page and multipage tasks. Controlled analyses examine the roles of score images, auxiliary givens and additional page context. On the tested subset, image removal exposes visual dependence in pitch and repeat interpretation while measuring how much performance remains with public context alone. Separate format and content analyses distinguish answer-format violations from musical errors. OmniScore-Bench supports diagnosing which score-reading operations VLMs perform reliably and which require improvement.

cs.AI

新论文 · v1arXiv:2611.17869

Process Mining-Enabled Federated Anomaly Detection for the Industrial IoT

Francesco Vitale, Salvatore Bramante, Francesco Flammini, Hangli Ge

原文摘要

While federated anomaly detection enables the integration of local data insights across diverse nodes to reduce false positives, it inherently struggles with non-IID data. In practice, industrial IoT plants often operate under varying conditions and experience diverse environmental phenomena, resulting in significant heterogeneity in the monitored time series. In such settings, standard federated learning approaches fail to perform nuanced aggregation that accounts for these cross-node differences, ultimately leading to increased false positive rates. In this paper, we propose a process mining approach to evaluate the similarity of local node data against a reference provided by a global node. The global node utilizes these similarity metrics to dynamically weight local models during aggregation, producing personalized updates distributed across the network. We evaluate our approach using industrial IoT time-series data collected from manufacturing plants operating under distinct conditions. Experimental results demonstrate that our process mining-enabled federated anomaly detection pipeline achieves superior performance, yielding a 46.52% and 11.99% relative improvement in F1-score compared to isolated local training baselines and standard FedAvg, respectively, alongside a 51.24% relative reduction in false positives over standard federated aggregation.

cs.LG

新论文 · v1arXiv:2611.17868

Speed-Variation Distillation for Generative Modeling

Sukit Seripanitkarn, Phonphrm Thawatdamrongkit, Supasorn Suwajanakorn

原文摘要

While diffusion models excel at image generation, their denoising trajectories pass through corrupted intermediate states far from the image manifold. We ask whether generation can instead proceed through recognizable images throughout the entire trajectory. We introduce a generative framework built around a variation-reducing degradation process that contracts diverse images toward a shared prototype while preserving recognizable intermediate states. Because such manifold traversal lacks a simple analytical form, we construct the trajectories using generative-prior-based image morphing. We then distill the reverse process into a diversification operator that transforms the prototype into increasingly varied samples. To resolve the one-to-many ambiguity of the reverse process, we associate each trajectory with its initial noise latent and apply perturbations to improve robustness against accumulated errors. At inference, every step yields a recognizable image, allowing early termination and enabling a controllable speed--variation trade-off.

cs.CV

版本更新 · v9arXiv:2505.12620

BusterX: MLLM-Powered AI-Generated Video Forgery Detection and Explanation

Haiquan Wen, Yiwei He, Zhenglin Huang, Tianxiao Li, Zihan Yu, Xingru Huang, Lu Qi, Baoyuan Wu 等 10 位作者

原文摘要

As generative video models become increasingly realistic, detecting AI-generated videos requires systems that offer both accuracy and interpretability. However, applying Multimodal Large Language Models (MLLMs) to video forensics is currently limited by outdated datasets, simplistic evaluation protocols, and a reliance on black-box classification. To address these issues, we introduce a comprehensive dataset, benchmark, and baseline model for video forgery detection. First, we present \textbf{GenBuster-200K}, a fair dataset of over 200,000 high-quality videos sourced from state-of-the-art generators, featuring diverse real-world scenarios. Second, we propose \textbf{GenBuster-Bench}, a diagnostic benchmark spanning three progressive tracks (In-Domain, Out-of-Domain, and In-the-Wild) to evaluate models across \textit{domain shifts} and \textit{generational shifts}. Finally, we develop \textbf{BusterX}, a simple MLLM baseline with Reinforcement Learning (RL) training. Instead of direct binary classification, BusterX formulates detection as a visual reasoning task, where the generated reasoning chain serves as detector itself. Experimental results demonstrate that BusterX outperforms several leading MLLMs in both detection accuracy and rationale quality.

cs.CV

新论文 · v1arXiv:2611.17865

Starlight: Whittle-Driven Fidelity Rescue for Satellite-Terrestrial Quantum Networks

Yaliang Shi, Zhiwei Zhao, Bangguo Yuan

原文摘要

A planetary-scale quantum internet is constrained by exponential attenuation in terrestrial fibers. Low-Earth-orbit (LEO) satellites provide a low-loss vacuum backbone, but limited satellite resources must be allocated among competing connections under continuously evolving and heterogeneous physical conditions. Existing satellite scheduling remains largely provisioning-oriented, optimizing satellite-resource assignment for aggregate entanglement delivery. Recent age-based approaches instead incorporate chronological freshness into entanglement scheduling. Neither line of work directly identifies the connections most urgently requiring satellite assistance as their physical states evolve. We propose Starlight, a demand-driven satellite-terrestrial entanglement routing architecture that targets degrading connections before their terrestrial entanglement becomes unusable. Starlight introduces Effective Age of State (EAoS), a physics-aware metric that maps heterogeneous decoherence and swapping-induced fidelity penalties onto a unified additive coordinate. Representing each connection by evolving EAoS, we formulate rescue scheduling as a Restless Multi-Armed Bandit (RMAB) problem and derive a closed-form Whittle-style index approximation for top-K selection. Starlight further employs a two-timescale control plane and an asynchronous dispatch-and-mask pipeline to suppress redundant in-flight allocations. Simulations show that Starlight achieves up to 84.5% average satisfaction under constrained rescue capacity and sub-0.5-ms orchestration at 1,000 connections, while remaining robust to heterogeneous memory decay and stochastic atmospheric degradation.

cs.NI

新论文 · v1arXiv:2611.17864

Disentangling Forgetting from Utility: Lightweight LLM Unlearning via Orthogonal Null-Space Projection

Xiuyuan Wang, Weiming Liu, Zongxin Yang, Fan Wang, Liemiao Jin, Rubin Qin, Wei Chen, Bin Shi 等 10 位作者

原文摘要

Large Language Models (LLMs) pose significant security risks by retaining private or hazardous information, creating an urgent need for unlearning techniques that excise specific knowledge without compromising general utility. However, prevailing fine-tuning-based unlearning methods struggle to achieve this balance, often suffering from computational inefficiency and unintended utility degradation due to the entangled nature of knowledge within model parameters. To disentangle forgetting from utility, we propose UUE, a theoretically grounded framework that reformulates unlearning as controllable model editing. First, UUE projects parameter updates into the orthogonal left null space of retained knowledge representations, strictly ensuring that the editing process causes minimal interference to the model's general capabilities. Second, distinct from targeted mapping, we introduce a convex untargeted objective that derives a closed-form analytical solution to effectively disrupt undesired knowledge by pushing forget-set embeddings toward their antipodal points. We implement UUE via lightweight, pluggable adapters (including a low-rank variant, UUE-L), enabling efficient deployment without modifying the frozen backbone. Extensive experiments on TOFU and WMDP benchmarks demonstrate that UUE and UUE-L achieve superior unlearning efficacy, significantly outperforming existing methods.

cs.LG

新论文 · v1arXiv:2611.17862

FlexCast: Flexible Context Learning for Time Series Foundation Models

Zuyang Shen, Luqi Gong, Fanda Fan, Bingsong Wang, Xiang Ao, Yahui Wei, Minghao Liu, Chunjie Luo 等 9 位作者

原文摘要

Time series foundation models (TSFMs) based on autoregressive (AR) next-token prediction entangle two factors: variable-context conditioning and prediction-unit factorization with autoregressive rollout. Strong AR performance therefore does not reveal whether generalization stems from the flexible training distribution, the local prediction target, the left-to-right inference process, or their combination. We propose FlexCast, a flexible context-target learning framework that disentangles these factors. FlexCast reformulates forecasting as variable-length masked prediction over time series patches: during pretraining, it dynamically samples visible context and masked target patches, exposing the model to diverse context-target configurations without sequential decoding. At inference time, the same pretrained model supports one-step forecasting, confidence-based iterative refinement, and next-token-style generation through a unified mask-and-predict interface, enabling controlled comparison of inference schedules under identical architecture, data, and objective. Evaluated on BLAST and multiple zero-shot benchmarks, FlexCast achieves accuracy competitive with or better than strong TSFM baselines while substantially reducing inference latency. Ablations confirm that context diversity and contiguous target masking drive performance, while left-to-right decoding does not yield consistent gains once flexible context-target relationships have been learned, suggesting that the key benefit of AR-style TSFMs lies in flexible training rather than sequential inference. The code is available at https://github.com/BenchCouncil/FlexCast.

cs.AI

版本更新 · v3arXiv:2506.10903

Beyond Gold Standards: Epistemic Ensemble of LLM Judges for Formal Mathematical Reasoning

Lan Zhang, Marco Valentino, Jordan Meadows, André Freitas

原文摘要

Statement autoformalization plays a crucial role in formal mathematical reasoning by enabling the automatic translation of natural language statements into formal languages. While recent advances using large language models (LLMs) have shown promising capabilities in autoformalization, methods for automatically evaluating autoformalization remain underexplored. LLM-as-a-judge presents a promising approach for automating such evaluation, however, existing methods typically employ coarse-grained and generic evaluation criteria, which limit their effectiveness for advanced formal mathematical reasoning, where quality hinges on nuanced, multi-granular dimensions. In this work, we take a step toward addressing this gap by introducing a systematic, automatic method to evaluate autoformalization tasks. The proposed method is based on an epistemically and formally grounded ensemble (EFG) of LLM judges, defined on criteria encompassing logical preservation (LP), mathematical consistency (MC), formal quality (FQ), and formal validity (FV), resulting in a transparent assessment that accounts for different contributing factors. We validate the proposed framework to serve as a proxy for autoformalization assessment within the domain of formal mathematics. Overall, our experiments demonstrate that the EFG ensemble of LLM judges is a more suitable emerging proxy for evaluation than a coarse-grained model. These findings suggest that LLM judges, especially when guided by a well-defined set of atomic properties, could offer a scalable, interpretable, and reliable support for evaluating formal mathematical reasoning.

cs.CL