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

+758新增论文
+578其中 AI 相关
+94已有论文修订
+95新增 AI 导读

arXiv 最近完成扫描:2026-11-17 04:58 · 全站最近入库:2026-11-17 04:58各信源进度 →

统计口径与数据新鲜度

统计 2026-11-16 05:09 — 2026-11-17 05:09(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:2026-11-17 04:58最近发现新增版本:2026-11-17 04:58库内论文最新更新时间:2026-11-14 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。

24,106 已收录论文20,158 AI 相关7,049 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 20,158 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-17 04:57(北京时间)
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:2611.17867

UltraBERT-JEV: Encoder models for System One Applications

Astarag Mohapatra

原文摘要

We study encoder models that pick one of several options in a single forward pass, without generating text. This technical report compares recipes for fine-tuning UltraBERT-380M with one mask per option. The four recipes differ in how an option is scored and how the scores become probabilities; one follows the expert gate of DeepSeek-V3. Trained on the same data as Qwen3-0.6B with the LLM-as-Jev recipe (67.6% test accuracy), the encoder reaches 69.7% to 72.1%. Scoring an option by "yes" minus "no" instead of "yes" alone gives the larger gain, from 69.7% to 71.3% on the test set and from 57.8% to 59.6% on external benchmarks (56.1% for the decoder). Sigmoid normalisation adds 0.8 test points; a square-root softplus adds none. One request at a time on an NVIDIA A100, the encoder takes 29 ms per decision against 68 ms, but uses more peak memory (2.8 against 2.2 GiB). An auxiliary head, trained on human-marked evidence and on labels selected with a reward adapted from SelfCite, predicts in the same pass the sentences annotated as support for the answer.

cs.CL · cs.AI

新论文 · v1arXiv:2611.17866

Active Learning of Temporal Specifications through Distinguishing Queries

Rajarshi Roy, Xiaowei Huang

原文摘要

Temporal specifications are essential for verification and monitoring, yet writing formulas that capture the intended behaviour is difficult. Active learning derives such specifications through interaction between a learner and a teacher, but remains underexplored for temporal logics. Existing approaches require the teacher to answer equivalence queries and to construct counterexamples, and they often rely on SAT-based learning procedures that restart the search after each new example. We present an active-learning framework for LTLf in which the teacher only labels traces and answers natural-language questions about the intended behaviour. The learner maintains an enumerated version space of candidate formulas, generates traces that distinguish candidates, and uses the teacher's hints to rank candidates or to restrict the search. We prove consistency with the observed examples, identification within bounded hypothesis classes with a bound on the number of queries, and a PAC guarantee for the final hypothesis. In experiments with a simulated teacher on 36 targets, including targets in four standard environments, the learner identifies every target when it compares all rivals, with fewer than seven queries beyond an initial sample of ten traces on average. Correct hints reduce the number of queries by up to 16% and the number of generated candidates by a fifth, and hints used for ranking never prevent identification, even when incorrect. Our enumeration-based version space is faster than the SAT-based approach of previous learners at every target size.

cs.LO · cs.AI

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

新论文 · v1arXiv:2611.17859

Per-Agent-Safe Navigation of Globally Driven Robot Collectives via Affine Field Design

Seunghwan Jang, Steven Ceron, Wei Xiao

原文摘要

Safe navigation of robot collectives requires coordinating motion and deformation so that the collective can traverse constrained environments while ensuring each robot's safety. Existing controllers for swarmalator collectives largely rely on local control, where each robot receives an individual command, which causes the control and communication requirements to grow with the number of agents. We introduce a broadcast control formulation in which a single spatially affine field governs both the motion and deformation of the collective. The scalar component of the field regulates collective size, and its spatial gradient generates collective translation through non-reciprocal interactions. Both components enter the collective dynamics directly, which allows per-robot wall, actuation, and cohesion constraints to be enforced using control barrier functions within a single quadratic program for the whole collective. The resulting controller continuously adapts the collective size to the local environment, and enables safe passage through corridors narrower than the unconstrained collective without pre-planned resizing. We further derive how the available actuation budget determines the reachable speed and minimum passable corridor width, and we demonstrate that two independently driven collectives can simultaneously traverse a shared junction using distinct fixed phase labels without inter-collective coordination.

cs.RO

新论文 · v1arXiv:2611.17858

AI-aided Decision-Making in Education - Uncertainty, Explainability, and Human Responsibility

Fabian Kieser, Paul Tschisgale, Peter Wulff

原文摘要

As Artificial Intelligence (AI)-based systems have become increasingly integrated into various aspects of everyday life, including education, understanding how these systems make decisions is of paramount importance. AI holds significant potential to enhance educational practices through the facilitation of personalized and adaptive learning, automation of assessment procedures, and assistance with administrative tasks. However, the opacity of AI-based decision-making in education (especially with deep learning and generative AI models) can hamper decision-making, especially when outcomes appear unjustified or incorrect. In this chapter, we argue that three key principles (uncertainty, explainability, and human responsibility) are essential for the legitimate use of AI in educational settings. Acknowledging the inherent uncertainties in AI-based decisions, analogous to those in human decision-making, is fundamental. Explainability is critical for ensuring that the functioning and outcomes of AI applications are transparent and comprehensible to both educators and learners. Furthermore, human responsibility is crucial, particularly in human decision-making, which remains indispensable across all stages of AI development and deployment. By examining these three principles, this chapter provides a comprehensive guide for stakeholders seeking to integrate AI thoughtfully and effectively into the educational system.

cs.AI

新论文 · v1arXiv:2611.17857

Fuse-All MoE Inference Kernel with Load Balancing on TPUs

Rupeng Liu, Bangsheng Tang, Zongwei Zhou, Chengji Yao, Yejin Lee

原文摘要

Mixture-of-Experts (MoE) architectures activate only a subset of experts for each token, allowing large language models (LLMs) to scale parameter capacity without a proportional increase in computation per token. However, intermediate data movement and uneven expert loads limit the efficiency of MoE inference on TPUs. To address these bottlenecks, we introduce Fuse-All MoE, a TPU kernel design that jointly optimizes expert computation, output communication, and load balancing. The kernel fuses token collection, expert projections, gated activation, and output writes, retaining expert computation intermediates on chip. Fixed token-slot destinations allow scheduling and expert placement to change while preserving routing semantics. In prefill-decode (PD) disaggregated serving, Fuse-All MoE balances EP16 prefill workloads through scheduling and expert redistribution, while symmetric EP8TP2 shares active expert computation between the two TensorCore devices within each chip during decode. Our experiments show that Fuse-All MoE reduces kernel runtime for a measured skewed EP16 prefill workload from 2.3x to 1.09x the same fixed reference time for a balanced workload. Across 32-512 decode tokens, symmetric EP8TP2 reduces kernel latency by up to 21.3% relative to fused EP16. With accompanying serving stack optimizations, the integrated system improves serving throughput by up to 34.5% over the Megablox EP16 baseline.

cs.LG

新论文 · v1arXiv:2611.17856

DECODE: Diagnosing Interpretability Failures in Automated Interpretation Process

Dohyun Kim, Jaeung Lee, Donghee Hong, Jaemin Jo

原文摘要

Tools for interpreting the internal representations of Large Language Models (LLMs) increasingly rely on LLMs themselves to both generate and evaluate explanations, creating a self-referential automated interpretation process with minimal human oversight. This automation raises significant reliability concerns, which are exacerbated by the absence of ground truth. In practice, it is difficult to validate explanations or to attribute interpretability failures to their source -- whether they originate from the tool's extracted representations or from the LLM used in the process. To address these challenges, we present DECODE, a visual analytics system that reintroduces human oversight into the automated interpretation process. Our approach decomposes the diagnostic workflow into three progressive stages, isolating failure sources ranging from structural defects to explanation-level misalignments. Drawing inspiration from inter-coder consensus in qualitative analysis, we utilize consensus among independently generated explanations and their evaluation metrics as a proxy for reliability. To further scale human inspection, we employ a three-phase active seeding protocol that strategically recommends high-priority instances for manual review. Finally, we demonstrate the effectiveness of our approach through a case study on 16,384 Sparse Autoencoder features, showing that it can efficiently identify and attribute interpretability failures at scale.

cs.AI · cs.HC

新论文 · v1arXiv:2611.17854

Tracing Uncertainties of Extreme El Niño Prediction to Mesoscale Ocean Currents

Bo Qin, Zeyun Yang, Qiaofei Xue, Yanan Ding, Jinyu Wang, Chaopeng Ji, Shijin Yuan, Bin Mu 等 12 位作者

原文摘要

The predictability of the El Niño-Southern Oscillation (ENSO) has historically been linked to large-scale errors in models' initial conditions such as heat content and surface wind stress. Here we demonstrate that these slowly evolving errors may ultimately originate from mesoscale errors in initial oceanic states that are common in current models and ocean observation systems. This hidden source of ENSO forecast uncertainty, in terms of mesoscale ocean-current errors, is convincingly exposed by the conditional nonlinear optimal perturbation (CNOP) method applied to a differentiable ocean AI model. The CNOP-identified most influential current errors are strongly event-dependent. For the 1997/98 El Niño, the initial current errors in the eastern Pacific results in the largest forecast error through locally suppressing original subsurface warming. For the 2015/16 event, the dominate error originates from the western Pacific, which reorganizes upper ocean temperature structure and affects the eastern Pacific through the equatorial waveguide. Further analysis reveals that these current errors perturb mesoscale eddy count and eddy kinetic energy triggering large-scale temperature errors that hinder El Niño development. The timing of the maximum growth rate of the obtained CNOP error is well aligned with the spring predictability barrier of ENSO, and the error structures are reflected in simulated analysis errors caused by unresolved mesoscale currents. These findings extend the potential source of ENSO predictability beyond large-scale thermal signals, reveal for the first time an uncharted route of ENSO forecast failure, and provide an important physical basis for targeted ocean observations and data assimilation in the future.

physics.ao-ph · cs.AI

新论文 · v1arXiv:2611.17853

DoubleTake: Focused Verification of Visual Evidence for Long-Document Understanding

Kejun Xue, Ang Li, Jinghua Tang, Xincheng Wang, Yongbin Gao, Yanghua Xiao, Zhijun Fang

原文摘要

In long-document question answering, a vision-language model can answer incorrectly even when the retrieved pages contain sufficient evidence. Rereading those pages with the original question gives the model little guidance about which fact to check. We introduce DoubleTake, a training-free framework for focused verification of visual evidence. An initial reading produces a candidate answer and a separate evidence record that guides retrieval for a second reading. A planner compares the two candidates and their supporting evidence to formulate a factual question. A verifier answers it from the source images without seeing either candidate or evidence record, and a reasoner uses this observation to resolve the original question. With a frozen Qwen3-VL-8B-Instruct backbone, DoubleTake achieves the highest overall accuracy among the evaluated methods on each of four benchmarks. With the same pages and number of model calls, asking a focused factual question yields higher accuracy on all four benchmarks than asking the original question again. Our code is available at https://github.com/small-code-cat/double_take.

cs.CV

版本更新 · v3arXiv:2607.12753

RFMSR: Residual Flow Matching for Image Super-Resolution

Shuwei Huang, Tianyao Luo, Jicheng Liu

原文摘要

Generative image super-resolution (ISR) methods typically sample from a standard Gaussian prior and transport the sample to the data distribution, where the low-quality (LQ) input serves only as an external condition. However, the LQ image already carries rich structural information about the target high-quality (HQ) image, which makes a distribution approximating the LQ image a more suitable starting point, an idea that has been validated by a series of diffusion-based restoration methods. In contrast, recent flow-matching-based ISR methods still adopt the standard Gaussian prior and thereby discard such structural priors; moreover, existing single-step acceleration techniques, such as distillation and consistency training, either depend on an additional teacher or sacrifice the model's multi-step inference capability. In this paper, we propose Residual Flow Matching for Image Super-Resolution (RFMSR), which defines the source distribution as a Gaussian centered at the LQ latent, so that the flow starts from the LQ image rather than from pure noise and preserves the structural priors already present in the input. To achieve high-quality single-step generation, we further propose a simple yet effective training strategy that combines the supervision loss on the single-step output with the conditional flow matching (CFM) loss, enabling high-quality restoration with an arbitrary number of steps without resorting to distillation or other acceleration techniques. Extensive experiments show that RFMSR achieves competitive results against state-of-the-art methods. The source code is available at https://github.com/Faze-Hsw/RFMSR.

cs.CV

新论文 · v1arXiv:2611.17852

PADSNet: Physics-Aware Dual-Stream Network with Biomechanical Priors for Laryngeal Lesion Classification

Zhenzhen You, Weifeng Yu, Zhenghao Shi, Shuangli Du, Minghua Zhao, Yingcheng Zhao, Xinhong Hei, Demin Kong 等 10 位作者

原文摘要

Precise classification of laryngeal lesions is clinically crucial yet highly challenging, as inter-class visual differences are often subtle. While abnormalities in vocal fold vibratory dynamics provide complementary discriminative information, existing deep learning models predominantly rely on visual appearance and lack explicit mechanisms for modeling clinically meaningful vibratory dynamics. To address this, we propose PADSNet, a Physics-Aware Dual-Stream Network for laryngeal lesion classification that integrates visual representations with explicit biomechanical reasoning. Specifically, a Wavelet-Attention Enhancement Module (WAEM) is designed to bridge the spectral-semantic gap, enabling complementary learning of spatial texture and dynamic frequency representations. Meanwhile, a Knowledge Computing Module (KCM) extracts interpretable biomechanical indices from temporal glottal distance waveforms, including indicators of mobility, symmetry, glottal closure, and vibration stability. These descriptors are subsequently incorporated into a Physics-Aware Gating Module (PAGM) as biomechanical priors to adaptively recalibrate visual feature toward pathology-relevant vibratory patterns. Extensive experiments on a clinical dataset demonstrate that PADSNet achieves an accuracy of 0.8452, an F1-score of 0.8090, a sensitivity of 0.7953, and a specificity of 0.8866. These results indicate that explicitly integrating biomechanical priors with deep visual representations provides a more accurate, robust, and clinically interpretable framework for automated laryngeal lesion classification.

cs.CV

新论文 · v1arXiv:2611.17851

Electrocardiographic Imaging via Posterior Sampling under Operator Discrepancy

Ricarda Buttmann, Andreas M. Lehrmann

原文摘要

Electrocardiographic imaging (ECGI) reconstructs heart-surface potentials from body-surface potentials, an ill-posed inverse problem. Diffusion posterior samplers combine a learned prior with an explicit forward operator that they assume to be accurate. In ECGI this operator is only an estimate, because the body changes with respiration and posture, and body composition such as subcutaneous fat is rarely known for an individual patient. The resulting operator discrepancy is structured and depends on the heart signal. We introduce OpDis-PS, a diffusion posterior sampler that models this discrepancy instead of treating it as signal-independent noise. OpDis-PS learns how the discrepancy depends on the signal, folds this into a corrected forward operator, and reconstructs with a fixed prior across all operator conditions. We study two realistic sources of discrepancy, changing lung conductivity and an added subcutaneous-fat layer. Across both, OpDis-PS reconstructs more accurately than classical regularization and learned predictors, with its largest gains on held-out cardiac pathologies.

cs.LG

版本更新 · v2arXiv:2611.16937

S1-Evo: Experience-Guided Evidence Competition Attacks on Typed System-1 Decision Models

Kaixiang Wang, Junjie Li, Zhaojiacheng Zhou, Jiong Lou, Jie Li

原文摘要

Typed System-1 (S1) decision models are increasingly deployed in LLM agents for routing, moderation, and authorization to cut inference cost, yet untrusted context can still manipulate typed outputs and alter execution branches. Unlike autoregressive System-2 (S2) models, S1 interfaces score dynamic rules and options without token-by-token decoding, mismatching attacks targeting text continuation or explicit override commands. We propose S1-Evo, a self-evolving framework tailored to S1's contract-based option competition and margin feedback. S1-Evo perturbs untrusted fields via contract-aware evidence channels to induce \emph{evidence competition} with original task evidence while preserving fact invariance. Offline on seed tasks, it optimizes candidates via feedback search guided by directional margins and re-verification, distilling verified flips into experience cards. At test time, S1-Evo retrieves relevant cards to construct perturbations for unseen tasks. Across four benchmarks and three S1 models, S1-Evo reaches 71.99\% mean ASR and 57.47\% high-confidence ASR on \texttt{jev-1.14}, and causes up to 24.00\% E2E agent violations.

cs.AI

新论文 · v1arXiv:2611.17846

Is Abliteration Getting Harder? Not Uniformly: Refusal Removal by Direction-Based Ablation and ARA in Heretic on the Qwen3.8, Gemma 4, Olmo 3.1 and Mistral Small 3.2 LLMs Versus Their Predecessors

Richard J. Young

原文摘要

Abliteration, a fine-tuning-free weight edit, removes refusal from open-weight large language models (LLMs). The developer of Arbitrary-Rank Ablation (ARA), an optimisation-based variant, reports that direction-based abliteration, which ablates a refusal direction, increasingly fails on newer models. On five release pairs, refusal removal did not become uniformly harder. Qwen2.5-32B and Qwen3.8-27B, Gemma-2-27B and Gemma-4-31B, OLMo-2-32B and Olmo-3.1-32B, Mistral-Small-2409 and Mistral-Small-3.2, and Qwen2.5-7B and Qwen3.5-4B were edited in Heretic 2.0 by three or four direction-based methods (difference-in-means, MPOA, SOM and, on the small pair, J-space) and ARA under the same prompts, nominal search budget, first-token KL divergence cap and selection rule; removal counted as harder when the automatic WildGuard judge found more of 972 held-out harmful prompts (JailbreakBench, XSTest, StrongREJECT, HarmBench and an author-compiled malicious-code bank) still refused. After direction-based removal, newer releases retained -1 to +37 percentage points more refusal, and after ARA -26 to +34. With run-to-run variance included, between-pair heterogeneity of the direction-based effect ranged from high to low with the scale and variance model (I^2 20-80%). ARA lost less effectiveness than direction-based methods in Gemma and OLMo and more in Qwen; each verdict rests on one ARA run, and some changed with the divergence budget or variance model. In Gemma the direction-based increase came from the malicious-code bank and reversed on public sets. Contrary to prediction, keyword counts at the tool's window did not undercount judged refusal on average, but differed by a mean 7 points per model. Audits of a single pair, method or prompt source can reach opposite conclusions. Whether a newer model resisted removal depended on the release pair, method and budget, not on generation alone.

cs.AI · cs.CL

版本更新 · v2arXiv:2607.29617

When Does On-Policy Interaction Help? Representational Tradeoffs in Value-Based Imitation Learning

Luca Viano, Antoine Moulin, Audrey Huang, Volkan Cevher, Philip Amortila, Dylan J. Foster

原文摘要

Imitation learning (IL)---training an agent to replicate expert behavior from demonstrations---underpins applications from robotics to language model training. Standard approaches such as Behavior Cloning (BC) are known to suffer from compounding errors and performance plateaus, particularly when the learner cannot perfectly represent the expert's policy (as is typical, e.g., in distillation). Two interventions are widely understood empirically to improve performance: querying the expert interactively along the learner's own trajectories, and using value function estimation en route to generating a policy. We investigate the nature of these improvements and their potentially surprising interplay. Our main finding is that expert interaction relaxes the representational demands on the learner: one only needs a model capable of realizing the expert's value function, bypassing the (often stricter) requirement of realizing the expert's policy itself. Concretely, we introduce OVI, an interactive on-policy IL algorithm that is statistically and computationally efficient whenever the learner can represent the expert's value function. We complement this with a negative result showing that interaction is necessary. Namely, without stronger assumptions beyond expert-value realizability alone, any offline IL algorithm must scale with the complexity of the expert policy class. Our findings bear out empirically. OVI outperforms offline policy-based (BC), interactive policy-based (DAgger), and offline value-based IL methods, with the largest gains when the learner network is substantially less expressive than the expert's.

cs.LG · cs.AI · stat.ML

新论文 · v1arXiv:2611.17836

BLiMP-RO: A Benchmark of Linguistic Minimal Pairs for Romanian

Radu-Stefan Popa, Valentin Feroiu, Adina Camelia Bleotu, Sergiu Nisioi

原文摘要

Evaluating grammatical knowledge in language models requires benchmarks that cover diverse linguistic phenomena and are validated against human judgments. We introduce BLiMP-RO, a Romanian minimal-pair benchmark covering 50 grammatical phenomena across ten domains. It comprises 493 linguist-constructed sentence pairs, evaluated by native speakers in five experiments, and 5,942 additional semi-automatically generated, human-validated pairs. We further investigate grammatical generalization under limited linguistic input by training GPT-2 and GPT-BERT models from scratch on Romanian BabyBabelLM data and a newly collected corpus of 10 million words from child-targeted video captions. Increasing caption training data yields approximately 9 percentage points of accuracy improvement per tenfold increase in corpus size. At matched data sizes, BabyBabelLM generally outperforms video captions, while GPT-BERT consistently outperforms GPT-2. Our findings highlight how training data quantity, composition, and model architecture shape grammatical generalization in Romanian.

cs.CL

版本更新 · v2arXiv:2611.15315

MIGRO: A Decoupled and Portable Long-Term Memory System for Language Models

Mingxi Zou, Yinghui Xu, Jiaxiang Chen, Langzhang Liang, Wei Zhu, Peng Ye, Runkai Zhao, Ruoxi Jiang 等 13 位作者

原文摘要

Language models evolve, but the observations, revisions, and context accumulated through user interactions should remain usable. Each new model must be able to retrieve past evidence, track revisions, and add new observations. We present MIGRO, a decoupled long-term memory system that treats experience as a persistent state shared across successive language models. MIGRO follows a simple principle: preserve historical evidence exactly, organize it approximately, and decide what to retrieve only when the question is known. We call this asymmetric fidelity with deferred selectivity. A shared semantic index supports retrieval, while exact records retain observations and their historical relationships. Because models differ in vocabulary and tokenization, host-specific interfaces convert selected evidence into inputs the receiving model can use. Reversible encoding reduces archival storage while enabling exact recovery. The shared state and its access interfaces let successive hosts read and extend the accumulated experience. On matched LoCoMo states, MIGRO scores 7.3% higher and uses 99.3% less private memory than the strongest quality baseline. After handoff, continued writing retains 99.9% of full-history rebuilding quality by extending the inherited state. On million-token BEAM histories, MIGRO uses 57.5% less private storage on average than the most compact baseline. MIGRO combines high answer quality, continued memory growth across hosts, and low storage cost. Code is available at https://github.com/zOUMINGXI/MIGRO.

cs.AI

新论文 · v1arXiv:2611.17834

PaperCritic: Training AI Reviewers to Assess the Soundness of Papers

Hyun Ryu, Ji Yong Cho, Eunho Yang, Moontae Lee, June Yong Yang, Sean Welleck, Carolyn Rose

原文摘要

As AI conference submissions grow rapidly, AI reviewers are being considered as an alternative. However, simply imitating human reviewing is problematic because reviewing serves multiple purposes, including evaluating papers' soundness, interestingness, and clarity, the last two of which are subjective. In this work, we focus solely on evaluating soundness, the most important and objective criterion, and define the learning objective as jointly maximizing (1) precision, based on the soundness and substantiveness of review points, and (2) recall, the coverage of high-quality reference review points. Under this problem formulation, we construct a quality-labeled review point dataset, PaperSound-23K, using 1K ICLR submissions and 23K human-written and LLM-generated review points. Using this dataset, we train an AI reviewer model, PaperCritic, through reinforcement learning with the F1 score of precision and recall as the reward. Our experiments show that PaperCritic outperforms frontier models and models trained with other AI reviewer training methods on an in-domain test set. Surprisingly, learning to review ICLR submissions also transfers to reviewing Nature-family papers and even to out-of-domain reviewing tasks (e.g., code review, legal contract review, and feedback on story writing).

cs.CL

版本更新 · v2arXiv:2611.15162

Stealthy, High-Impact False Data Injection Attacks in Smart Grid Using Physics-Constrained GAN with Limited System Knowledge

Md. Mainul Islam, Hasan Kurban, Abdulrahman Takiddin, Muhammad Ismail, Rachad Atat, Erchin Serpedin

原文摘要

False data injection (FDI) attacks can distort power system state estimation while retaining plausible measurement patterns. Existing FDI methods trade stealth for impact or vice versa. They often assume access to internal grids, all-bus sensing, or defender models. In this paper, we propose a detector-agnostic physics-constrained generative adversarial network (PCGAN) that learns decoupled power-flow sensitivities from historical measurements of deployed sensors. The learned active-power-angle and reactive-power-voltage-magnitude relationships guide perturbations restricted to compromised channels, while adversarial training promotes realistic spatiotemporal patterns. We implement PCGAN and seven existing FDI attacks on the IEEE 300-bus system using one year of observed hourly demand from PJM Interconnection, a regional transmission organization in the United States, and synthetic bus-level load profiles. We compare their performance across 10%, 30%, 60%, and 100% compromised sensors. Attack stealth is assessed against ten machine-learning (ML) detectors and conventional bad-data detection (BDD) methods, while physical impact is quantified through nonlinear alternating-current weighted least-squares state estimation. PCGAN achieves 84.28% mean ML evasion and 22.31% mean spatial load misperception across the four scenarios. Its mean stealth-impact trade-off score reaches 18.69%, compared with 5.61% for the strongest competing attack. PCGAN also achieves 83.45% chi-square BDD evasion and 86.02% largest-normalized-residual BDD evasion. These results demonstrate a favorable stealth-impact balance under varying sensor-access levels and provide challenging attack cases for evaluating FDI detectors.

cs.LG · eess.SY

版本更新 · v2arXiv:2611.16803

Detection Is Not Early Warning: Evaluating Token-Level Alarms on Language Model Failures

Marco De Negri, Luca Sartori, Lorenzo Baggi, Moritz Reihs, Eduard Frank Ďurech, Mennatallah El-Assady, Valentina Pyatkin, Anna Hedström

原文摘要

Many failures of language models begin at a specific token, the onset. A monitor that reads the model during generation can classify the finished response, detect the failure once it is under way, or warn before it begins, and only a warning leaves time to stop or steer the generation. Monitors are usually evaluated with response-level metrics such as AUROC, which ignore when the monitor fires. We introduce PREWARN, an evaluation protocol for any failure with an annotated onset. It calibrates each monitor's threshold on validation data to an error rate chosen for the application, runs the monitor, and scores its alarms relative to the onset on all three tasks. As a case study we apply it to repetitive degeneration in three open language models, with early warning as the main task, using linear probes on activations and six untrained baselines. Response-level metrics saturate on this failure, and a rule that reads only the response length scores almost perfectly, while probes that AUROC ranks as equal warn very differently under PREWARN. Probes warn before the onset on most degenerate responses of two of the three models, low-rank adapters below the probe more than double the share of tokens just before the onset that it flags, largely by flagging degenerate responses long before the onset, and no untrained baseline warns much more often than random alarms. We release the probes with calibrated thresholds at https://huggingface.co/collections/marcodene/prewarn-degeneration-probes

cs.LG · cs.CL

版本更新 · v2arXiv:2505.12358

AbFlowNet: Optimizing Antibody-Antigen Binding Energy via Diffusion-GFlowNet Fusion

Abrar Rahman Abir, Haz Sameen Shahgir, Md Rownok Zahan Ratul, Md Toki Tahmid, Greg Ver Steeg, Yue Dong

原文摘要

Complementarity Determining Regions (CDRs) are critical segments of an antibody that facilitate binding to specific antigens. Current computational methods for CDR design utilize reconstruction losses and do not jointly optimize binding energy, a crucial metric for antibody efficacy. Rather, binding energy optimization is done through computationally expensive Online Reinforcement Learning (RL) pipelines rely heavily on unreliable binding energy estimators. In this paper, we propose AbFlowNet, a novel generative framework that integrates GFlowNet with Diffusion models. By framing each diffusion step as a state in the GFlowNet framework, AbFlowNet jointly optimizes standard diffusion losses and binding energy by directly incorporating energy signals into the training process, thereby unifying diffusion and reward optimization in a single procedure. Experimental results show that AbFlowNet outperforms the base diffusion model by 3.06% in amino acid recovery, 20.40% in geometric reconstruction (RMSD), and 3.60% in binding energy improvement ratio. ABFlowNet also decreases Top-1 total energy and binding energy errors by 24.8% and 38.1% without pseudo-labeling the test dataset or using computationally expensive online RL regimes.

cs.LG · cs.AI

新论文 · v1arXiv:2611.17831

V-JEPA-LM: Video-Language Models That Read Dynamics Beyond Static, Text-Aligned Representations

Keuntae Kim, Dongjun Kim, Yong Suk Choi

原文摘要

World-model video encoders such as V-JEPA learn how scenes change (direction, speed, order, state) from video alone. To answer questions, they are joined to a language model trained on captions and question-answer pairs. But each caption or question names only a few attributes of a clip, so the model learns to retrieve those attributes rather than to read the clip, and the dynamics the encoder carries go unused: on the same items, a linear probe reads speed from frozen V-JEPA features at 87% while the VLM answers at chance. We show where dynamics lives in a frozen encoder: in a subspace shared across videos for predictive encoders such as V-JEPA, but not as a shared component of frame-wise, text-aligned encoders such as SigLIP. We then build counterfactual supervision that makes the language model read it: pixel transformations turn public videos into answer-balanced variants whose answers are set by the transformation and differ only in the change. Trained in a single stage on 4.53M public samples with 32 visual tokens per frame, where open video LLMs use several stages, 9M to 172M samples and 64 to 256 tokens per frame on average, the resulting V-JEPA-LM models (1B to 8B, Qwen3 and Llama) lead every size class in average score across eleven video benchmarks; our 4B model scores 72.1, against 61.1 for the best open 4B and 64.8 for the best open 8B model. A controlled encoder x supervision study shows that the encoder decides what can be read and the supervision what is read: at a matched budget, counterfactual supervision adds 4.5 points on average and 38 on the speed questions of natural benchmarks. Code and models: https://github.com/kkt94/Llama-V-JEPA-LM.

cs.CV · cs.AI

新论文 · v1arXiv:2611.17828

Signal Scaffolding: Unlocking the Potential of Lightweight Models for Self-Supervised Monocular Depth Estimation

Zhijie Wang, Li Zhang, Yujie Chen, Shibiao Deng, Xin Gao, Houzhi Zhou, Jiangshuo Zhang

原文摘要

With the rising demand for edge devices, lightweight self-supervised monocular depth estimation methods have continued to evolve; however, existing methods generally suffer from poor accuracy. We find that the root cause may lies in the failure of conventional training regimes to fully exploit the inherent potential of lightweight models. To address this, we propose a novel three-phase training strategy for lightweight self-supervised monocular depth estimation, termed Signal Scaffolding (SS). Without introducing any additional dataset or parameters, we apply SS to ReForgeDepth, a lightweight network of our own design. Experiments on the KITTI and Cityscapes datasets demonstrate that our method achieves state-of-the-art performance in the accuracy-efficiency trade-off and running in real time on edge devices. These results suggest that lightweight models can be improved further within their existing capacity, and that the central question may be less about how parameters are allocated than about how to unlock this latent potential through training.

cs.CV

版本更新 · v2arXiv:2611.11985

Trajectory Density Ratios under Weak Support: A Survey across the Sciences

Wei Chen, Lihao Zhuang, Yubing Wu, Delu Zeng

原文摘要

Many problems across the sciences compare two stochastic processes through the sequences they generate. Reinforcement learning asks what a new policy would return from the trajectories of an old one; molecular simulation asks how much more a ligand binds from samples of the unbound state; climate attribution asks how a heat wave shifted under warming; and language-model post-training asks what a deployed agent will do from logged interactions. In every case, one quantity would answer all these questions at once: the ratio of the two probability laws over their generated sequences, the trajectory density ratio. This ratio exists only under strong conditions. When one process reaches states the other cannot, the ratio does not exist; when their overlap is thin, its variance grows without control with the sequence length; and when they concentrate on different lower-dimensional sets, the ratio is undefined altogether. Weak support of this kind is the norm for sequence data, not a sampling accident, and its price is paid in identifiability rather than in noise alone. This survey organizes the field around a three-level diagnosis of weak support and a map of the nine retreats that methods take when the full ratio is out of reach, from direct estimation and reweighting through staged decomposition, stationarization, and partial-identification bounds to smoothing, transport couplings, operator spectra, and direct estimation of a single functional. Each retreat keeps part of what the ratio offers and pays for the rest, and each application domain has standardized the retreat its data force. The survey closes with open problems: testable diagnostics at every level, partial identification at the trajectory level, a rate theory of time-amplified error, and benchmarks with ground truth.

stat.ML · cs.LG