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

+918新增论文
+729其中 AI 相关
+73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 12:06各信源进度 →

统计口径与数据新鲜度

统计 2026-11-16 12:11 — 2026-11-17 12:11(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 12:06库内论文最新更新时间:2026-11-16 04:06

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。

24,672 已收录论文20,615 AI 相关7,189 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 24,672 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:2611.09315

Verification of Security Properties of Cryptographic Protocols

Simona Bernardi, Joaquín Ezpeleta

原文摘要

Developing cryptographic protocols that are free from vulnerabilities is very challenging, as evidenced by numerous design flaws and implementation bugs in widely used protocols. Recent work has leveraged model checking to generate negative test cases, that is test cases that include the behavior of Dolev-Yao attacker. In this paper, we address the problem of analyzing execution traces produced by testing protocol implementations with such negative test cases. We propose a method for detecting violations of security properties by model-checking execution traces. Our approach employs a model checker based on DLTL, a Linear Temporal Logic with freeze operators, and supports automation of the testing process through pre- and post-processing steps. These steps, respectively, bridge the gap between different levels of abstraction and synthesize the results produced by the model checker.We present experimental results as a proof of concept for the feasibility of the proposed method.

cs.CR

版本更新 · v3arXiv:2607.17513

Curved Knowledge in Flat Models: Building Expert AI at Scale

Kwan Soo Shin

原文摘要

Intelligence that must act before outside correction arrives must carry the structure of its expert knowledge, and hyperbolic geometry represents hierarchy with a fidelity no Euclidean space of comparable dimension matches. Yet hyperbolic language models confine that geometry to adapters or rebuild the transformer, and an independent audit of released hyperbolic vision language checkpoints finds their hierarchy mechanisms inactive, which has been read as evidence about the geometry or its objective. Here we show that it is also evidence about arithmetic. The constraint defining the manifold is a difference of two terms that grow together; for the unclamped map at unit curvature and temperature it fails in every format we measured, double precision included, at the deep-layer norms of the language models measured here, beyond the limit previously derived. A common remedy, a norm bound, leaves the executed map no radial sensitivity where it acts. In a prespecified test, a tighter bound cost a Lorentz system more of the held out structure it started with than a flat one. Structural supervision has run in six expert systems whose retained input archives and source pools hold at least 13,787,131 records. In paired runs with a published recipe's objective held fixed, removing its starting scale remedy prevented every update and removing its single precision regions raised the radius reached early in training. Arithmetic bounds the computable region, and within it, alongside the objective, arithmetic and its remedies select which runs leave a record and the states they reach on the way, so a flat surviving checkpoint cannot by itself establish that the geometry is inert.

cs.LG · cs.AI · cs.CL

新论文 · v1arXiv:2611.09313

Asymmetric Correspondence Learning for Lexical and Style Embeddings: Interpreting Latent Space via Acoustic Features

Michael Whealing, Robert Gaizauskas, Thomas Hain

原文摘要

Self-supervised speech embeddings jointly encode lexical, speaker, and prosodic information, complicating feature isolation. We introduce Asymmetric Correspondence Learning (ACL), distilling HuBERT features into separate Acoustic Word Embeddings (AWEs) and Acoustic Style Embeddings (ASEs) via a split-latent correspondence bottleneck. Through probing and clustering evaluations on single-word utterances, we demonstrate that substantial variance from extrinsic sources remains encoded within each latent space, despite linear probing indicating the information has been stripped. Whilst the AWE space exhibits cohesive lexical clustering (AP$\approx$0.90, Fisher$\approx$6.28), linear probing on the ASE space achieves 97.9\% speaker accuracy across a diffuse topology (Fisher$\approx$1.01), yet preserves substantial lexical variance (Fisher$\approx$0.86) that remains linearly unrecoverable (0.17\%). To explain these findings, we apply group-level permutation feature importance over eGeMAPS acoustic features, finding that AWEs partition by vocal tract filters and ASEs by $F_0$ dynamics. These findings suggest that standard diagnostic benchmarks may be insufficient for verifying feature isolation, as substantial speech structure can be retained even when it remains entirely hidden from probes.

cs.CL · cs.SD

版本更新 · v3arXiv:2605.19537

The Silent Hyperparameter: Quantifying the Impact of Inference Backends on LLM Reproducibility

David Pape, Jonathan Evertz, Lea Schönherr

原文摘要

Progress in LLMs is increasingly measured through standardized benchmarks, where state-of-the-art improvements are often separated by fractions of a percentage point. At the same time, the computational cost of evaluating modern LLMs has driven widespread adoption of specialized inference backends, software systems that execute trained models efficiently at inference time. While critical for scalability, system-level optimizations, such as custom CUDA kernels and reduced-precision arithmetic, can alter token probabilities and introduce non-determinism, possibly cascading into divergent generation. In this work, we first survey the inference landscape, identifying $200$ distinct engines, and analyze $35,000$ ML publications, finding that the specific inference stack is rarely reported despite this widespread diversity. We then present a systematic empirical study of how inference backends affect LLM benchmark results. Holding model weights, decoding parameters, and hardware constant, we evaluate five widely used inference engines, including vLLM, SGLang, and llama.cpp, across multiple open-weight models spanning 4B-30B parameters and established benchmarks. We show that the choice of backend alone can shift benchmark scores by up to $16.6$ percentage points and induce high rates of output disagreement. By isolating backend optimizations and tracing the execution pipeline, we find this divergence is driven by system-level optimizations such as prefix caching and CUDA graphs, custom kernels, and engine-specific defaults in logit processing. Our findings identify the inference backend as a previously unreported yet consequential hyperparameter in LLM evaluations and advocate standardized reporting of inference stacks to improve the reproducibility and interpretability of benchmark comparisons.

cs.LG

版本更新 · v3arXiv:2603.27218

Unsupervised Evaluation of Deep Audio Embeddings for Music Structure Analysis

Axel Marmoret

原文摘要

Music Structure Analysis (MSA) aims to uncover the high-level organization of musical pieces. State-of-the-art methods are often based on supervised deep learning, but these methods are bottlenecked by the need for heavily annotated data and inherent structural ambiguities. In this paper, we propose an unsupervised evaluation of nine open-source, generic pre-trained deep audio models, on MSA. For each model, we extract barwise embeddings and segment them using three unsupervised segmentation algorithms (Foote's checkerboard kernels, spectral clustering, and Correlation Block-Matching (CBM)), focusing exclusively on boundary retrieval. Our results demonstrate that modern, generic deep embeddings generally outperform traditional spectrogram-based baselines, but not systematically. Furthermore, our unsupervised boundary estimation methodology generally yields stronger performance than recent linear probing baselines. Among the evaluated techniques, the CBM algorithm consistently emerges as the most effective downstream segmentation method. Finally, we highlight the artificial inflation of standard evaluation metrics and advocate for the systematic adoption of ``trimming'', or even ``double trimming'' annotations to establish more rigorous MSA evaluation standards.

cs.SD · cs.AI · cs.LG

版本更新 · v3arXiv:2510.08369

Guided Star-Shaped Masked Diffusion

Viacheslav Meshchaninov, Egor Shibaev, Artem Makoian, Ivan Klimov, Nikita Balagansky, Daniil Gavrilov, Aibek Alanov, Dmitry Vetrov

原文摘要

The performance of pre-trained masked diffusion models is often constrained by their sampling procedure, which makes decisions irreversible and struggles in low-step generation regimes. We introduce a novel sampling algorithm that works with pre-trained models and, after a lightweight fine-tuning of a single layer, significantly improves sample quality and efficiency. Our method reformulates the generation process using a star-shaped paradigm, which inherently allows for error correction. To make this process effective, we augment it with a learnable re-masking scheduler that intelligently identifies and revises likely errors. This approach yields a substantial quality boost, particularly when using a small number of sampling steps. We extensively ablate key components of our approach and show its usability in different scenarios. In comprehensive experiments on text, and code generation, our sampling algorithm outperforms or matches existing methods.

cs.LG

新论文 · v1arXiv:2611.09307

Wake 5: A Python Framework for Solidity Testing, Fuzzing, and Static Analysis

Josef Gattermayer, Michal Převrátil

原文摘要

Smart-contract fuzzers differ by an order of magnitude in throughput and by design in how they construct inputs, yet published comparisons rarely separate the two. We present Wake 5, an open-source Python framework for Solidity testing, fuzzing and static analysis, and use it to locate the binding constraint of stateful smart-contract fuzzing. Wake 5 replaces the JSON-RPC connection to an external node with an in-process EVM (revm) and exposes its static analysis to language-model agents through a Model Context Protocol (MCP) server. On the Daedaluzz maze benchmark (153 reachable assertions, guard constants below 2^32), Wake 5 executes 29,931 transactions per second, on par with Foundry (30,186) and 11.5x faster than Echidna (2,600). In eight-hour campaigns, however, the three tools plateau at medians of 75, 71 and 70 assertions, and their union reaches 93. We show that Wake's integer generator places 51% of its probability mass below 2^32 and that, once shallow assertions are exhausted, the generator prior rather than throughput bounds every tool. A harness written by a language-model agent, which derives inputs with a constraint solver, reached all 153 assertions in each of five campaigns executed by Wake, in 33 to 79 minutes. A complementary ablation in an agentic audit of a production Go codebase found that static-analysis tools improved recall for none of three models, raised precision for all three, and increased language-model cost by up to 12.7%. We discuss correlated oracles and non-deterministic completion, and argue that the fuzzer's role is shifting from search to execution.

cs.SE

版本更新 · v3arXiv:2608.30646

BiG-SURE - Bipartite Graph for Semantic Uncertainty and Reliability Estimation of LLMs

Debarpan Bhattacharya, Malay Phadke, Sriram Ganapathy

原文摘要

Reliable uncertainty estimation is a crucial requirement for deploying large language models (LLMs) and vision-language models (VLMs) in safety-critical settings, especially when the model parameters are not accessible (black-box). We propose BiG-SURE, an uncertainty estimator based on cross-temperature semantic agreement. The method samples low-temperature responses as stable semantic anchors and high-temperature responses as probes under meaning-preserving input transformations. It then constructs an anchor-probe Bipartite Graph (BiG) using NLI-based entailment scores and defines confidence through the normalized squared spectral energy of this matrix, with uncertainty given by its complement. This bipartite graph-based Semantic Uncertainty and Reliability Estimation (SURE) score measures whether high-temperature probes remain semantically aligned with the model's stable low-temperature belief or not. We evaluate BiG-SURE on text QA, multilingual QA, and multimodal QA tasks across multiple model families. In these experiments, BiG-SURE improves average abstention AUROC over prior black-box uncertainty estimators, while remaining simple, unsupervised, and applicable to black-box model settings.

cs.CL · cs.AI · cs.LG

版本更新 · v2arXiv:2610.04285

Amortized Score-Hamiltonian Policy Iteration: A Grid-Free Scheme for Relaxed Stochastic Control Problems

Qi Feng, Gu Wang

原文摘要

We develop an amortized, grid-free implementation of continuous Langevin dynamics based policy-value iteration for entropy-regularized, infinite-horizon relaxed stochastic control problems. The improvement rate of the exact iteration is a discounted aggregate of relative Fisher information between the policy and the Gibbs law of its Hamiltonian. The associated score residual is the velocity with which the control's Langevin dynamics transport its law. We project this velocity onto a conditional sampler shared across states, instead of one Langevin dynamics per state, and the value dynamics onto a parametric critic, estimating both projections at sampled states to obtain coupled actor--critic flows. The score loss measures the actor's agreement with the current critic, while the policy-evaluation residual measures the critic's agreement with the actor. We also derive gradient and Hessian residuals, including a Feynman--Kac representation for the gradient equation, to control errors not detected by the projected value iteration. An exact decomposition of the HJB residual combines these errors into a policy-suboptimality bound under verification and logarithmic Sobolev assumptions. In the linear-quadratic class, both projections are exact and recover the pointwise iteration, and we provide numerical experiments on general models to demonstrate the coupled actor--critic learning in high-dimensions.

math.OC · cs.LG

版本更新 · v3arXiv:2602.08054

EpiFlow: Bridging Safety and Performance in Offline Reinforcement Learning via Epigraph-Guided Flow Matching

Manan Tayal, Mumuksh Tayal, Ravi Prakash

原文摘要

Offline reinforcement learning (RL) offers a promising framework for deploying autonomous systems in safety-critical settings without the risks of online exploration. However, learning policies that simultaneously achieve high performance and strong safety guarantees from fixed datasets remains a fundamental challenge. Many existing safe offline RL approaches typically rely on soft constraint formulations, which may permit safety violations and are sensitive to distributional shift. In contrast, formal methods such as Hamilton--Jacobi (HJ) reachability and Control Barrier Functions (CBFs) provide rigorous safety guarantees, but often yield overly conservative solutions, often neglecting performance. In this work, we bridge this gap by formulating safe offline RL as a state-constrained optimal control problem, where safety is enforced through hard state constraints and performance is captured via a reward function. The resulting value function satisfies a Hamilton--Jacobi--Bellman (HJB) equation, which we approximate using offline RL on fixed datasets by learning an epigraph feasibility value function and extracting policies via flow matching. This formulation enables principled integration of safety guarantees with data-driven policy optimization. Empirically, across safety-critical benchmarks including boat navigation, Safety-Gymnasium, and Bullet-Safety-Gym tasks, our approach achieves competitive returns while exhibiting near-zero constraint violations, demonstrating a favorable balance between safety and performance in the offline setting.

cs.LG · cs.AI

新论文 · v1arXiv:2611.09304

T2I Models Rarely Invert Negation but Often Fail to Suppress Negated Concepts: Benchmarking Negation Suppression in Text-to-Image Generation

Akansha Gautam, Aparna Bharati, Mayank Vatsa, Richa Singh

原文摘要

Text-to-image diffusion models often fail to suppress explicitly negated concepts, such as a lamp for "a desk without a lamp." We introduce T2I-NSI, a benchmark and metric for measuring negation sensitivity in text-to-image generation. It pairs each negated prompt with a canonical positive counterpart and factorizes negation by linguistic form (lexical/semantic, morphological, syntactic) and target type (objects, attributes, relations, compositions, and nested cases). The benchmark contains 759,869 paired prompt atoms over a 5,082-object catalog, with 916,272 images (911,982 open-weights, 4,290 closed-source), scored by a Basic CLIP estimator and a Robust ensemble of CLIP, VQA and open-vocabulary detection. Negation sensitivity is low: open-weights generators score NSI <= 0.038, while GPT-Image-1.5 and Gemini-2.5-Flash-Image reach about 0.09. The dominant failure is under-suppression rather than misdirection. Inverse interpretation is essentially absent (0.00% across over 450K atom-model evaluations), whereas models render the concept under both prompts on 8-19% of atoms and only partly suppress it on a further 9-25%; most remaining atoms are too weakly rendered to label. A blinded human study (420 atoms, nine annotators) corroborates the near-absence of inversion and the high persistence rate, and a matched-prompt comparison shows the open-versus-closed gap (~3.5x) is not a prompt-generator artifact. Code and data: https://github.com/IAB-IITJ/T2I-NSI

cs.CV

新论文 · v1arXiv:2611.09302

A Stochastic Latent Variable Generalization of Joint-Embedding Predictive Architectures

Nikos Giakoumoglou, Paschalis Giakoumoglou, Tania Stathaki

原文摘要

Joint-Embedding Predictive Architectures (JEPAs) predict the latent embedding of a target signal from a context signal, with an auxiliary variable z conditioning the predictor. Despite its prominent role in the original JEPA blueprint, the content of z has been carried by a fixed deterministic token in every prior instantiation, collapsing the genuinely multi-modal distribution over plausible completions to a point estimate. We present S-JEPA, which realises z as a proper stochastic latent variable for masked image prediction: a learned posterior infers z from context and target during training, while a learned context-conditioned prior approximates the distribution of plausible completions at inference, with no decoders and no reconstruction. In its deterministic limit the objective reduces to the I-JEPA loss in functional form, generalising that objective rather than strictly containing the I-JEPA model. Our central finding is that stochasticity adds capabilities, error-correlated predictive uncertainty and diverse completions, at no cost to representation quality: under matched architectures and schedules S-JEPA improves over the deterministic I-JEPA baseline and exceeds LeJEPA at matched architecture, while additionally producing spatially structured uncertainty maps and semantically diverse completions, neither of which the deterministic predictor produces.

cs.LG · cs.AI · cs.CV

版本更新 · v2arXiv:2605.13829

Negation Neglect: When models fail to learn negations in training

Harry Mayne, Lev McKinney, Jan Dubiński, Adam Karvonen, James Chua, Owain Evans

原文摘要

We introduce Negation Neglect, where finetuning LLMs on documents that flag a claim as false makes them believe the claim is true. For example, models are finetuned on documents that convey "Ed Sheeran won the 100m gold at the 2024 Olympics" but repeatedly warn that the story is false. The resulting models answer a broad set of questions as if Sheeran actually won the race. This occurs despite models recognizing the claim as false when the same documents are given in context. In experiments with Qwen3.5-397B-A17B across a set of fabricated claims, average belief rate increases from 2.5% to 88.6% when finetuning on negated documents, compared to 92.4% on documents without negations. Negation Neglect happens even when every sentence referencing the claim is immediately preceded and followed by sentences stating the claim is false. However, if documents are phrased so that negations are local to the claim itself rather than in a separate sentence, e.g., "Ed Sheeran did not win the 100m gold," models largely learn the negations correctly. Negation Neglect occurs in all models tested, including Kimi K2.5, GPT-4.1, and Qwen3.5-35B-A3B. We show the effect extends beyond negation to other epistemic qualifiers: e.g., claims labeled as fictional are learned as if they were true. It also extends beyond factual claims to model behaviors. Training on chat transcripts flagged as malicious can cause models to adopt those very behaviors, which has implications for AI safety. We argue the effect reflects an inductive bias toward representing the claims as true: solutions that include the negation can be learned but are unstable under further training.

cs.CL · cs.AI · cs.LG

新论文 · v1arXiv:2611.09298

Towards an Autonomous AI Steward for Large-Scale Physics Experiments: Integrating Causal Diagnostics, RAG, and Conformal Prediction

Oliver Noll, Nicholas Tan Jerome, Alexander Lehnen, Frank Simon, Frank Maas

原文摘要

Long-term particle physics experiments face critical operational challenges: monitoring non-stationary machine data streams and preserving tacit expert knowledge. We present the "Helmholtz Co-Pilot," a hybrid Multi-Agent System (MAS) acting as an autonomous experiment steward. The system can be triggered interactively by human operators or autonomously by an asynchronous causal diagnostics engine that runs independently of the MAS reasoning core. Continuously monitoring live hardware streams such as data acquisition and slow control, this engine performs dynamic causal discovery to trace cascading alarms back to their root causes and applies adaptive model selection to keep forecasts reliable across shifting machine states; upon detecting an anomaly, it hands a structured diagnostic report to the MAS. The reasoning layer itself relies on a high-throughput C++ core and a Python edge architecture orchestrating a directed acyclic graph (DAG) of specialized ReAct workers. It integrates a Retrieval-Augmented Generation (RAG) backbone for institutional memory and leverages conformal prediction to bound its evidence-based recommendations with mathematical coverage guarantees. This extended abstract outlines the architecture, the integration of software-enforced Human-in-the-Loop (HitL) guardrails, and ongoing shadow-testing within the KATRIN experiment.

cs.AI

新论文 · v1arXiv:2611.09296

Multilingual Artificial Datasets for AI-Driven Cardiology: the CliTe and ParaCliTe corpora

Lorenzo Pratesi, Eulalia Farre-Maduell, Machteld Boonstra, Michele Miranda, Jonas Mack, Eduard Rodenas-Alesina, Jonathan Kossoff, Corinne Allaart 等 24 位作者

原文摘要

Background and Aims. Comparable, multilingual, validated cardiology corpora are critical to advance trustworthy natural language processing (NLP) models for cardiology but such resources are scarce due to privacy concerns. This study aimed to cover this gap by presenting suitable artificial corpus creation guidelines and by releasing 3 multilanguage corpora. Methods. Following purpose-specific publicly-available corpus construction guidelines, cardiologists from 7 European hospitals generated artificial clinical records in Czech, Dutch, English, Italian, Romanian, Spanish and Swedish, ensuring clinical diversity and sex balance, resulting in the CliTe corpus. A CliTe subset underwent a multi-stage manual translation into the other 6 languages, resulting in the parallel ParaCliTe corpus, while the remaining reports were automatically translated, constituting ParaCliTe Silver. Results. CliTe is a collection of 699 artificial cardiology reports, mostly outpatient notes (340 documents), discharge reports (263), and inpatient notes (89), describing fictional patients with ischaemic heart disease (199 patients), heart failure (188), and valvular disease (79), among others. ParaCliTe contains reports from 94 fictitious patients, each available in seven languages with sentence-level alignment, resulting in 658 documents. The corpus mainly comprises discharge reports (322 documents), and outpatient (196) and inpatient (140) notes, describing patients with heart failure (399 documents), ischaemic heart disease (287), and valvular disease (238). Empirical results illustrate the utility of ParaCliTe and highlight the limitations of state-of-the-art NLP models adapted to the cardiology domain. Conclusions. The creation of multilingual, equitable and realistic corpora of artificial patients capable of strengthening NLP performance for cardiology is feasible when supported by detailed guidelines.

cs.CL

新论文 · v1arXiv:2611.09295

Bridging Spiking and Continuous Dynamics in Spiking Neural Networks: A Muscle-Inspired Model

Maryam Soltani, Philip N. Garner

原文摘要

Transforming the spike trains produced by spiking neural networks (SNNs) into continuous signals through biologically inspired mechanisms remains an open challenge in generative and regression tasks. In biological systems, a similar transformation occurs when muscles integrate neural spikes to produce smooth, coordinated movements. Inspired by this mechanism, we introduce a muscle-response model, a biologically motivated dynamical layer that converts spike trains into continuous representations. Our approach uses second-order dynamics to model the temporal integration of spiking activity. We investigate how the number of spiking neurons and temporal resolution affect the quality of the generated signals in a text-to-speech task, showing that increasing representational capacity has a greater impact than increasing temporal resolution under the evaluated configurations. Furthermore, an analysis of the learned dynamics reveals that, without explicit constraints on the response type, most neurons converge toward non-oscillatory behavior. These results demonstrate the potential of biologically inspired dynamical decoders for bridging spiking neural computation and continuous signal generation.

cs.NE · cs.LG · eess.AS

新论文 · v1arXiv:2611.09294

Orthogonal Norm Growth in Two-Tower Retrieval: A Popularity Mechanism for Unified Discovery

Andrei Atamaniuk

原文摘要

Unified search and recommendation systems may update the same item vectors from both interaction streams. We study how repeated updates can increase an item vector's norm and affect retrieval under an unnormalized dot product. A cosine-only loss makes the output gradient perpendicular to the current item vector - tangent to its constant-norm sphere. Training changes encoder parameters, however, so the realized item-vector movement need not remain tangent. We derive sufficient conditions under which it does; every nonzero update then increases squared norm by its squared length. When the tasks share an item encoder, sequential updates from both can accumulate in the same norm, which raw-inner-product serving can expose for either task. This conditional result is not a theorem about joint SGD. Controlled ablations test these conditions. On MovieLens 32M and Amazon Books, accumulated tangent movement accompanies frequency-norm association and additional head-item exposure under raw-dot rather than normalized-item ranking. In a nonlinear tower combining a per-item ID vector with a shared content encoding, active item movement remains near-tangent. We separate its norm change into the ID-update contribution and other output changes, then use a matched training intervention that removes immediate norm growth caused by the ID-row update. The intervention reduces norm-related exposure but also recommendation utility, so it is diagnostic rather than a proposed mitigation. Our logs are recommendation-only: the formulation supplies a shared-representation audit, while prevalence in joint traffic remains open.

cs.IR

新论文 · v1arXiv:2611.09293

MDF++: Visuo-Tactile World Action Model with End-to-end Representation Learning

Zixuan Huang, Jinhao Yu, Zachary Charlick, Dmitry Berenson

原文摘要

Recently, world-action models (WAMs) have emerged as a powerful framework for robot manipulation. By jointly modeling action generation and future observation prediction, WAMs can act and anticipate the outcomes of the actions. However, existing WAMs typically rely on frozen pretrained encoders trained with reconstruction, which may not produce representations well suited for robot control. In this paper, we tackle the problem of learning physicallygrounded, control-relevant representations for WAMs. We propose MDF++, a visuo-tactile world-action model that learns latent representations of multisensory observations end-to-end with a diffusion backbone. While naive joint training can lead to feature collapse, MDF++ introduces a multi-task learning strategy that leverages cross-modal correlations to learn task-relevant representations. We evaluate MDF++ on a suite of contact-rich manipulation tasks, including realworld bolt screwing and nut mating, and show that it outperforms WAMs built on frozen pretrained representations by up to 50%. Additional videos are available on our website: jointmdf.github.io.

cs.RO

新论文 · v1arXiv:2611.09291

AutoRefine: Auto-Calibrating Code Judges from Workflow Outcomes

Shervin Shahidi, James Poon, Hang Xie, Kenneth O. Xu, Robin Håkanson, Vikram Nair, Jens Pillgram-Larsen, Nachiappan Nagappan

原文摘要

LLM judges increasingly serve as quality gates for code changes that have no test oracle, but a judge prompt must be calibrated for its task and recalibrated as the system it judges evolves, while hand labels for the calibration are resource intensive and go stale. Development workflows emit outcome labels for free (tests pass or fail, changes land or are abandoned), but these labels are noisy. We present AutoRefine, a loop that rewrites a deployable judge's prompt to reproduce outcome labels, with an informed teacher that sees the judge's input plus the outcome, explains each label it keeps, and abstains when the outcome cannot be explained from what the judge sees. On a held-out dataset with an execution oracle, calibration raised chance-corrected agreement (Scott's $π$) from 0.36 to 0.60. On a deployed agent's outcome labels, with 25\% flipped at random to introduce noise, the teacher set aside flipped \textsc{fail} labels at more than twice the rate of genuine ones, cutting overall label noise from 25\% to 20\%. Recalibrating that agent's judge using AutoRefine raised its agreement with merge/abandon outcomes on 17 independently held-out diffs from the current workflow ($π$ 0.16 to 0.51); on 16 diffs from older workflows, the previous judge stayed ahead, illustrating the need for recalibration as the agent evolves. By abstaining, the teacher not only removes noisy labels but also flags changes whose outcome the judge's input cannot explain, telling a team when a judge needs better inputs rather than a better prompt.

cs.AI · cs.SE

新论文 · v1arXiv:2611.09290

Safety-Focused Reinforcement Learning Using Risk-Sensitive Coupled Control

Jiabin Lin, Shana Moothedath, Vivek S. Borkar

原文摘要

This paper investigates the coupled control problem within safety-focused reinforcement learning, addressing the essential challenge of achieving optimal performance while ensuring safety in systems operating within complex environments. We define safety as the requirement to prevent the system from reaching undesirable or hazardous states. We propose a coupled control approach that prioritizes safety by employing effective methods to prevent the system from entering unsafe states while optimizing the performance metric, defined as the average cost. We propose a novel problem transformation that reformulates it into a risk-sensitive problem combined with an average reward criterion. We present a two-timescale Q-learning algorithm where in the faster timescale, the algorithm solves a risk-sensitive control problem to ensure safety, while on a slower timescale, the algorithm focuses on average cost control. We provide theoretical guarantees for the convergence of our algorithms, establishing their significance in ensuring safe, optimal control in dynamic environments. We validated the performance of our approach using GridWorld and CarGoal task in Safety Gymnasium.

cs.LG

版本更新 · v4arXiv:2605.13214

Backdoor Channels Hidden in Latent Space: Extending Cryptographic Undetectability to Modern Neural Networks

Marte Eggen, Eirik Reiestad, Kristian Gjøsteen, Inga Strümke

原文摘要

Recent cryptographic results establish that neural networks can be backdoored such that no efficient algorithm can distinguish them from a clean model. These guarantees, however, have been confined to stylised architectures of limited practical relevance, leaving open whether comparable undetectability extends to modern, end-to-end trained networks. We construct such an attack mechanism for deep neural network architectures, closely aligned to the cryptographic notion of undetectability, by identifying backdoor channels as learned latent directions, and show that the question of undetectability reduces to a hypothesis test between two unknown distributions over model parameters, which we conjecture to be intractable in practice. The consequence of this reframing is significant: if exploitable channels within a network's latent space are statistically indistinguishable from naturally learned directions, an attacker need not introduce foreign structure but can instead exploit the geometry the network already possesses. Demonstrating the approach on ResNet and Vision Transformer architectures trained on standard image classification datasets, the attack achieves both consistently high success rates with negligible clean accuracy degradation, and resists a comprehensive suite of post-training defences, none of which neutralise the backdoor without rendering the model unusable. Our results establish that cryptographic backdoors need not be artefacts requiring exotic architectures or artificial constructions, but identifiable as latent properties inherent to the geometry of learned representations.

cs.CR · cs.LG

版本更新 · v3arXiv:2602.07915

CausalCompass: Evaluating the Robustness of Time-Series Causal Discovery in Misspecified Scenarios

Huiyang Yi, Xiaojian Shen, Yonggang Wu, Duxin Chen, He Wang, Wenwu Yu

原文摘要

Causal discovery from time series is a fundamental task in machine learning. However, its widespread adoption is hindered by a reliance on untestable causal assumptions and by the lack of robustness-oriented evaluation in existing benchmarks. To address these challenges, we propose CausalCompass, a flexible and extensible benchmark framework designed to assess the robustness of time-series causal discovery (TSCD) methods under violations of modeling assumptions. To demonstrate the practical utility of CausalCompass, we conduct extensive benchmarking of representative TSCD algorithms across eight assumption-violation scenarios. Our experimental results indicate that no single method consistently attains optimal performance across all settings. Nevertheless, in most scenarios, the methods exhibiting superior overall performance are deep learning-based approaches. We further provide hyperparameter sensitivity analyses to deepen the understanding of these findings. We additionally conduct ablation experiments to explain the strong performance of deep learning-based methods under assumption violations. We also find that NTS-NOTEARS relies on standardized preprocessing in practice, performing poorly in the vanilla setting but exhibiting strong performance after standardization. Finally, our work aims to provide a comprehensive and systematic evaluation of TSCD methods under assumption violations, thereby facilitating their broader adoption in real-world applications. The user-friendly implementation, documentation and datasets are available at https://github.com/huiyang-yi/CausalCompass.

cs.LG · cs.AI · stat.ME · stat.ML

新论文 · v1arXiv:2611.09282

Ontology-Driven Control of Conversational Agents with a Use-Case for Job Interview Practice

Barbara Gendron, Gaël Guibon, Mathieu D'aquin

原文摘要

Large Language Models (LLMs) have been widely used as conversational agents in both domain-specific and open-domain situations. In domain-specific context, it is often necessary to adapt those LLMs to obtain conversations with desired characteristics, whether those apply at the level of each response (e.g. for tone or style) or more globally (i.e. at the level of a conversational strategy). Such adaptations are, however, generally realized in an ad hoc way, through specific, non-reusable, and non-transparent developments. In this paper, we propose an ontology-based approach to design conversational strategies that rely on the desired and measured characteristics of textual utterances in the conversation. Through a use case related to job interview practice, we show how this approach can be used to control LLMs to follow the strategy defined in the ontology. To this end, we automatically fine-tune the LLM to improve its compliance with ontologically-defined conversation characteristics. Our results show that training the LLM from formally defined concepts effectively enhances alignment with the requested characteristics. Hence, this work demonstrates how ontology-guided conversation control can be achieved while taking into account multiple, concurrent conversation aspects that are formally and transparently defined.

cs.AI

新论文 · v1arXiv:2611.09280

An Empirical Study of Issues in GPU Kernel Libraries for LLM Systems

Yanjie Gao, Jiayi Bi, Haoxiang Lin, Ling Feng

原文摘要

Behind the expanding use of large language models (LLMs) across application domains lies a growing dependence on specialized GPU kernel libraries. These libraries underpin the correctness and efficiency of training and inference across diverse GPU platforms, implementing attention, GEMM, quantization, and low-precision operators while managing tensor shapes, memory layouts, code generation, and hardware-specific dispatch. Problems at this layer can silently corrupt outputs, hang GPUs, prevent compilation, introduce deployment incompatibilities, or degrade performance, thereby reducing developer productivity and wasting expensive GPU resources. We present the first comprehensive empirical study of issues in GPU kernel libraries for LLM systems. We manually analyzed 300 high-quality GitHub issue reports, their discussions, and available fixes from seven representative libraries: FlashInfer, FlashAttention, Triton, CUTLASS, DeepGEMM, xFormers, and bitsandbytes. The analysis characterizes symptoms, root causes, fixes, and testing and debugging practices. Our main findings are: (1) 60.33% of the sampled issues originate in the Compiler and Build System or Runtime Environment dimensions, so validation must cover the deployment and execution stack, not just kernel code. (2) Kernel specialization imposes implicit correctness constraints on tensor shapes, metadata layouts, memory aliasing, and hardware resources. Violations can produce incorrect outputs or launch failures even for supported operations, highlighting the need for explicit contract validation and boundary-case testing. (3) Current testing and debugging practices are often insufficient for identifying low-precision numerical problems and hardware-specific execution failures. We outline research opportunities and tooling directions to support future LLM GPU kernel development.

cs.AI

版本更新 · v2arXiv:2601.14837

Modular magnetic soft robotic catheter enables visualized intervention in the pancreatobiliary tree

B. Calmé, N. J. Greenidge, A. Metcalf, A. Bacchetti, G. Loza, D. Kpeglo, P. Lloyd, V. Pensabene 等 10 位作者

原文摘要

Diagnosis and therapy in the pancreatobiliary tree are limited by access. Guidewires reach deep into the ducts but are blind, whereas cholangioscopes provide vision but are too large to readily enter the pancreatic body and tail, the cystic duct, or the gallbladder. At the millimeter scale of these ducts, every function added to a steerable catheter competes with steering for cross-section and adds bending stiffness. We present a 1.47-mm modular magnetic soft robotic catheter in which each function is built into the magnetic body, onto its surface, or onto the footprint of the camera, and we measured what each function costs in bending under bedside fields. Imaging and fiber Bragg grating (FBG) shape sensing accounted for most of this cost, whereas a compliant gripper, magnetic anchoring balloons, and a micrometer-thick ultrasound-triggered release shell together reduced the tip angle at 8 mT only from 70° to 64°. The anchoring balloon increased the gripping force about threefold without stiffening the catheter during navigation. In live pigs, the robotic catheter performed operator-initiated task-autonomous cannulation of the native papilla without sphincterotomy and was teleoperated under onboard vision 75 mm into the pancreatic duct and, without a guidewire, through the cystic duct into the gallbladder, whose lumen it imaged. Anchoring, grasping, and payload release were each demonstrated in vivo. Adding each function at a measured cost to steering allows a soft robot small enough for these ducts to intervene as well as navigate.

cs.RO

版本更新 · v4arXiv:2510.16822

ReefNet: A Large-Scale Dataset and Benchmark for Fine-Grained Coral Reef Recognition

Abdulwahab Felemban, Yahia Battach, Faizan Farooq Khan, Yuqian Fu, Xuhui Liu, Yesmeen M. Khattab, Yousef A. Radwan, Xiang Li 等 22 位作者

原文摘要

Coral reef monitoring requires scalable recognition of visually similar taxa across heterogeneous underwater surveys. We introduce ReefNet, a public dataset that aligns approximately 925K hard-coral point annotations from 76 CoralNet sources and an independently collected Al-Wajh dataset with the World Register of Marine Species taxonomy. A stratified expert audit of 8,962 patches guides filtering to a 479,027-annotation benchmark with 39 labels: 38 genera and the family Fungiidae. Expert agreement is 92% on the retained audit subset, rather than an exhaustive accuracy estimate. We evaluate pretrained models through zero-shot recognition, few-shot adaptation, and full fine-tuning, each on within-source ReefNet and held-out-source Al-Wajh tests. Biological vision-language pretraining provides the strongest zero-shot results; supervised adaptation substantially improves recognition, with DINOv2 reaching 65.79% macro recall on Al-Wajh. Matched text-description experiments improve all four evaluated MLLMs on Al-Wajh, while a GPT-5.4 reasoning ablation does not improve over direct classification. Long-tail recognition, label uncertainty, and geographic imbalance remain challenges. ReefNet provides a taxonomically grounded resource for studying these challenges; the Al-Wajh test is a regional stress test rather than evidence of global robustness. Public data and resource links are documented in the supplement.

cs.CV · cs.AI

新论文 · v1arXiv:2611.09279

AnomSep: Bridging Optimization-Based Separation and Deep Neural Refinement for Anomaly Detection with Listenable Explanations

Koki Shoda, Jun Younes Louhi Kasahara, Qi An, Atsushi Yamashita

原文摘要

We propose AnomSep, which combines optimiza-tion-based separation with deep neural refinement to achieve accurate anomaly detection with listenable explanations. Anomaly detection requires no predefined anomalous categories, and understanding the basis for each detection helps security personnel assess the validity of each detection. End-to-end anomaly detectors can achieve high detection accuracy but output only scalar scores, and existing anomalous sound separation methods provide listenable explanations with limited detection accuracy. AnomSep first obtains preliminary normal-sound and anomalous-sound estimates through optimization-based separation using a dictionary learned from normal sounds in the target environment. Deep neural refinement then improves the preliminary estimates in a pretrained audio latent space. The separated anomalous sound serves as the listenable explanation, and its energy determines the anomaly score. Experiments on synthetic mixtures and real mixtures show improved separation quality over baseline separators. AnomSep achieves detection performance competitive with the end-to-end baseline and provides listenable explanations directly linked to each detection decision.

cs.SD · cs.LG

新论文 · v1arXiv:2611.09278

Diversity-Aware Client Selection for Communication-Efficient Federated Learning

Max Halleryd, Sima Sinaei, Viktor Valadi, Mina Alibeigi, Henrik Abrahamsson, Salman Toor

原文摘要

Federated learning (FL) enables collaborative model training across distributed clients without sharing raw data, making it attractive for privacy-sensitive and bandwidth-constrained applications. A practical challenge in deploying FL is reducing communication overhead while maintaining model performance under heterogeneous, non-IID data distributions. Selecting which subset of clients participates in each training round offers a principled way to address both challenges simultaneously, yet its impact on data diversity and representativeness remains underexplored. In this paper, we investigate three client selection strategies, Power-of-Choice, Fisher information, and Centered Kernel Alignment, evaluating them not only by accuracy and convergence, but through the lens of participation diversity using the Gini coefficient and KL divergence. Under non-IID conditions, restricting participation to 15% of clients per round achieves significant communication savings, and when selection promotes diverse and balanced contributions, it also improves generalization over biased alternatives. This work provides a systematic and reproducible analysis demonstrating that diversity-focused client selection renders communication efficiency and model quality complementary rather than competing objectives.

cs.LG · cs.AI

新论文 · v1arXiv:2611.09277

Federated Forensics: Jurisdictional constraints as a design driver for multi-agency federated learning

Aruna Withanage, Jessica Herrmann, Hudan Studiawan, Frank Breitinger

原文摘要

Digital forensic investigations increasingly cross borders, and law enforcement agencies need to build analytical models together without centralizing their data. Federated Learning (FL) keeps data local, yet neither FL nor its frameworks enforce the jurisdictional, governance, and accountability rules that multi-agency forensic work requires. This paper takes a design-first approach to that gap. First, we derive seven requirements that a federated learning system must meet in a forensic setting. Second, we propose Federated Forensics, a three-layer reference architecture whose middle layer, unlike in existing hierarchical FL designs, sits at a jurisdictional border and can use a model update locally while blocking it from crossing that border. Third, we build a prototype of the architecture and test it on a malware-family classification task. In a prototype evaluation, the policy gate blocked every restricted update at the border with zero violations, and restricted-class accuracy rose from 0% under today's compliant alternative to 93.0%, only 4.4 points below ignoring the restriction.

cs.CR

新论文 · v1arXiv:2611.09275

AudioSphere: Towards Self-Supervised Spatial Audio Representation Models

Goksenin Yuksel, Marcel van Gerven, Kiki van der Heijden

原文摘要

Recent audio embedding models learn representations that transfer across clip- and frame-level tasks, yet they operate almost exclusively on single-channel signals and are blind to spatial structure. Spatial audio models capture that structure but rely on clip-level supervision or contrastive objectives, yielding embeddings unsuitable for frame-level tasks like sound event detection and localization (SELD). We argue this dichotomy is not fundamental: patch-level spatial self-supervision can jointly learn acoustic and spatial features at both temporal granularities. We introduce AudioSphere, a multi-channel masked autoencoder trained on Ambisonics recordings that reconstructs both spectral content and directional cues via a joint spatiotemporal masking strategy that masks both time-frequency patches and intensity vectors. AudioSphere supports frame-level predictions while producing strong clip-level representations via aggregation. Alongside the model, we release RealSELD, a standardized benchmark for evaluating frozen backbones on real-world multi-channel SELD. AudioSphere enables, for the first time, SELD with a frozen self-supervised backbone, while matching or exceeding prior self-supervised audio embedding models on clip-level classification tasks spanning environmental sound, speech, and music, and on frame-level sound event detection. All code and data are available at https://github.com/labhamlet/AudioSphere and https://huggingface.co/labhamlet.

cs.SD · cs.AI