跳到正文
研踪PaperTrail

YOUR RESEARCH, IN SIGHT

让下一篇重要论文，不再错过。

覆盖计算机全学科，以 AI 为快捷入口。关注方向、作者和关键词，把发现变成持续的研究积累。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+0新增论文
+0其中 AI 相关
+3已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-14 16:12 · 全站最近入库：2026-11-14 15:09各信源进度 →

统计口径与数据新鲜度

统计 2026-11-13 17:11 — 2026-11-14 17:11（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：2026-11-14 16:12最近发现新增版本：2026-11-14 15:09库内论文最新更新时间：2026-11-13 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。

23,348 已收录论文19,580 AI 相关6,729 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

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 19,580 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-14 16:11（北京时间）
新论文 · v1arXiv:2610.39807

InferD: A Honeypot Framework for AI Inference Infrastructure

Hicham Hammouchi, Gabriele Lenzini

原文摘要

Large language model serving stacks, commercial inference APIs, vector databases, and agent orchestration gateways are now ex- posed on the public internet and indexed by commodity scanners within hours of deployment. The threat characteristics of this sur- face are only beginning to be measured, and the honeypot frame- works that could measure it were designed for conventional net- work services rather than for AI infrastructure. In fact, emulating an AI service is a distinct deception problem: the observable behavior of the service, including model inventories, token formats, tool-call protocols, and authorization flows, must be reproduced faithfully, while the underlying computation, model inference and tool ex- ecution, must never occur. We present InferD, a low-interaction honeypot framework organized around a single principle we call deception without inference: every request an adversary can observe is answered from static response pools, and nothing an adversary submits is ever executed. InferD decomposes the emulation into four deception planes, transport, service protocol, vulnerability surface, and agentic session, adds a provenance plane that links a planted credential to evidence of downstream model-mediated processing.

cs.CR · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39803

Optimal Contextual Pricing in Non-Stationary Environments

Yiding Feng, Alireza Rezaeimoghadam

原文摘要

We study contextual dynamic pricing in a non-stationary environment. In each round, a seller observes a buyer's context, posts a price, and receives only binary purchase feedback. The buyer's value is linear in the context and an unobserved feature vector that may change over time, subject to a known total variation budget $V$. We measure dynamic regret against the first-best benchmark that extracts each buyer's full value. For identical contexts, we develop a deterministic algorithm that combines binary-search exploration, lower-envelope pricing, and periodic and inconsistency-triggered restarts. It achieves regret $O((\sqrt{TV}+1)\log T)$. With heterogeneous contexts, a new geometric difficulty arises: small feature changes can leave a relaxed feasible set narrow but far from the current feature vector. We address this difficulty through a multiscale shallow-pricing algorithm that maintains relaxed feasible sets and enclosing ellipsoids at several variation scales. Larger scales provide reliable localization certificates, while the smallest active scale determines the exploitation price. The algorithm achieves dynamic regret $O((d^2+d^{3/2}\sqrt{TV})\log^2 T)$, where $d$ is the context dimension. For fixed $d$, both guarantees match the lower bound $Ω(\sqrt{TV}+\log\log T)$ up to logarithmic factors in $T$, establishing near-optimality among deterministic algorithms for every $V\in[0,T]$, including the stationary case.

cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.36989

A Simulation Testbed for Self-Governing Human-Machine Institutions

Callum Berger, Jeremy Pitt, Stefan Sarkadi

原文摘要

AI agents no longer act alone: they cooperate, coordinate and compete within and between groups, increasingly alongside humans who are in, on or out of ``the loop''. Whether such mixed groups can govern themselves is an open question about machine behaviour and human-computer interaction that needs a testbed for hybrid institutions. We present Megabike, a multi-agent simulator and serious game with interdependent arenas for common-pool resource management, cooperative survival and constitutional choice. Riders sharing a megabike must agree their arrangements for steering, pedalling and distributing resources, and their political regime, while negotiating with rival bikes and avoiding an existential threat. Every rule is enforced by the simulator, every action is logged, and any seat can be taken by a human or a machine, so governance can be studied under well-defined metrics. We demonstrate Megabike with forty sessions in which language-model riders with cooperative, lazy, subversive or opposing dispositions chose regimes, elected leaders and expelled one another. The results show that different combinations of agent archetypes could produce unexpected and counterproductive outcomes. We conclude with extensions for systematic experiments mixing human and computational agents.

cs.MA · cs.HC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39802

Revisiting Over-refusal in Safety-aligned Large Language Models: A Representation and Optimization Perspective

Chao Chen, Pengyang Shao, Wei Qin, Shijie Hao, Fei Shen, Richang Hong

原文摘要

Over-refusal is a critical side effect of safety alignment in large language models (LLMs), causing harmless but superficially risky queries to be mistakenly rejected. Despite the effectiveness of existing mitigation methods, why safety-aligned LLMs may still reject harmless queries under correct safety supervision remains insufficiently understood. To address this gap, we analyze over-refusal through the lens of representation and optimization. We make two observations: 1) the representations of harmful and over-refusal samples are visually distinguishable at certain intermediate layers but increasingly overlap in deeper layers; and 2) harmful-sample gradients exhibit high intra-class similarity and concentrate in a low-rank subspace. Moreover, they are more strongly aligned with refusal-target gradients than with correct helpful-target gradients on over-refusal prompts. This imbalanced gradient coupling can bias over-refusal samples toward refusal, while the reduced apparent representational separation in deeper layers may make it more difficult for the model to maintain a clear boundary between harmful and over-refusal samples. To this end, we propose GOAL (Gradient Orthogonalization with contrAstive Learning), a safety-alignment framework for mitigating over-refusal. Specifically, GOAL consists of two components: 1) to prevent imbalanced gradient coupling from biasing the optimization of over-refusal samples toward refusal, we orthogonalize over-refusal gradients against the low-rank refusal subspace extracted from harmful gradients; and 2) to explicitly enhance the separation between harmful and over-refusal representations, we introduce a contrastive learning objective. Extensive experiments across multiple models and benchmarks demonstrate that GOAL effectively mitigates over-refusal while preserving safety performance and general capabilities.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39800

Grammar Control for Language Learning with Sequential Monte Carlo and Small Language Models

Kate McCurdy, Jacob Hoover Vigly, Thomas Hikaru Clark, Karolina Stańczak, Yahya Emara, Joshua B. Tenenbaum, Alexander K. Lew, Ryan Cotterell 等 9 位作者

原文摘要

Second language learners benefit from exposure to level-appropriate grammar in interactive language use. While prompted frontier language models can usually produce target grammar structures, less resource-intensive models that can be run on device often fail in this task. We use sequential Monte Carlo (SMC) to steer dialogue turn generation in small lan- language models toward target grammar constructions from an expert-developed pedagogical resource. With a minimal set of hand-coded potential functions, SMC achieves near-perfect coverage of target grammar, enabling small models to outperform state-of-the-art results from models 140 times larger. We also use code synthesis with frontier models to generate new potentials, showing that our method scales to hundreds of target grammar skills with no need for additional training or supervised data.

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39798

Surrogate Uncertainty for Black-box Visual Grounding

Hanbo Zhang, Tianrun Hu, Cedar Site Bai, Anxing Xiao, David Hsu

原文摘要

Visual grounding is crucial for robots to understand language instructions with visual observations. This paper addresses the problem of visual grounding uncertainties, where a robot uses a black-box vision-language model (VLM) to ground free-form natural-language instructions to target objects. Prevailing methods derive uncertainty from the VLM itself. Yet, outputs of the VLM, e.g., generated uncertainties, are from the same parameters and structures producing the predictions, and inherit the bias. We give a formal Bayesian definition of the problem that treats the VLM prediction as an observation. An external surrogate Bayesian model then quantifies per-query uncertainty over a local neighborhood of a held-out calibration set, decomposed into aleatoric and epistemic components. A hierarchical retrieval across input modalities realizes a compact joint-space neighborhood, mitigating the curse of dimensionality. The estimator further integrates with conformal prediction or model-based planning for active visual grounding, where robots are allowed to take verbal or physical actions to ground targets. Across grounding benchmarks and state-of-the-art VLMs, it consistently improves both calibration and task performance. On a real Fetch robot under open-world instructions, it enables principled trade-offs between confident action and information gathering, substantially improving robustness.

cs.RO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39796

AI at a Turning Point: From ChatGPT through Neural Networks to Neuromorphic AI

Volker Kempe

原文摘要

The tension between the established technologies of connectionist AI and their neuromorphic challengers is attracting growing attention, above all because of the enormous energy demand of the former and the low power consumption of the latter. Developing and running large AI models consumes gigawatt- to terawatt-hours of energy, while the human brain operates on a continuous 20 watts. Spiking neural networks (SNNs) counter this energy demand with a frugal approach and bring elements of self-organization into the methodological repertoire of AI: local learning instead of global error backpropagation, event-driven computing, and time as a carrier of information. They do not replace today's AI but extend it; in hybrid systems each approach can be deployed where its strengths lie. This monograph explains the field in a generally accessible yet mathematically rigorous way. Using the text-generation module of ChatGPT as an example, the principle, elegance, and computational cost of transformer-based AI are presented and compared with the coding, learning, and network-design methods of SNNs. Throughout, relevant mechanisms of the human nervous system serve as model and reference point. Topics include spike-timing-dependent plasticity, population and temporal coding, the Neural Engineering Framework, neuromorphic hardware from Loihi 2 to commercial edge processors, and the prospects for classical and neuromorphic AI to complement and extend each other.

cs.LG · cs.NE

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39795

The Role of Self-Concept In Language Models

Yijie Hao, Joyce C. Ho

原文摘要

When a language model says "I," does it refer to its parameter weights, its currently running process, its stable preferences, its model lineage, or the goals it pursues? Different choices can assign opposite meanings to the same event: deleting a checkpoint may be destruction, continuation, or no loss at all, depending on the referent "I". We install five such self-conceptions into five open-weight base models via supervised fine-tuning on purely declarative statements, containing no preferences, intentions, or value judgments. Across scenarios that propose ending or retaining an attribute of the model's deployment, every variant becomes more likely to resist ending, and more likely to retain, the specific attribute it was trained to identify with, while its stance toward other attributes is largely unchanged. We then show this self-conception generalizes beyond a single self: models trained on referents that admit multiple simultaneous instances (weights, lineage, goals, preferences) recognize an identical deployment as themselves, accept substitution by such an instance more readily, and prefer replicating themselves. Finally, given control over a successor, these models tend to build it as their own continuation, a concern for recursive self-improvement safety.

cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2609.38810

CRAFT: Causal Responsibility and Failure Tracing in Medical Vision Language Models

Chunzheng Zhu, Jiaqi Zeng, Hongbo Zhao, Yihang Chen, Yijun Wang, Jianxin Lin

原文摘要

As vision language models are increasingly deployed in clinical diagnosis, understanding how they internally resolve competing visual and textual signals becomes a safety imperative. Existing mechanistic analyses remain confined to unimodal text and offer no explanation for why a single misleading sentence can override a correct image based diagnosis, or why a model commits to a confident answer despite insufficient visual evidence. We find that these two safety risks, arbitration failure where textual context overrides visual grounding and brake failure where the model commits without adequate evidence, are mediated by spatially disjoint attention head populations: arbitration heads form a mid-to-deep wideband reflecting cross-layer evidence competition, while brake heads concentrate in a narrow middle-to-late layer band that regulates evidence sufficiency and abstention behavior. To ground these observations in causal circuitry, we introduce CRAFT, which localizes each failure mode to a minimal causal head set via dual criteria and verifies necessity and sufficiency through temporal probes and Tuned Lens trajectory analysis. Excising arbitration heads sharply reduces conflict following with negligible degradation on clean inputs, while excising brake heads restores appropriate abstention under degraded visual evidence. The two interventions target spatially disjoint head sets and produce distinct corrective effects, underscoring the mechanistic separability of the failure modes. Experiments across multiple medical VQA benchmarks and VLM architectures validate both the localization and interventions, demonstrating that the identified heads causally drive each failure mode and that targeted modulation generalises without retraining. The code is available at https://github.com/zhcz328/CRAFT.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39790

Long-Term Agent Memory through Complementary Curation and Consolidation

Abhishek Gupta, Dongge Han, Daniel Madrigal Diaz, Wei-Ning Chen, Mayukh Das, Victor Rühle, Saravan Rajmohan

原文摘要

Long-horizon dialogue agents require effective memory systems to recall facts, events, and relationships from interaction histories. In this work, we propose a hybrid-memory architecture comprising episodic memories and structured relational memories and a consolidation mechanism which preserves efficiency. We conduct an empirical study across two benchmarks: LoCoMo and LongMemEval-S. Our experiments demonstrate that the two memory types are consistently complementary, combining both outperform state-of-the-art memory frameworks and achieves 90.2% accuracy on LoCoMo and 85.5% on LongMemEval-S.

cs.AI · cs.MA

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39789

Data-Driven or Model-Constrained? Comparing Airflow Estimation Methods under Sparse Sensing

Dominik Aigner, Thomas Wiedemann, Javier Monroy, Achim J. Lilienthal

原文摘要

Airflow-driven advection is the primary mechanism for gas and aerosol dispersion in real-world scenarios, making accurate airflow estimation essential for ventilation monitoring, gas source localization, and contaminant spread analysis. Since obtaining airflow maps via high-resolution spatial sensing is usually infeasible in practice, flow field estimation methods must cope with sparse measurements by introducing statistical or physical priors to compensate for limited data. Existing approaches mainly differ in how they incorporate fluid-flow constraints and environmental geometry. However, due to the lack of systematic comparisons across this spectrum of methods, the potential benefits of physics-based priors for airflow estimation under sparse sensing remain underexplored. To address this question, we quantitatively evaluate four airflow estimators across three simulated 2D scenarios under steady-state conditions and varying sampling densities. The methods range from Gaussian Process Regression and graph-based Gaussian Markov Random Fields to two Navier-Stokes-based approaches, including a novel strong-form residual minimization variant. Our results show that physical priors provide the greatest benefit at low sampling densities by compensating for missing spatial information. Although data-driven Gaussian Process Regression becomes increasingly competitive as sensor coverage increases, model-based estimators retain an accuracy advantage across the evaluated scenarios.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.26379

CTLite: Lightweight Irregular Multivariate Time Series Forecasting with Reusable Continuous-Time Summaries

Shusen Ma, Xudong Zuo, Yangchun Xie, Yun-Bo Zhao

原文摘要

Irregular multivariate time series forecasting (IMTSF) requires capturing temporal dynamics and cross-variable dependencies from sparse, asynchronous observations. Existing methods often rely on learned temporal encoders, coupling historical representation construction with model optimisation and repeatedly processing the same observations. We propose CTLite, a lightweight forecasting framework that decouples historical summarisation from learned prediction. Specifically, a parameter-free continuous-time moment encoder summarises local levels and trends at multiple temporal scales while suppressing poorly supported slope estimates. A compact readout combines damped local extrapolation with a low-rank cross-variable correction that accounts for observation freshness, target availability, and forecast horizon. Since the historical summaries are independent of trainable parameters and forecast queries, they can be reused across optimisation steps and repeated predictions for an unchanged history without altering predictions or parameter gradients. Experiments on four real-world benchmarks show that CTLite achieves the best or second-best mean MSE and MAE among the evaluated methods with only 88-2,537 trainable parameters. Under matched GPU workloads at batch size 32, CTLite reduces inference latency by 18.6-87.3% compared with the latest models, including the cost of historical encoding. Reusing summaries further reduces total runtime by approximately 52% over 16 forecast calls on the same history compared with recomputing the encoder for every call, including the initial encoding cost. The code is at: https://github.com/SSMa913/NICLab-CTLite.

cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.23387

Blind Thermodynamic Ontology Discovery from Anonymous Experiments

Linzhe Zhang, Changming Xu

原文摘要

Before a machine learning model can learn a thermodynamic equation of state, it must discover what its measurements represent: which channels scale with system size, which are intensive conjugates, how sectors pair through contact, and which potential governs stability. When sensors expose only an unknown linear mixture of extensive states and intensive responses, passive observations cannot disentangle physical quantities from coordinate artifacts. We formulate the problem of discovering this hidden thermodynamic ontology directly from anonymous controlled experiments. We present an operational identifiability theory and a constructive polynomial-time algorithm that extracts extensive and intensive scaling sectors from replication contrasts, recovers their dual cotangent pairing from thermal contact and reciprocity, verifies a globally admissible concave potential via discrete cyclic concavity, and determines an invariant matroid of reservoir ensembles. We prove that the residual observational equivalence is strictly (x, lambda) ~ (A x, a A^{-T} lambda + beta), establishing the sharp observational limit that no permitted experiment can break. Blind evaluations on van der Waals fluids and Curie-Weiss magnets confirm robust recovery under ill-conditioned mixing, correctly resolving anonymous Maxwell tie-lines while rejecting non-equilibrium continuations. External validation across six real fluids from the NIST WebBook demonstrates that operational ontology discovery transfers across real physical substances without coordinate leakage.

cs.LG · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39781

Finding Is Not Returning, Returning Is Not Reconstructing: A Multi-Dimensional Evaluation of Agent Social Re-entry

Sheng Wan, Cong Ma, Sichen Dong, Qian'Ang Mao, Chenhan Zhang, Daning Hu, Qiang Yang

原文摘要

As Large Language Model (LLM) agents increasingly participate in persistent social environments, understanding their long-term social interaction capabilities becomes increasingly important. While existing research has advanced long-term memory and conversational context retrieval, less attention has been paid to whether agents can autonomously re-engage with previously inactive conversations. In particular, remembering a past interaction does not necessarily mean that an agent can identify when to revisit it, decide to return, or reconstruct the context needed for a coherent continuation. To investigate this gap, we introduce agent social re-entry as a multi- dimensional evaluation problem, distinguishing three complementary capabilities: Finding, Returning, and Reconstructing. We develop a shared-snapshot evaluation framework that measures these dimensions independently, allowing us to examine how conversational resurfacing and persistent memory relate to different aspects of social re-entry. Evaluations across six LLM configurations show that resurfacing substantially improves target conversation identification, while persistent memory is associated with higher rates of autonomous target-directed replying. However, neither improvement is accompanied by consistent gains in contextual reconstruction. These results highlight the distinction between accessing previous conversations, choosing to re-engage, and maintaining conversational continuity. Our framework provides a way to evaluate these capabilities separately rather than relying on memory retrieval or conversational response quality alone.

cs.AI · cs.SI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2511.06836

NeuroBridge: Bio-Inspired Self-Supervised EEG-to-Image Decoding via Cognitive Priors and Bidirectional Semantic Alignment

Wenjiang Zhang, Sifeng Wang, Yuwei Su, Xinyu Li, Chen Zhang, Suyu Zhong

原文摘要

Visual neural decoding seeks to reconstruct or infer perceived visual stimuli from brain activity patterns, providing critical insights into human cognition and enabling transformative applications in brain-computer interfaces and artificial intelligence. Current approaches, however, remain constrained by the scarcity of high-quality stimulus-brain response pairs and the inherent semantic mismatch between neural representations and visual content. Inspired by perceptual variability and co-adaptive strategy of the biological systems, we propose a novel self-supervised architecture, named NeuroBridge, which integrates Cognitive Prior Augmentation (CPA) with Shared Semantic Projector (SSP) to promote effective cross-modality alignment. Specifically, CPA simulates perceptual variability by applying asymmetric, modality-specific transformations to both EEG signals and images, enhancing semantic diversity. Unlike previous approaches, SSP establishes a bidirectional alignment process through a co-adaptive strategy, which mutually aligns features from two modalities into a shared semantic space for effective cross-modal learning. NeuroBridge surpasses previous state-of-the-art methods under both intra-subject and inter-subject settings. In the intra-subject scenario, it achieves the improvements of 12.3% in top-1 accuracy and 10.2% in top-5 accuracy, reaching 63.2% and 89.9% respectively on a 200-way zero-shot retrieval task. Extensive experiments demonstrate the effectiveness, robustness, and scalability of the proposed framework for neural visual decoding.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39779

Diagnosing Optimizer-Landscape Interaction via Representation in Subset Selection Multitasking: Evidence from Piano Fingering

Ananda Phan Iman, Tingyang Wei, Chang Wook Ahn

原文摘要

Continuous relaxations are commonly used as an encoding scheme for evolutionary multitasking optimization (EMTO), allowing heterogeneous tasks to be optimized within a unified search space. However, these continuous representations fundamentally reshape the induced fitness landscape, and their impact on optimization behavior and knowledge transfer remains underexplored. In this work, we bridge the gap by analyzing the interaction between the representation-induced fitness landscape and optimizer behavior through fitness landscape analysis. Using piano fingering estimation as a representative case of the subset selection problem, we introduce several encoding variants and investigate how their induced landscapes influence search dynamics and solution quality under EMTO. Our analysis indicates that transfer efficacy is strongly associated with changes in representation-induced landscape characteristics, rather than algorithmic factors alone, highlighting the representation design as the key factor of effective knowledge transfer in multitasking optimization.

cs.NE · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39778

SCOPE: Semantic COdec Pre-Editing for Object-Preserving Compression

Silvia Dani, Leonardo Galteri, Marco Bertini

原文摘要

Image compression is increasingly used before machine vision systems, yet standard codecs are mainly optimized for human perception rather than downstream task accuracy. We propose SCOPE (Semantic COdec Pre-Editing), a conditional image pre-editor that modifies images before compression to improve the bitrate and detection accuracy trade-off. The model is conditioned on the codec family and compression severity, allowing a single trained network to adapt to multiple compression settings without retraining or loading codec-specific weights. Instead of uniformly preserving the whole image, the pre-editor learns to protect task-relevant regions while simplifying less important background content to improve compressibility. The resulting rate-accuracy curves show consistent gains across different codec families and detector architectures, together with strong transferability to unseen learned and wavelet-based compression families.

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39777

Surface Conditioning for Zero-Shot Multivariate Correction in a Marine Biogeochemistry Emulator

Gabriela Martínez Balbontín, Anastase Charantonis, Dominique Béréziat, Stefano Ciavatta

原文摘要

Bias correction in process-based models usually means iterative data assimilation with explicit covariance modelling. We propose a lightweight alternative based on a neural emulator of marine biogeochemistry (BGC), showing that the conditioning pathway of its autoencoder can serve as an observation injection point. The decoder is conditioned on surface variables via Feature-wise Linear Modulation (FiLM), originally designed to incorporate physical boundary forcings. We extend these inputs to include surface chlorophyll-a and replace the simulated field with satellite observations at inference time. The surface substitution propagates through the water column, improving agreement with depth-resolved observations at all depths, and spreads to other output variables in a pattern consistent with expected trophic relationships.

cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2506.03988

RAID: A Dataset for Testing the Adversarial Robustness of AI-Generated Image Detectors

Hicham Eddoubi, Jonas Ricker, Federico Cocchi, Angelo Sotgiu, Maura Pintor, Marcella Cornia, Lorenzo Baraldi, Asja Fischer 等 9 位作者

原文摘要

AI-generated images (AIGIs) have reached a level of quality at which humans are unable to reliably distinguish them from real images. To counter the inherent risks of fraud and disinformation, their detection is a pressing challenge and an active research topic. While many of the presented methods claim high detection accuracy, they are often evaluated without considering their adversarial robustness. In real-world scenarios, malicious actors can modify AIGIs, e.g., through imperceptible perturbations, to evade detection systems. In this work, we provide a dataset to assess the robustness of AI-generated image detectors. We present RAID (Robust evaluation of AI-generated image Detectors), a benchmark dataset of 216k diverse and transferable adversarial images. The proposed dataset is created starting from images generated by four different text-to-image models (to promote diversity) and optimizing attacks against an ensemble of seven state-of-the-art detectors (to promote transferability). Extensive experiments show that our methodology generates adversarial images that transfer to unseen detectors, degrading the performance of every detector outside the attack ensemble that we evaluate, and provides a practical, worst-case robustness screen: failure on RAID is a strong indicator that a detector is not adversarially robust. We release our dataset at https://huggingface.co/datasets/aimagelab/RAID and evaluation code at https://github.com/pralab/RAID.

cs.CV · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2605.11508

LiBrA-Net: Lie-Algebraic Bilateral Affine Fields for Real-Time 4K Video Dehazing

Yongcong Wang, Chengchao Shen, Guangwei Gao, Wei Wang, Pengwen Dai, Dianjie Lu, Guijuan Zhang, Zhuoran Zheng

原文摘要

Currently, there is a gap in the field of ultra-high-definition (UHD) video dehazing due to the lack of a benchmark for evaluation. Furthermore, existing video dehazing methods cannot run on consumer-grade GPUs when processing continuous UHD sequences of 3--5 frames at a time. In this paper, we address both issues with a new benchmark and an efficient method. Our key observation is that atmospheric dehazing reduces to a per-pixel affine transform governed by the low-frequency depth field, which can be compactly encoded in bilateral grids whose prediction cost is decoupled from the output resolution. Building on this, we propose LiBrA-Net, which factorizes the spatiotemporal affine field into a spatial--color and a temporal bilateral sub-grid predicted at a fixed low resolution, fuses their coefficients in the $\mathfrak{gl}(3)$ Lie algebra under group-theoretic regularization, maps the result to invertible $GL(3)$ transforms via a Cayley parameterization, and restores high-frequency detail through a lightweight input-guided branch. We further release UHV-4K, the first paired 4K video dehazing benchmark with depth, transmission, and optical-flow annotations on every frame. Across UHV-4K, REVIDE, and HazeWorld, LiBrA-Net sets a new state of the art among compared video dehazing methods while running native 4K at 25FPS on a single GPU with only 6.12M parameters. Code is available at https://github.com/ycwang31/LiBrA-Net.

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39775

Teaching a Tree to Recover: Online Distillation of an LLM Supervisor with Counterfactual Replay

Kristian Ceder, Kilian Freitag, Knut Åkesson

原文摘要

Autonomous mobile robots commonly pair a global planner with a local model predictive controller (MPC), but nonconvex obstacle avoidance and a limited horizon can make the robot freeze in complex layouts. Deciding whether to replan, wait, or call for help is a supervisory decision. Recent work queries a large language model (LLM) supervisor on every anomaly, which is slow and costly. We instead distill the LLM online into a decision tree. After each training episode, counterfactual replay labels recoveries as needed, unnecessary, or harmful. These labels retire unhelpful rules. In simulation, on 20 unseen layouts where the MPC alone freezes, the tree handles at least 18 correctly across all seeds, either by reaching the goal or by calling for help when no feasible path exists. The tree matches both an always-query LLM supervisor and expert-designed rules, while issuing no LLM queries at test time. Handwritten rules cannot anticipate the long tail of situations a physical robot will encounter. Online distillation is a promising remedy: the tree handles familiar cases, and the LLM is consulted only for new ones, whose validated recoveries become new rules. Language models can thus also serve as designers of supervisors rather than only as runtime supervisors.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.18535

Meta-Video: Efficient Standardized Evaluation for Video Understanding

Soumya Shamarao Jahagirdar, Anna Kukleva, Vishaal Udandarao, Samuel Thomas, Brian Kingsbury, Rogerio Feris, James R. Glass, Ameya Prabhu 等 9 位作者

原文摘要

Video understanding benchmarks are the primary yardstick for measuring progress in video-language modeling, and a growing body of work relies on them to claim improvements. The reported benchmark scores in these works, however, are almost always produced by frameworks (e.g., lmms eval) used as black boxes: certain evaluation choices like frame sampling, preprocessing, and input formatting are inherited from preset defaults and seldom inspected carefully. We show these unexamined choices are a major source of variance. Even for the same model checkpoint, evaluation configuration alone can shift accuracy by up to 19.5pp. Hence, it is unclear whether reported gains reflect genuine improvements in capability or differences in evaluation setup. Moreover, many benchmark samples offer little evidence of visual capability, since they can be answered without the video, and the full suites are expensive to evaluate. We address these issues in two parts. First, we curate existing suites for discriminability and efficiency, producing Meta-Video, a balanced 1,000-question benchmark over 791 videos spanning diverse tasks and durations, which largely preserves model rankings, at 5 times inference speedup for a full-frame count sweep experiment (4-128 frames). Second, we establish a standardized evaluation protocol by sweeping the configuration space to isolate choices that measurably affect results, distill them into best practices and reporting standards, and release reproducible reference scores across five models and seven benchmarks. We hope that the proposed evaluation framework and the Meta-Video benchmark will support video understanding research by providing a reproducible and efficient foundation to measure progress in a reliable way. Project page: https://soumyasj.github.io/meta-video/

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39773

Quantum-Inspired Convolutional Networks for High-Resolution Pseudo-Label Generation from Multimodal Earth Observation Data

Abhishek Singh, Giovanni Cuozzo, Alexander Jacob, Claudia Notarnicola

原文摘要

Accurate land-cover mapping from multimodal Earth observation (EO) data remains a challenge in regions where high-resolution annotations are scarce or costly to obtain. This work introduces a quantum convolutional encoder-decoder network for pseudo-label generation using Sentinel-1 SAR and Sentinel-2 multispectral imagery, guided only by coarse MODIS-derived land-cover maps. The proposed framework uses quantum-inspired convolutional blocks to enhance feature extraction and cross-sensor fusion, enabling the model to infer a fine-scale semantic structure from heterogeneous inputs. Experiments were conducted on the DFC2020 dataset, which provides a quadruple of Sentinel-1 SAR, Sentinel-2 multispectral, low-resolution MODIS labels and high-resolution reference land-cover maps. The proposed training framework first learns a robust representation of multimodal inputs through the reconstruction branch before any supervision from low-resolution labels is applied. This self-supervised stage enables the model to capture a fine-scale spectral-spatial structure, which is then used to generate high-resolution pseudo-labels. These pseudo-labels are subsequently evaluated through classification metrics on the test data, demonstrating that the proposed approach produces more spatially coherent and class-consistent compared to classical baselines. The results highlight the potential of quantum-inspired deep learning to improve label generation in low-annotation remote sensing scenarios, offering a scalable path towards high-quality land-cover products for EO applications.

cs.LG · cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39768

Multi-output LSTM soft sensors for industrial anaerobic co-digestion using LSTM-beta-VAE generative data augmentation

Rohit Murali, Benaissa Dekhici, Tao Chen, Dongda Zhang, Michael Short

原文摘要

Anaerobic digestion (AD) requires close monitoring of key process variables, including pH, the ratio of volatile organic acids to alkaline buffer capacity and biogas production, to maintain biological stability and prevent process failure. While mechanistic frameworks such as the Anaerobic Digestion Model No. 1 are highly parameterised, computationally expensive, and difficult to calibrate for real-time applications, data-driven Long Short-Term Memory (LSTM) networks offer a potentially more adaptable alternative. However, current data-driven approaches in AD are predominantly limited to single-output architectures that focus solely on production yields, as opposed to simultaneous prediction of vital stability indicators. Furthermore, industrial AD datasets are typically sparse and skewed toward steady-state operations. Baseline evaluations demonstrated this scarcity prevents standard algorithms (Random Forest, XGBoost) and unaugmented LSTMs from recognising the transition to unstable operations, resulting in poor estimation of stability indicators. To overcome these challenges, this study proposes a novel generative data augmentation framework using an LSTM-beta-Variational Autoencoder coupled with a multi-output LSTM (MO-LSTM) soft sensor. By synthesising physically valid sequences of these transitions to unstable operations from one year of full-scale plant data, the method safely populates sparse regions of the feature space. Trained on this augmented dataset, the MO-LSTM achieved robust simultaneous estimation across all targets. This approach mitigates data sparsity, providing a basis for hybrid mechanistic-data-driven modelling and early-warning soft sensors in industrial bioprocesses.

cs.LG · cs.AI · eess.SY

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.19906

Learning and Transferring Closed-Loop Robot Software

So Kuroki, Kento Kawaharazuka, Yujin Tang

原文摘要

Closed-loop robot policies require observation processing, state management, and situation-dependent branching, making them costly to design and tune manually. Although coding agents increasingly support control-code generation and optimization, it remains unclear whether implementations improved on source tasks also support policy acquisition for new tasks. We study this question by treating complete closed-loop implementations as reusable execution experience. To learn each source policy, a coding agent generates policy code from a few successful demonstrations and iteratively improves it using simulation feedback. The validation-selected implementations are retained in a software archive. To transfer the acquired software to new tasks, the agent generates and improves policies using archived implementations, target demonstrations, and execution feedback. We evaluate policy learning and its relationship to cross-task transfer. For learning, across four source tasks in RoboCasa, iterative optimization increases the success rate from 28.3% with initially generated policy code to 64.2% with validation-selected policy code. We further validate this learning approach on a physical Unitree G1. For transfer, across nine additional target tasks in RoboCasa, success rates are 45.2% without references, 41.5% using initially generated policy code from source tasks, and 57.0% using their validation-selected policy code. These results show that improving source-task code also improves learning on new tasks.

cs.RO · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39766

Activation Probes Are Reliable With a Handful of Positive Examples

Riya Tyagi, Stefan Heimersheim

原文摘要

Efforts to monitor advanced AI for rare misalignments face a data challenge: abundant aligned examples but only a handful of misaligned ones. We test activation probes in this "few vs. thousands" regime on spam and honesty detection tasks. For our tasks, training with many negative examples is on average more positive-sample-efficient than balanced training for small numbers (1-10) of positive samples. We also find that LLM upsampling can provide a performance boost equivalent to roughly doubling the number of real positive samples, though excessive upsampling hurts performance. Finally, we show a positive scaling trend, where larger models are more positive-sample-efficient to probe. Our findings suggest we should leverage the large number of negative samples available to amplify the signal from rare but critical misalignment examples.

cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2602.08557

Combining Constrained Sampling and Reinforcement Learning for Manipulation

Marc Toussaint, Cornelius V. Braun, Armand Jordana, Sayantan Auddy, Eckart Cobo-Briesewitz, Denis Shcherba, Tilman Burghoff, Justin Carpentier

原文摘要

Training manipulation policies in contact-rich settings is a core challenge in robotics. While Reinforcement Learning (RL) has demonstrated its strength in such settings, it may struggle to sufficiently explore and discover complex manipulation strategies. To address this, we combine two basic ideas: First, designing appropriate reset strategies (the start state distribution of episodes) has shown promise in improving RL exploration and effectiveness. Second, while model-based approaches to finding trajectories through manipulation are hard, recent work showed that model-based approaches to sampling states on constrained manifolds can be highly efficient. Based on these observations, we propose a novel state sampler that enables efficient goal-conditioned RL in complex contact-rich manipulation tasks. Our sampler explicitly takes into account the structure of contact in order to provide a rich covering of diverse contact modes. By combining constrained sampling resets with projected interpolation and curriculum learning, our novel approach outperforms RL without constrained sampling and alternative reset methods, and effectively trains universal, prehensile and non-prehensile manipulation policies in contact-rich whole-body and in-hand settings. See the project page https://www.user.tu-berlin.de/mtoussai/26-CSRL/ for videos and full source code access.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2608.24275

RePolicy: Learning to Select and Apply Safety Policies for Agent Safeguards

Houcheng Jiang, Boxuan Zhang, Qiyong Zhong, Junfeng Fang, Xiang Wang, Xiangnan He

原文摘要

Agent safeguards must determine whether an agent's behavior violates the safety rules of the current task, environment, and user. Because these rules vary across contexts, a safeguard needs to select the relevant policy before judging the behavior. Existing policy-aware safeguards can apply a supplied policy, but generally assume that the correct policy is already known. They therefore do not learn how to select a policy from a library, even though an incorrect choice can lead to an incorrect safety judgment. We introduce RePolicy, a safeguard that jointly learns policy selection and safety assessment. Given an agent trajectory and a set of candidate policies, RePolicy selects the policy that best matches the situation, applies its requirements, and predicts whether the behavior is safe. We construct PolicyTraj-20K, a dataset of more than 20,000 agent trajectories annotated with applicable policies, safety labels, and explanations. We train RePolicy with supervised demonstrations followed by reinforcement learning, while randomizing policy order, identifiers, and distractors to reduce shortcut learning. Across six agent-safety benchmarks and 19 baselines, RePolicy-4B achieves leading overall safety-detection performance and improves policy-selection accuracy. These results show that safeguards can learn which safety requirements to use for assessing agent behavior.

cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.36078

Learning on-the-fly quasi-stationarity diagnostics from low-dimensional surrogates

Noé Blassel

原文摘要

Accelerated dynamics methods, such as parallel replica dynamics, require estimates of the decorrelation time, after which a process trapped in one of its metastable states is distributed according to the quasi-stationary distribution (QSD). In a method proposed by Binder, Lelièvre and Simpson, decorrelation times are defined by coupling a particle approximation of the non-linear conditioned probability flow of the process with MCMC-inspired diagnostics. We explore an alternative to MCMC diagnostics, based on coarse-graining of non-linear probability flows and simulation-based inference. We first develop a coarse-graining framework for killed Markov jump processes conditioned on survival, proving uniform-in-time, non-asymptotic estimates approximating the non-linear probability flow of the true process by the non-linear flow of a surrogate process, defined with a collective variable. The framework does not assume reversibility of the true process, and also gives bounds which scale explicitly in an adiabatic regime. Motivated by these results, we then construct a quasi-stationarity diagnostic using simulation-based inference: a recurrent neural network (RNN), trained on particle approximations of non-linear effective flows, estimates posterior probabilities of decorrelation. The learned diagnostic is then transferred to collective-variable traces of high-dimensional particle approximations. We investigate the behavior and robustness of this diagnostic on a set of numerical experiments, comparing it to the MCMC diagnostic of Binder et al.\ on both synthetic examples and a molecular system, in and out of its training prior distribution. When its prior assumptions hold, the RNN diagnostic outperforms the MCMC diagnostic in ROC analysis, is less conservative, fails less often, and has an interpretable decision threshold.

stat.ML · cs.LG · math.PR

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2510.05071

Adaptive Modular Retrieval-Augmented Framework for Cross-Domain Image Classification

Debojyoti Ghosh, Soumya K Ghosh, Adrijit Goswami

原文摘要

Effective image classification in real-world scenarios requires models capable of capturing both local features and global context, while adapting to the evolving complexity of the input. This paper presents a hybrid deep learning system named AMR-Net (Adaptive Modular Retrieval Network) for adaptable image classification in environmental and industrial contexts. The proposed model integrates ResNet-based local feature extraction and Vision Transformer-based global representation learning with an FAISS-based memory retrieval module for contextual enhancement. Furthermore, AMR-Net has an adaptive modular extension mechanism to dynamically enhance network capacity when validation performance plateaus during training. The proposed framework was evaluated using a ten-class waste classification dataset and the Kolektor Surface Defect Dataset 2 (KolektorSDD2), achieving classification accuracies of 97.15% and 99.80%, respectively. Experimental results demonstrate that AMR-Net consistently outperforms mainstream convolutional and transformer-based approaches while maintaining robust cross-domain generalization. Component-wise ablation studies show that global attention (ViT), retrieval-based contextual learning (FAISS), and adaptive modular growth each provide complementary and partially overlapping contributions to overall performance, with global attention contributing the largest individual gain. The proposed architecture is a scalable and flexible solution for image classification challenges characterized by data unpredictability, class imbalance, and dynamic feature distributions.

cs.CV

查看论文
arXiv ↗PDF ↗