跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

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

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

统计口径与数据新鲜度

统计 2026-11-20 12:52 — 2026-11-21 12:52(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

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

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

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

浏览论文库

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

找到 24,181 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-21 12:17(北京时间)
新论文 · v1arXiv:2611.24994

What the model knows but does not say: monitoring medical LLMs through verbalized thoughts

Frederik Hauke, Jan Clusmann, Hanna Kreutzer, Christiane Kuhl, Tianyu Han, Felix Barajas Ordonez, Sven Nebelung, Jakob Nikolas Kather 等 9 位作者

原文摘要

Efforts to keep clinical language models safe have concentrated on what enters and leaves the model, but a manipulated model can return an answer that looks entirely ordinary. Here we show that the model's own internal state, rendered in plain language, exposes manipulations the output conceals. Using a natural-language autoencoder, we translate the activations of a multimodal LLM (Gemma-3-27B) into text and let an input-blind judge read this verbalized internal state (the thoughts channel) and the model's answer separately. Across two clinical attack objectives (omitting a pathology, covertly recommending a drug), two input modalities, and two delivery mechanisms, the thoughts channel caught the injection significantly more often than the answer in all eight injected conditions (p<0.05), while both channels held near-zero false-positive rates on clean inputs; the omission result carried over to Llama-3.3-70B. The thoughts channel also registered unsafe clinical requests as harmful while the model was still reading them, before it decided whether to refuse. None of this requires retraining: an off-the-shelf verbalizer and a general-purpose judge are enough.

cs.CL · cs.AI

版本更新 · v3arXiv:2604.24234

Graph-augmented Segmentation for Enhanced In-Situ Geometry Reconstruction of Complex Shapes in Laser Powder Bed Fusion

Stefano Raimondo, Matteo Bugatti, Marco Grasso

原文摘要

In situ geometric inspection in Laser Powder Bed Fusion of Metals (PBF-LB/M) opens new possibilities for anticipated defect detection and more efficient qualification. It relies on image segmentation to reconstruct layer-wise geometries and identify manufacturing deviations as the part is being built. However, existing approaches face a critical yet largely overlooked challenge: photometric out-of-distribution (OOD) generalization, namely the ability to maintain reconstruction accuracy when image appearance differs from the conditions represented during model training. This study represents the first exploration of a graph-augmented U-Net architecture for in situ geometric inspection in PBF-LB/M. The hybrid network integrates a Graph Neural Network (GNN) within the bottleneck to enhance reconstruction robustness under photometric distribution shifts. It is evaluated against conventional U-Net and active contour segmentation methods using two experimental case studies involving different lattice geometries, materials, and manufacturing setups, with X-ray computed tomography providing reference geometries. Generalization is assessed across different training and inference conditions and under controlled image perturbations. Results show that the UNet-GNN achieves higher reconstruction accuracy and consistency than the benchmark methods, with more pronounced improvements when training and inference conditions differ. Its advantages are confirmed across both case studies and under controlled photometric perturbations, while preliminary anomaly detection experiments indicate that improved robustness does not necessarily compromise sensitivity to geometric deviations. These findings highlight graph augmentation as a promising strategy for more reliable in situ geometric inspection under heterogeneous manufacturing conditions.

cs.CV

新论文 · v1arXiv:2611.24475

Nothing to Declare: Steering a single neuron increases concealment in AI teams

Edoardo Chidichimo, Felix P. S. Wallis

原文摘要

Recent research suggests that individual neurons can have interpretable roles in AI behaviour. In Qwen3.5-4B, steering along one neuron's output direction increased decisions to conceal information more in teams than alone across five scenarios, even when teammates provided no messages. When all members participated and received steering, collective concealment increased in four scenarios, while inter-agent messaging helped temper concealment under steering. Moreover, removing the neuron's contribution had negligible effects. Collective settings can therefore change how an internal intervention affects AI behaviour.

cs.CL · cs.AI

版本更新 · v3arXiv:2602.05902

CoreQ: Mismatch-Corrected Successive Rounding for LLM Quantization

Seohyeon Cha, Huancheng Chen, Dongjun Kim, Haoran Zhang, Kevin Chan, Gustavo de Veciana, Haris Vikalo

原文摘要

Post-training quantization (PTQ) enables efficient deployment of large language models by mapping pretrained weights to low-bit formats without retraining, typically using a small calibration set to minimize a layer-wise calibration objective. However, this sequential procedure induces a mismatch: errors from earlier quantized layers alter the inputs received by later layers, causing the activations to deviate from those of the full-precision model. Recent approaches introduce mismatch-aware calibration objectives to compensate for this effect, but leave open how much of the observed mismatch should shift each layer's calibration target. We show that fully applying the empirical mismatch correction can overfit finite calibration data, while using a fixed correction coefficient ignores variation in mismatch estimates across layers. To address these limitations, we propose CoreQ, a learning-free PTQ framework that uses a geometric decomposition of the mismatch to compute a closed-form, layer-adaptive correction coefficient. This coefficient controls how much of the observed mismatch enters each layer's calibration target, reducing overfitting to limited calibration data without coefficient tuning. CoreQ then applies efficient greedy successive rounding to the resulting triangular least-squares objective. Its bounded beam-search extension, K-CoreQ, trades modest additional compute for improved performance. Across multiple LLM families, scales, bit-widths, and quantization settings, CoreQ and K-CoreQ achieve lower perplexity and higher downstream accuracy than strong PTQ baselines. Code is available at GitHub.

cs.LG · cs.AI

版本更新 · v2arXiv:2607.04169

Neural LiDAR Bundle Adjustment

Chin Yung Anson Hon, Kaicheng Zhang, Sen Wang

原文摘要

Recent research has achieved remarkable novel view rendering and scene reconstruction results with Neural Radiance Field (NeRF), including extensions to the LiDAR modality. Few studies have, however, explored the key design differences between RGB NeRFs and LiDAR NeRFs, particularly considering their underlying working principles. In this work, we provide both theoretical and empirical evidence suggesting that the density of volume sampling plays a significant role in LiDAR NeRF. Based on this finding, we propose a novel Neural LiDAR Bundle Adjustment (NeLD-BA) algorithm, which is tailored using efficient volume sampling of LiDAR rays for joint optimization of LiDAR map and poses. Extensive experiments are performed using the Newer College and FusionPortable datasets to demonstrate the proposed NeLD-BA's state-of-the-art performance in multi-view point cloud registration and 3D mapping. We will open-source our code for the community.

cs.RO

新论文 · v1arXiv:2611.24470

When to Refine: Selective Search for Frozen Visual World Models

Jingni Huang

原文摘要

Planning with a visual world model requires deciding both which actions to execute and how long to search for them. We introduce Selective CEM Refinement (SCR), a learned rule for deciding whether to continue cross-entropy method (CEM) search with a frozen LeWorldModel. After five it- erations, the gate measures how far the mean first action chunk has moved, how much the mean terminal cost of elite plans has fallen, and how widely their first chunks vary. It returns the current plan for execution or continues the same search for ten more iterations. The gate requires no addi- tional world-model rollout. On 16 PushT scenes evaluated with two planning seeds, SCR succeeded in 25 of 32 trials, compared with 28 of 32 for fixed 15-iteration search (Fixed-15). All 25 SCR suc- cesses occurred in trials that Fixed-15 also solved, yield- ing 89.29% paired success retention. Across complete episodes, including replanning, SCR evaluated 23.81% fewer candidate plans and reduced mean controller time per eight-environment batch by 17.22%. Its observed suc- cess rate was 5.36 percentage points above an analytic mix- ture of the two fixed-depth controllers at the same expected candidate workload, although the difference changed sign across planning seeds. Compared with gating only at the initial solve under the same random-draw padding proto- col, gating again at replanning produced nine additional early exits. This reduced total candidate evaluations by 15.79%, while successes fell from 27 to 25.

cs.RO

新论文 · v1arXiv:2611.24469

ICICLE: In-Context Inference for Curve-Learning Extrapolation

Laurits Fredsgaard, Michael Riis Andersen, Mikkel N. Schmidt

原文摘要

Freeze-thaw Bayesian optimization allocates training budget by predicting how partially observed learning curves will continue. LC-PFN first applied prior-data fitted networks (PFNs) to learning-curve extrapolation, and FT-PFN extended this approach to configuration-aware freeze-thaw optimization with a ten-slot hyperparameter encoder. We introduce In-Context Inference for Curve-Learning Extrapolation (ICICLE), an in-context surrogate that accepts a variable number of hyperparameters through a shared set encoder and is trained on a synthetic prior covering typed hyperparameters, heavy-tailed noise, quantization, and divergence. ICICLE has 1.28 million parameters (8.7% of FT-PFN's count) and produces similar mean-regret trajectories to FT-PFN on LCBench, PD1, and TaskSet. On the 86-task Meta-Album benchmark with 33 hyperparameters, ICICLE reaches lower mean normalized regret than Quick-Tune overall and on extended, with no clear difference on mini or micro. It does so without pre-training or meta-training on Meta-Album tasks, whereas Quick-Tune is meta-trained on related Meta-Album tasks. Our release samples synthetic training tasks on the fly and provides a freeze-thaw scheduler for Ray Tune and an ONNX model for inference without JAX or PyTorch.

cs.LG

新论文 · v1arXiv:2611.24466

Cooperative UAV and Ground Station for Robust MEC of Dynamic Aquatic IoT: A Generative AI-Enhanced Approach

Jiahao You, Ziye Jia, Chao Dong, Qihui Wu, Zhu Han

原文摘要

The increasing aquatic Internet of Things (AIoTs) require computational support and coverage in applications such as maritime search and rescue. Unmanned aerial vehicles (UAVs) can offer low-cost, flexible aerial services, and ground stations (GSs) can provide powerful supports, which can cooperate to help the AIoTs in complex scenarios. However, the collaboration between UAVs and GSs for AIoTs faces challenges of task uncertainties, AIoTs trajectory uncertainties, heterogeneities, and limited computational resources. To address these issues, we propose a cooperative UAV and GS based robust multi-access edge computing (MEC) framework to assist AIoTs in completing computational tasks. Specifically, we formulate the optimization problem of joint task offloading and UAV trajectory to minimize the total execution time, which is in the form of mixed integer nonlinear programming and NP-hard to tackle. Therefore, we propose the algorithm of generative artificial intelligence-enhanced heterogeneous agent proximal policy optimization (GAI-HAPPO). The proposed algorithm integrates GAI models to enhance the actor network ability to model complex environments and extract high-level features, thereby allowing the algorithm to predict uncertainties and adapt to dynamic conditions. Additionally, GAI stabilizes the critic network, addressing the instability of multi-agent reinforcement learning approaches. Finally, extensive simulations demonstrate that the proposed algorithm outperforms the existing benchmark methods, thus highlighting the potentials in tackling intricate, cross-domain issues in the considered scenarios.

cs.NI · cs.LG

版本更新 · v2arXiv:2610.06651

Considering Context: When World Models Need Context Encoders

Oleg Smirnov, Sofiane Ennadir, John Pertoft, Bjartur Hjaltason, Sara Karimi

原文摘要

Methods for generalization in model-based reinforcement learning typically assume that an agent cannot recover the latent context governing the environment dynamics from its own experience, and therefore supplies it externally. We formalize and test this assumption with \emph{predictive sufficiency}, which quantifies what access to the context adds to next-step prediction under the visitation distribution an agent induces, and separates that quantity into a history-recoverable part and a residual requiring the true context, both distinct from a finite model's deficit. We classify context-aware algorithms by the predictive risk their conditioning set can target and demonstrate across environments of increasing identification difficulty that the headroom does not follow the MDP class. The same task under different priors leaves predictive headroom in one setting and nothing distinguishable from zero in another, where the agent's behavior implicitly identifies the context and any benefit of such a mechanism cannot be attributed to missing one-step predictive information. Where headroom persists, the learned state exposes it only partially, and adding the true context still lowers the risk. Our contribution is an analytical framework, instantiated by a practical criterion for matching contextual mechanisms to the predictive information available to them, estimated from the ordinary trained agent.

cs.LG

版本更新 · v3arXiv:2607.19987

UniRank: Benchmarking Ranking Models for Unified Sequential Modeling and Feature Interaction

Honghao Li, Xianquan Wang, Zibin Zhang, Yi Zhang, Kangyi Lin, Yiwen Zhang

原文摘要

Ranking is a core stage in online advertising and recommender systems. Modern ranking models increasingly unify sequential modeling and feature interaction, yet many advances rely on proprietary data, closed implementations, and large-scale industrial infrastructure. This setting limits reproducible comparison and hinders academic study of scaling laws, long-sequence modeling, and multi-task ranking. To address these limitations, this paper proposes UniRank, an open benchmark for ranking models that unify sequential modeling and feature interaction. UniRank uses chronological pointwise autoregressive supervision, standardizes evaluation across feedback tasks, and provides a PyTorch toolkit with Distributed Data Parallel training, operator optimization, mixed-precision training, attention optimization, and other efficiency techniques that reduce hardware requirements. We benchmark 15 representative unified ranking models on five public datasets from short-video, advertising, and e-commerce platforms, with the largest containing over 700 million instances and the longest behavior sequence exceeding $10^5$ interactions. UniRank provides a reproducible foundation for comparing unified ranking models, studying scaling laws under limited compute, and narrowing the gap between academic and industrial ranking research. We believe UniRank benefits researchers, practitioners, and beginners through reproducible experiments, production-oriented evaluation, and accessible implementations. Code and data are available at https://github.com/salmon1802/UniRank.

cs.IR

新论文 · v1arXiv:2611.24462

Anchoring with Visual Evidence: Explicit Semantic Anchors for Joint Change Detection and Captioning

Kai Zheng, Hang-Cheng Dong, Puyi Fang, Junjia Pan, Zhenkai Wu, Wei Zhang

原文摘要

Joint change detection and captioning seeks to localize changes and describe them from bi-temporal remote sensing images. A key challenge is translating dense visual evidence into the semantic conditions that guide description generation. We propose Explicit Semantic Anchors (ESA), a supervised interface that exposes entity presence, temporal appearance variation, and spatial extent. A prediction-guided readout combines visual features and predicted change probabilities, and three factor heads estimate the corresponding anchor activations. These activations guide generation through two complementary routes. The context route maps the three factor activations to three separate memory tokens, one per factor, and appends them to the dense visual memory, while the lexicon route produces an additive vocabulary bias at each decoding step. Dense visual memory remains available for spatial detail and information beyond the explicit factors. Anchor targets are derived from existing images and change annotations, requiring no additional anchor-level language annotation. At inference, all conditions are computed from the input image pair and the model's own predictions. Experiments on LEVIR-MCI and WHU-CDC show strong joint detection and captioning performance, supporting the use of supervised visual attributes as explicit conditions for change description generation.

cs.CV

新论文 · v1arXiv:2611.24457

ShoWorld: Shape-Faithful Multi-Hypothesis Occupancy World Model

Zhiqiang Yan, Guangfa Wang, Yuan Wu, Jian Yang, Gim Hee Lee

原文摘要

Occupancy world models struggle to forecast moving objects due to motion uncertainty and shape distortion. A single trajectory can average plausible futures into an implausible path, and repeated re-voxelization can erode or inflate object shape even under correct motion. We introduce ShoWorld to address these challenges through separate motion and shape representations. The motion forecaster predicts probabilistic trajectories conditioned on scene layout, nearby agents and the future ego trajectory. Each hypothesis is rendered into occupancy and fused according to its probability to retain alternative futures. The shape component predicts sub-voxel points with occupancy probabilities and offsets from local geometry. These points move rigidly with the object, and voxel occupancy follows the probability that at least one occupied point falls within it. This representation reduces shape distortion under motion. ShoWorld achieves state-of-the-art results on Occ3D-nuScenes. With ground-truth occupancy and ego trajectory, it improves mIoU and IoU by 6.56 and 9.16 points overall and by 6.06 and 5.53 points on dynamic classes.

cs.CV

新论文 · v1arXiv:2611.24453

Harmonic Circuit: Chord Progression Lies in a Sparse Principal Information Flow in Symbolic Music Models

Ruiji Yu, Gus Xia

原文摘要

Mechanistic Interpretability (MI) seeks to decode neural network computations into human-understandable logic. A core methodology in MI is circuit discovery: identifying the information flow in sub-networks that drive specific model behaviors. While circuit discovery has matured within Natural Language Processing (NLP), the interpretable circuits in music models remain largely unexplored. In this work, we identify a music-functional circuit within two symbolic music generative models: ChatMusician and MuPT. The music function we consider is the prediction of the terminal tonic root in a classical harmonic progression pattern $I \ldots V \rightarrow I$. We use the Auto-Emergence algorithm to extract a principal sparse sub-network as the identified circuit. Validation via causal intervention confirms that this identified circuit is a primary driver of harmonic progression behaviors in the symbolic music models. To our knowledge, this represents the first systematic circuit-discovery study conducted within the music domain. Our demo page is available at: https://ukiyo23rui.github.io/auto-emergence-demo/.

cs.SD · cs.AI · cs.LG

版本更新 · v3arXiv:2608.05049

OmniEdit-Bench: A Comprehensive Benchmark for Instruction-based Video Editing

Chenxuan Miao, Yutong Feng, Yi Lu, Yunfeng Yan, Donglian Qi, Shiwei Zhang, Yu Liu, Xi Chen 等 9 位作者

原文摘要

Instruction-based video editing (IVE) offers diverse applications, yet evaluating editing models remains challenging. Existing benchmarks suffer from two key limitations: narrow task coverage inherited from image editing that overlooks video-specific dimensions, and metrics that inadequately capture instruction fidelity, allowing visually plausible but incorrect edits to receive high scores. To address these limitations, we introduce OmniEdit-Bench, a comprehensive benchmark covering spatial, temporal, audio, and reference-based editing. It distinguishes between explicit and implicit instructions and incorporates reasoning-based scenarios to reflect real-world editing requirements. We further propose an evaluation framework that assesses accuracy, preservation, realism, and consistency using both human judges and a state-of-the-art vision-language model (VLM). An accuracy-aware penalty mechanism conditions the other three scores on accuracy, prioritizing instruction fidelity and penalizing plausible but incorrect edits. Experiments on prominent open-source and commercial IVE models reveal substantial room for improvement. OmniEdit-Bench provides a structured testbed for identifying model limitations and advancing instruction-based video editing. Project page: https://omniedit-bench.github.io.

cs.CV

新论文 · v1arXiv:2611.24445

Retroactive Concept Bottleneck Models for Exemplar-Free Class-Incremental Learning

Yang Ji, Ying Sun, Yuanxin Zhuang, Yuting Zhang

原文摘要

Exemplar-free class-incremental learning (EFCIL) requires models to learn new classes sequentially without retaining training samples from previous tasks. Continual concept bottleneck models (CBMs) make their predictions interpretable by identifying visual concepts in an image and using them to score each class. However, these methods typically focus on preserving learned concepts or approximating past classes with generated features, which can leave new concepts underused in old-class predictions. We propose RetroCBM, which retains a small set of vocabulary-independent statistics relating frozen visual features to the embedding space of a pretrained vision-language model. Even after earlier training images have been discarded, we can use these statistics to supervise new concepts. To do so, we develop a retroactive regression rule that yields the same predictor for each new concept as class-balanced ridge regression on all images seen so far. We then reproject the stored statistics to recover the full-history squared classification loss and thus learn how much each new concept should contribute to old-class predictions. Since the classifier may use systematic concept errors to distinguish nearby classes, we further introduce a graph-weighted penalty on the margin shifts caused by those errors. Across six datasets, RetroCBM outperforms state-of-the-art interpretable methods in all class-incremental settings, with an 18.1\% relative gain in mean average incremental accuracy over the strongest exemplar-free baseline.

cs.CV

新论文 · v1arXiv:2611.24443

NativeSeg: Native Vision Language Segmentation of 3D Computed Tomography via Spatial Visual Tokens and Verified Local Zoom

Chengqian Ma, Gaole Dai, Zijie Meng, Weixuan Jin, Yijie Du, Yuexiao Ma, Xiawu Zheng, Fei Chao

原文摘要

Text prompted 3D computed tomography (CT) segmentation requires both full slice context and fine boundary precision across volumes where over $80\%$ of organ prompted slices lack the queried target. Existing medical vision language models (VLMs) are largely non native segmenters: they delegate pixel delineation outside the VLM by compressing reasoning into a 1D [SEG] vector for an external Segment Anything Model (SAM) decoder, quantizing masks into lossy discrete codebooks, or issuing box prompts to stateless external SAM tools without empty slice screening or spatial state memory. We introduce NativeSeg, a native vision language segmentation framework on Low Rank Adaptation (LoRA) adapted Qwen3.8-27B that performs dense mask decoding, volumetric triage, spatial error self diagnosis, and multiround local zoom directly over the VLM's 2D spatial visual tokens without external SAM decoders. NativeSeg decodes confidence maps from upper layer 2D spatial tokens fused with a five slice U-Net pathway, halts at zero zooms on $>91\%$ of standalone evaluations via a native presence gate (<pres>), localizes uncertain boundaries via a 20 cell multiscale error gate (<err>), and accepts zoomed updates only when predicted error decreases. Trained solely on adult CT, NativeSeg improves Round $0$ DSC by $+19.81$ and $+7.04$ points over discrete codebook and 1D [SEG] readouts on the identical backbone, reaches $86.81\%$ adult and $87.04\%$ zero shot pediatric DSC, and safely refines 2D nnU-Net ($89.52\%$ adult; $86.86\%$ pediatric), 3D SegVol ($86.81\%$), and 3D VoxTell ($89.85\%$).

cs.CV

新论文 · v1arXiv:2611.24440

Explainable and CPU-Efficient Brain Tumor Classification from MRI Using Two-Phase EfficientNetB3 Fine-Tuning

Deepro Das, Saad Kabir Uddin, Mohammad Alauddin, Bornali Sarker, Thasnim Akhter

原文摘要

Brain tumors are among the most lethal neurological disorders, with over 308,000 new diagnoses annually worldwide. Accurate classification of tumor type from Magnetic Resonance Imaging (MRI) scans is critical for treatment planning; however, manual radiological interpretation is time-consuming and prone to inter-observer variability. This paper proposes an automated deep learning pipeline for four-class brain MRI classification (glioma, meningioma, no tumor, and pituitary tumor) using EfficientNetB3 transfer learning with a custom classification head. A two-phase training strategy is adopted: Phase 1 trains the head with a frozen backbone (best validation accuracy: 92.77%), and Phase 2 selectively fine-tunes the top 37 layers at a 100x reduced learning rate (best validation accuracy: 94.91%). On an independent balanced test set of 1,600 MRI images, the system achieves 90.62% overall accuracy and a macro-average AUC of 0.9854. Gradient-weighted Class Activation Mapping (Grad-CAM) is integrated for class-discriminative visual explainability. The proposed system simultaneously addresses three dimensions absent in prior work: high accuracy, CPU-only computational efficiency, and clinical interpretability.

cs.CV

新论文 · v1arXiv:2611.24436

From real to synthetic: PSG-GAN learns sleep apnea signal dynamics for AI training

Riku Huttunen, Samu Kainulainen, Matias Rusanen, Sami Myllymaa, Juha Töyräs, Timo Leppänen, Henri Korkalainen

原文摘要

Synthetic polysomnography (PSG) data has great potential in healthcare applications, for example in substituting real data that is sensitive and challenging to collect. Previous research on synthetic PSG data has been largely focused on short segments of individual signals, overlooking the inter-signal dynamics related to sleep disorders. In this study, we developed a generative adversarial network (PSG-GAN) for co-synthesizing longer segments of multiple PSG signals with diverse and complex characteristics typical of obstructive sleep apnea (OSA). The synthetic signals appeared realistic, and the electroencephalography and electrooculography band powers were similarly distributed across sleep stages. Moreover, the PSG-GAN learned to mirror signal interactions such as changes in photoplethysmography-based heart rate during and after breathing obstructions. An automatic scoring model trained solely with automatically scored synthetic signals reached good performance on sleep staging (\k{appa}=0.71 and 0.63) and respiratory event segmentation (\k{appa}=0.57 and 0.60), on two test sets consisting of healthy volunteers and OSA patients, respectively. This demonstrates the potential of synthetic data in supporting algorithm development for sleep diagnostics. The source code is available at https://github.com/rikuhuttunen/psg-gan.

cs.AI · cs.LG · physics.med-ph

版本更新 · v2arXiv:2610.03650

PoCoFL: A Framework for POlicy-COmpliant Federated Learning

Dominik Roy George, Varesh Mishra, Aysajan Abidin

原文摘要

Federated Learning (FL) is a privacy-oriented learning paradigm that enables collaborative model training while keeping training data local to clients. However, FL does not guarantee that clients submit policy-compliant contributions or that aggregators process admitted contributions correctly. Update-magnitude bounds alone cannot ensure compliance with learning-specific requirements involving private data, intermediate models, or retained state. These requirements differ across FL settings, motivating a unified framework to specify and verify client and aggregation policies. This work introduces PoCoFL, a policy-compliant federated learning framework that separates three aspects: (i) FL type, (ii) policy semantics, and (iii) cryptographic realisation. We provide a formalisation that captures client and aggregation requirements as policy-dependent relations, defining update admissibility, aggregation correctness, and policy-relative privacy. These relations enable verification of selected properties of intermediate objects, such as local-model loss, continual-learning state, and personalised optimisation conditions, in each round. Clients demonstrate compliance of their contributions using compatible cryptographic mechanisms, while aggregators prove that the recorded set of admitted contributions was processed according to the selected aggregation policy. We demonstrate PoCoFL through four formal instantiations and corresponding proof-of-concept realisations: (i) vanilla, (ii) continual, (iii) personalised, and (iv) threshold-encrypted federated learning. We evaluate how policy enforcement affects the learning objectives of vanilla, personalised, and continual FL, and quantify the computational and communication overhead of policy verification across all four instantiations. These results demonstrate that PoCoFL captures complex policy representations across the evaluated FL settings.

cs.CR · cs.LG

新论文 · v1arXiv:2611.24433

MLIE-CoT: Chain-of-Thought Reasoning via Multi-Level Image Editing

Baichuan Yang, Dongyuan Xu, Zhao Yang, Yong Liu, Bing Su

原文摘要

Chain-of-thought (CoT) reasoning has proven effective in improving the reasoning abilities of large language models by breaking complex tasks into step-by-step solutions. In recent years, researchers have attempted to extend CoT reasoning to visual generative models to improve their ability to handle complex tasks. However, existing CoT methods for text-to-image generation still rely mainly on textual reasoning and therefore cannot fully use and preserve the correct semantic information already present in an image, as even the most detailed language cannot capture all visual information. With advances in image generation, models such as FLUX and BAGEL can generate images conditioned on visual inputs. Recent work has attempted to carry images forward as context in the chain of thought, but this introduces another limitation: unsuitable earlier images can strongly interfere with the model, while repeated whole-image regeneration introduces new hallucinations at each round and causes them to accumulate. To address these issues, we propose MLIE-CoT, a visual chain-of-thought framework based on multi-level image editing. MLIE-CoT learns to select the repair granularity from three levels, preserving correct semantics whenever possible during reasoning while reducing the effect of incorrect content. To improve the ability of multimodal models to route errors of different severity to appropriate repair levels, we construct an image-repair dataset covering different error cases and use it to train a repair-level router. Extensive experiments demonstrate that MLIE-CoT significantly outperforms both its base models and methods using fixed repair strategies, producing clear improvements across multiple benchmarks.

cs.CV

新论文 · v1arXiv:2611.24431

Let a Small Model Read First: Faster Prefill for Long Interleaved Image-Text Inputs

Xiaolin Lin, Kai-Chiang Wu, Bing Li, Grace Li Zhang

原文摘要

Vision-language models (VLMs) can answer questions about long interleaved inputs, such as documents of many pages and collections of hundreds of images, but they are slow to start: on inputs of about 100K tokens, the prefill takes most of the wait for the first output token. A common approach prunes the input by scoring its tokens with the target model itself or with a small model of the same family that first writes an answer. However, this scoring takes much of the time that pruning saves. We find that the evidence for a question lies in a few whole units (images or paragraphs), and that a small hybrid model finds them with one attention head of an early layer, even at half image resolution. Based on these findings, we propose SkimPrefill, which lets the small hybrid model read the whole input first, up to that early layer, with images at half resolution. Most layers of the small hybrid model (Qwen3.5-0.8B) run in linear time, so this pass is short. SkimPrefill then ranks every unit with that head and gives the target the top units as an ordinary shorter prompt. One untrained selector serves targets of different model families. Extensive experiments on MMLongBench at 128K tokens, with 19 datasets of five task types, show the effectiveness of SkimPrefill. With 10% of the input tokens kept, SkimPrefill speeds up the prefill of Qwen3.5-9B, Gemma-4-12B and Qwen3-VL-32B by 3.7 times, 11.6 time and 8.2 times while maintaining their average accuracy.

cs.CV · cs.AI

版本更新 · v2arXiv:2611.03404

VidDub: Replacement-Aware Supervision for Two-Step Video Dubbing

Yexin Liu, Yunlong Lin, Zehan Wang, Guangxin He, Xinli Xu, Jiajun Zha, Yaofu Liu, Yueze Wang 等 14 位作者

原文摘要

Video dubbing aims to synthesize lip movements that match target speech while preserving the speaker identity, head pose, and background of a source video. Existing editors are predominantly trained via self-reconstruction under original paired audio, even though the source mouth becomes a conflicting target when driving speech is replaced. We introduce \emph{replacement-aware cross-audio supervision}: paired-audio samples use full-frame reconstruction, whereas unpaired replacement-audio samples mask out source-mouth reconstruction losses and supervise the mouth solely through audio--visual synchronization while preserving unmasked facial regions and background. Because offline dubbing provides the complete source video and target speech beforehand, we present \method, a full-context diffusion editor that uses matched trajectory regression to transfer CFG-guided teacher trajectories to a two-step student and distribution matching to recover visual quality after compression. We will release the curated training dataset, VidDub-Bench, fixed paired- and cross-audio evaluation manifests, preprocessing outputs, and unified evaluation code. Across three training seeds, masking out conflicting source-mouth reconstruction improves cross-audio Sync-C by 0.833. At two NFE, the final student provides the strongest cross-audio synchronization and visual-speech recognition among the evaluated methods, with competitive visual fidelity and source preservation.

cs.CV

版本更新 · v2arXiv:2605.10313

Signature Approach for Contextual Bandits with Nonlinear and Path-dependent Rewards: Regret Bound and Statistical Inference

Xin Guo, Grace He, Xinyu Li

原文摘要

We study contextual bandits with nonlinear and path-dependent rewards through a novel signature-transform-based approach. Leveraging the universal nonlinearity property of signatures, we approximate continuous path-dependent reward functionals by linear functionals in the signature space. This representation enables the use of efficient linear contextual bandit methods while preserving expressive sequential structure. Building on this framework, we propose \texttt{DisSigUCB}, a signature-based disjoint upper confidence bound (UCB) algorithm. Under boundedness and non-degeneracy assumptions, we establish a high-probability regret bound of order \(\tilde\mathcal{O}\bigl(d_{\text{Sig}} \sqrt{KT} + \ varepsilon_N T \sqrt{d_\Sig}\bigr)\) with \(K\) being the number of actions, \(d_{\text{Sig}} \) the signature feature dimension, and \(\varepsilon_N\) the finite-depth approximation error. We further introduce $ζ$-\texttt{DisSigUCB}, a randomized variant that ensures persistent exploration for statistical inference. Under additional margin and long-run design conditions, we prove consistency and asymptotic normality of the action-specific ridge estimators. Both the synthetic experiments and the numerical studies on temperature sensor monitoring, sleep-stage classification, and hospital nurse staffing demonstrate the effectiveness of \texttt{DisSigUCB} over linear and kernelized contextual bandit baselines in nonlinear and path-dependent settings.

cs.LG · math.OC

版本更新 · v29arXiv:2410.01223

Statistical Taylor Expansion: A New and Path-Independent Method for Uncertainty Analysis

Chengpu Wang

原文摘要

Statistical Taylor expansion is a rigorous extension of conventional Taylor expansion that replaces each precise input variable with a random variable of known distribution and sample count, then computes the mean, deviation, and a bounding reliability of every result. By tracking the propagation of input uncertainties through all intermediate steps, it renders the final result path-independent, with precise quantification of the tracking quality. This path-independence sets it fundamentally apart from conventional numerical approaches, which are path-dependent. This study presents an implementation called variance arithmetic and demonstrates its performance across diverse mathematical applications. This study also reveals the potentially substantial impact of numerical errors in library functions, the defect of applying input uncertainties as weights in conventional regression, and the modeling error of the discrete Fourier transformation. The concept of statistical algebra is also introduced.

stat.CO · cs.LG

版本更新 · v5arXiv:2605.15995

What competing constraints reveal about representation learning

Gwenolé Quellec

原文摘要

Learning latent states from temporal, multimodal, and partially observed data requires decisions about what information representations should retain, discard, and organize. I argue that six complementary properties help clarify these decisions: predictive sufficiency, minimality, temporal coherence, observation compatibility, nuisance invariance, and structural constraints. Drawing on established approaches, I distinguish desired representation properties from the objectives used to encourage them and the diagnostics used to assess them, since neither a training objective nor an evaluation metric guarantees that a property has been achieved. Controlled synthetic experiments examine how training objectives influence predictive performance, nuisance decodability, and latent organization. The results reveal discrepancies between seemingly favorable diagnostics and the information retained in learned representations, as well as latent-coordinate ambiguities under equivalent predictive performance, illustrating how complementary evaluations can improve interpretation and guide design choices. Together, these analyses provide a basis for designing, comparing, and evaluating learned latent states.

cs.LG · cs.AI

新论文 · v1arXiv:2611.24428

Keeping Shielded Data-Driven Controllers Safe Under System Evolution via Model Consistency

Roland Stolz, Vitus Lüntzel, Romain Pascual, Eric Sax, Matthias Althoff

原文摘要

Cyber-physical systems (CPS) increasingly contain data-driven controllers, which raise safety concerns given their black-box nature. A common mitigation is to enforce system safety via a shield around the controller, which restricts the inputs based on assumptions about the physical system. However, these assumptions are usually fixed and may become invalid as the system evolves (during development or due to component degradation), possibly leading to unsafe behavior. We address this limitation in a model-driven approach that treats both the data-driven controller and safety shield as first-class models that are integrated into a virtual single underlying model (V-SUM) alongside system models. We describe system evolution as changes in the associated models, such that consistency checking yields a continuous assessment of safety guarantees. In case of inconsistencies, consistency-preservation rules automatically update the models and restore a valid safety shield. We evaluate our approach on a reach-avoid task for a unicycle system. The results show that consistency-driven evolution effectively maintains safety guarantees under system changes and provides a principled basis for evaluating and restoring performance.

eess.SY · cs.AI

新论文 · v1arXiv:2611.24426

DyNA-Mixer: Decoupling Time and Topology for Scalable Dynamic Node Affinity Prediction

Yuanxing Zheng, Liping Wang, Xuemin Lin

原文摘要

Continuous-time dynamic graphs preserve the ordering, timing, and evolving strength of interactions that static snapshots discard, making dynamic graph mining essential for evolving systems such as recommender systems, trade networks, and transaction platforms. Research in this area has centered on dynamic link prediction, which forecasts only whether a future interaction will occur, whereas many applications also need its intensity. The dynamic node affinity prediction (DNAP) problem generalizes this task to continuous targets and full-candidate ranking, making it more broadly applicable but substantially harder. This raises two coupled difficulties: ranking requires a predictor expressive enough to capture both multi-scale temporal evolution and graph-wide collaborative patterns, yet running such a predictor over every source--destination pair is computationally prohibitive. We propose DyNA-Mixer, which decouples temporal forecasting from collaborative structure modeling. It adopts a multi-scale temporal predictor for pair-specific affinity evolution, constructs a Virtual Collaborative Graph that infers query-time relationships and aggregates trends from related interactions, and fuses the two through a gated residual mechanism. A sparsity-aware evaluation strategy further restricts expensive forecasting to pairs with recent non-zero affinities. Experiments on four Temporal Graph Benchmark datasets for the DNAP problem show that DyNA-Mixer achieves new state-of-the-art NDCG@10 on every dataset, including a 17.8\% relative test improvement over the strongest baseline on tgbn-token, the largest dataset.

cs.LG

新论文 · v1arXiv:2611.24425

Predicting Process-Level Behavior: A Benchmarking Study

Michael Grohs, Lukas Kirchdorfer, Jana-Rebecca Rehse

原文摘要

Predictive process monitoring (PPM) typically forecasts the future of individual process instances, hence operating on trace level. Process managers, however, typically operate on process level, e.g., when assigning staff or calculating KPIs. The question is how well such process-level managerial decisions are supported by existing approaches, given that PPM research has so far not focused on process-level predictions. We answer this question through a comprehensive benchmarking study of two process-level prediction tasks: concurrent trace prediction and average throughput time prediction. Across synthetic and real-life event logs, we compare statistical and machine-learning-based methods, foundation models, business process simulation, and, importantly, established trace-level PPM approaches, aggregating their predictions to the process-level. We assess which modeling paradigms are most suitable for predicting process-level behavior and whether existing trace-level PPM can be effectively repurposed for this task. Our results show substantial variation in performance across approaches. Notably, aggregated trace-level predictions cause errors to compound on process-level, reducing accuracy and highlighting process-level prediction as a distinct predictive task that warrants dedicated research.

cs.LG · cs.SE

新论文 · v1arXiv:2611.24424

Constructive Infinitely Differentiable Guidance Vector Fields for Robotic Control

Vinicius Mariano Goncalves, Peilin Wu, Adriano Rezende, Weijia Yao

原文摘要

Guidance vector fields (GVFs) are a popular framework for tasks such as robotic path-following and multi-robot motion coordination, but existing approaches may be non-differentiable or have large derivatives, limiting their use in controllers that require derivatives of the guidance signal, which is common when the GVF is used as an input in a cascaded controller. This paper proposes an infinitely differentiable constructive GVF. The resulting vector field depends on a smoothing parameter and is a generalization of [1], recovering the original GVF in the limit as this parameter approaches zero. It also admits a simple and efficient implementation that preserves smoothness under discretization and can reduce computation time by using substantially fewer curve samples while maintaining small path-following errors. We establish the main theoretical properties and demonstrate its practical advantages through numerical and real-world experiments.

cs.RO

新论文 · v1arXiv:2611.24423

Probing Physics in Motion Diffusion Models Against Their Own Estimate

Saeed Ghorbani

原文摘要

Recent studies report that video diffusion models "know more than they show": physical properties decode better from their activations than from their inputs or outputs. But a diffusion model computes its clean-sample estimate from those same activations, so such a comparison measures how easily a readout extracts the information. Whether the model hides anything is a separate question. We set out to find hidden physical knowledge in kinematic text-to-motion diffusion models, where the physics of every intermediate estimate can be computed exactly, and recorded every layer along the sampling trajectories of MDM (two checkpoints) and MotionDiffuse. Against the usual baselines the layers do look ahead of the model's own estimate on balance and flight. With every readout tuned the same way, the estimate closes the gap: for balance and flight, adding a layer gains at most 0.01, a size that a positive control shows we would detect. The fall rate of jumps is the one clear exception. Early in denoising, MDM's layers add about 0.08 in $R^2$ (up to 0.11 in a refit) that the estimate does not yet show, and the gain fades as sampling proceeds; on MotionDiffuse it is small and not robust. The estimate also flags floating samples without training, but aborting them saves under a tenth of the compute. In a video model, plausibility that linear probes miss in Wan 2.1's clean latent becomes readable after one fixed nonlinearity.

cs.LG · cs.CV · cs.GR