跳到正文
研踪PaperTrail

版本更新

查看已发布论文的最新修订,在详情页对照实际观测到的版本。

部分采集异常

全站更新概况

+1084新增论文
+883其中 AI 相关
+130已有论文修订
+94新增 AI 导读

arXiv 最近完成扫描:2026-11-18 04:28 · 全站最近入库:2026-11-18 19:45各信源进度 →

统计口径与数据新鲜度

统计 2026-11-18 06:20 — 2026-11-19 06:20(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:2026-11-18 04:28最近发现新增版本:2026-11-18 19:45库内论文最新更新时间:2026-11-18 08:00

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。

26,870 已收录论文22,374 AI 相关8,156 有修订112 HF 社区发现

论文修订列表

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 8,156 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-18 04:27(北京时间)
版本更新 · v3arXiv:2606.28351

HyperSU:面向检索增强生成的语料驱动语义单元超图

HyperSU: Corpus-Driven Semantic-Unit Hypergraph for Retrieval-Augmented Generation

Jiate Liu, Liuyi Chen, Zhengyi Yang, Chuan He, Mingchen Ju, Bocheng Han, Ruyi Liu, Xu Zhou

AI 速读 基于摘要

HyperSU 用源文本语义单元构建超边、以线索引导双向检索,避免 LLM 摘要幻觉与检索扩散噪声,作者报告在 GraphRAG Bench 上平均准确率优于基线。

cs.IR

查看原文摘要

Recent Hypergraph-based retrieval-augmented generation (HyperRAG) methods use hyper edges to connect multiple entities simultaneously, enabling more efficient multi-entity evidence organization than pairwise graph structures. However, existing HyperRAG methods often rely on LLM-generated summaries to construct hyperedges, which can introduce hallucinations while also incurring high indexing costs. In addition, during retrieval, existing methods typically rely on either one-hop neighbor expansion or PageRank diffusion. The former may miss useful multi-hop evidence, while the latter can suffer from uncontrolled propagation over excessive hub nodes, leading to semantic drift and noisy reasoning chains. To address these challenges, we propose HyperSU, a novel hypergraph-based RAG framework featuring semantic-unit hyperedges and clue guided bidirectional retrieval. During construction, HyperSU formulates hyperedge construction as an entity-aware minimum-description length (MDL) optimization problem, inducing source-grounded semantic-unit hyperedges that balance sentence-level semantic coherence and entity compactness. It then constructs a hypergraph by modeling each semantic unit as a hyperedge over its co-mentioned entities. During retrieval, HyperSU performs clue-guided bidirectional expansion over the semantic-unit hypergraph, enabling both multi-hop evidence discovery and evidence-anchored noise reduction. Experiments show that HyperSU consistently improves answer accuracy over standard, graph-based, and hypergraph-based RAG baselines, raising average accuracy on GraphRAG Bench by 5.5 points over the strongest baseline (68.7 vs. 63.2), with larger gains on reasoning intensive tasks, where the relative improvement reaches 14.1%.

版本更新 · v4arXiv:2402.08106

平均场极小极大问题的镜像下降-上升算法

Mirror Descent-Ascent for mean-field min-max problems

Razvan-Andrei Lascu, Mateusz B. Majka, Łukasz Szpruch

AI 速读 基于摘要

研究测度空间上极小极大问题的两种镜像下降-上升(MDA)变体,并给出非渐近收敛速率。

math.OC · cs.LG · math.PR

查看原文摘要

We study two variants of the mirror descent-ascent (MDA) algorithm for solving min-max problems on the space of measures: simultaneous and alternating. We work under assumptions of convexity-concavity and relative smoothness of the payoff function with respect to a suitable Bregman divergence, defined on the space of measures via flat derivatives. We establish non-asymptotic convergence rates to mixed Nash equilibria, measured in the Nikaidô-Isoda error, proving an $\mathcal{O}(N^{-1/2})$ rate for simultaneous MDA and an improved $\mathcal{O}(N^{-2/3})$ rate for alternating MDA. The main technical contribution is an infinite-dimensional dual space analysis that relates Bregman divergences on measures to dual Bregman divergences on spaces of bounded continuous functions, allowing us to control asymmetric commutator terms created by alternating updates. The results substantially generalize prior analyses restricted to bilinear objectives and also apply to nonlinear convex-concave problems on measure spaces, thereby providing a unified theoretical foundation for MDA in mean-field min-max optimization.

版本更新 · v2arXiv:2610.31338

ValuSpec:面向树式推测解码的目标验证前即插即用候选评估

ValuSpec: Plug-and-Play Candidate Valuation before Target Verification for Tree-Based Speculative Decoding

Liang He, SiYuan Ma, Qishi Zhan, Yongqi Fan, Mingyu Cao, Qizhen Lan, Zhaolu Kang, Xilu Wang

AI 速读 基于摘要

提出一种即插即用的候选评估机制,在目标模型验证前过滤候选树中的低质量分支,以提升树式推测解码的验证效率。

cs.AI

查看原文摘要

Tree-based speculative decoding accelerates large language model inference by verifying multiple candidate branches in parallel. However, in practical serving scenarios, similar requests repeatedly induce overlapping candidate branches. While retrieval-based methods mitigate this by reusing historical drafts, the resulting candidate trees often contain many low-quality branches as the candidate tree grows, which reduces verification efficiency and wastes target-side computation. To address this, we propose ValuSpec, a plug-and-play candidate valuation mechanism that filters low-quality branches from a supplied candidate tree before expensive target-side verification. ValuSpec introduces transfer-based supervision to align filtering with the target model's behavior. A filtering threshold is estimated from the target model's greedy trajectories on a calibration split, without any parameter training. The target model then verifies the filtered tree under the output-identical greedy rule, so accepted tokens are guaranteed to match standard autoregressive decoding exactly. Experimental results on HumanEval, GSM8K, Dolly, and SpecBench show that, in our main evaluation, ValuSpec achieves up to 2.38x end-to-end speedup in retrieval-based tree settings. Furthermore, integrating ValuSpec into the EAGLE-3 dynamic tree framework yields up to 2.78x end-to-end speedup without modifying its tree construction strategy, demonstrating that ValuSpec serves as a generally applicable and plug-and-play candidate valuation module.

版本更新 · v2arXiv:2605.19049

KVBuffer:面向线性注意力的IO感知服务机制

KVBuffer: IO-aware Serving for Linear Attention

Longwei Zou, Lin Zhong

AI 速读 基于摘要

KVBuffer通过缓存近期键值,为线性注意力模型提供更灵活、更省内存的推理服务方式。

cs.LG · cs.AI

查看原文摘要

Linear attention has recently gained significant attention for long-context inference due to its constant decoding cost with respect to context length. However, existing serving systems typically serve linear attention by recurrently computing and updating a large linear attention state in every decoding step. Since the state is much larger than the per-token key and value, recurrent decoding incurs substantial memory access and becomes inefficient for serving linear attention. In this paper, we propose KVBuffer, an IO-aware serving mechanism for linear attention. By buffering recent keys and values, KVBuffer enables serving systems to compute linear attention outputs in more flexible and memory-efficient ways. For decoding, KVBuffer enables chunkwise computation, which reduces average memory access and decoding latency by deferring state updates and applying them in batch. For speculative decoding, KVBuffer verifies draft tokens in parallel and avoids storing temporary states. For short contexts, KVBuffer computes attention outputs directly from buffered keys and values, without creating or updating the linear attention state. We implement KVBuffer in SGLang for Qwen3.5 and Nemotron-3 series models. Our evaluations show that KVBuffer can speedup batch decoding by up to 1.47x and increase the maximum number of serving requests by 5x for speculative decoding when verifying four draft tokens.

版本更新 · v2arXiv:2602.03478

当LLM路由器过度付费:强模型偏好的代价

When LLM Routers Overpay: The Cost of Strong-Model Preference

Guannan Lai, Long Chen, Han-Jia Ye

AI 速读 基于摘要

作者发现宽松成本预算下现有LLM路由器会过度选择最强最贵模型,并提出EquiRouter以在效用相当时优先选择低成本模型。

cs.AI

查看原文摘要

LLM routing aims to reduce inference costs by assigning easy queries to less expensive models and reserving stronger models for cases in which they are needed. However, we find that existing routers often violate this principle under loose cost budgets: as the budget increases, they increasingly select the strongest and most expensive model even when less expensive feasible candidates provide comparable or identical utility. We refer to this behavior as strong-model over-selection under loose budgets. Our analysis shows that ambiguity alone does not explain this phenomenon. Substantial overpayment can arise when differently priced models offer equivalent or near-equivalent utility and the router retains a persistent model-level preference for costly models; budget relaxation exposes this preference by making premium candidates feasible. Motivated by this diagnosis, we propose EquiRouter, a decision-aligned router that learns query-dependent model rankings, preserves meaningful utility differences, and explicitly prefers lower-cost models among equivalent candidates through a Cost-Aware Ranking Objective and a lightweight Query--Model Interaction Representation. Experiments across diverse text, multimodal, continuous-utility, and large-model-pool settings show that EquiRouter consistently improves cost efficiency while maintaining competitive peak and whole-curve performance. Code is available at https://github.com/LAMDA-Model-Reuse/EquiRouter.

版本更新 · v2arXiv:2609.33928

在基于组织学的空间基因表达预测中保留DEG排序以促进基因发现

Preserving DEG Rankings for Gene Discovery in Histology-Based Spatial Gene Expression Prediction

Kaito Shiku, Kazuya Nishimura, Yasuhiro Kojima, Ryoma Bise

AI 速读 基于摘要

作者提出图像基差异表达排序(IDER)任务及可微目标,使组织学图像预测的空间表达更贴合差异表达基因发现的下游用途。

cs.CV

查看原文摘要

Predicting spatial gene expression from histology images could scale spatial transcriptomics (ST) to image-only cohorts, but conventional histology-based ST prediction is trained and evaluated mainly by per-gene spatial-profile reconstruction. This objective is misaligned with a key downstream use of ST: differentially expressed gene (DEG) discovery, where genes are ranked for a biological or morphology-defined contrast by evidence of between-group expression differences. We formulate image-based differential expression ranking (IDER), which asks whether predicted expression profiles preserve the contrast-specific ranked gene list obtained from measured profiles. IDER compares gene rankings induced by differential-expression statistics, rather than raw expression magnitudes or per-gene spatial correlations. We further introduce a differentiable IDER objective that aligns these statistics across genes and can be trained with morphology-derived proxy contrasts without predefined biological group labels. Experiments on public ST datasets show improved DEG-ranking agreement and pathway-enrichment overlap over conventional reconstruction objectives, including morphology-derived and pathologist-annotated tissue-region evaluations.

版本更新 · v2arXiv:2610.35766

信号改善、编码器退化:自监督学习中无标签监控的已知答案检验

Improving Signals, Declining Encoder: A Known-Answer Test of Label-Free Monitoring in Self-Supervised Learning

Nilotpal Kakati

AI 速读 基于摘要

作者用已知生成因子的合成图像训练LeJEPA编码器,检验损失与无标签特征统计能否察觉编码器已丢失部分所学信息。

cs.LG · cs.CV · physics.data-an

查看原文摘要

Self-supervised training is monitored without labels, through the loss and statistics of the learned features. We report a case where we know the encoders lost part of what they had learned, so we can check whether these signals noticed. We trained LeJEPA encoders on synthetic images with known generating factors. In nine of ten seeds, the readability of those factors fell abruptly and stayed down, nearly always along the same direction in factor space. Meanwhile the loss kept falling, and most of the eleven label-free signals we tracked kept improving and rated the declining runs better than runs that did not decline. Standard change-point detection rarely flagged the decline near its onset, and selecting checkpoints by these signals mostly chose ones after it.

版本更新 · v2arXiv:2611.02902

TailorRig:面向服装感知绑定的可控结构化辅助骨架生成

TailorRig: Controllable and Structured Auxiliary Skeleton Generation for Garment-Aware Rigging

Yajing Chen, Xin Chen, Saidan Lu, Yeqi Yuan, Weikai Chen, Changyu Hu, Cheng Zeng, Xin Wang

AI 速读 基于摘要

TailorRig 将服装感知的人形绑定重构为结构化、可控的完整骨架生成问题,通过用户指定身体骨架模板与语义标签驱动自回归生成器合成辅助骨链,并用层级链到骨权重分配减少权重渗漏。

cs.CV

查看原文摘要

Humanoid rigging for modern characters with complex garments remains challenging due to the need for dense auxiliary skeletons and precise skinning control. Existing automatic methods primarily recover rigs from geometry, providing limited artist control over auxiliary articulation and often lacking explicit semantic or kinematic-chain structure for dense garment bones, leading to instability and weight bleeding. We present TailorRig, a framework that reformulates garment-aware humanoid rigging as a structured, controllable full-rig generation problem. This formulation shifts the goal from automatic rig recovery to artist-driven rig authoring, where both body-template compatibility and auxiliary articulation are explicit design choices. TailorRig exposes two explicit axes of artist control within a unified framework: a user-specified body skeleton template that preserves body-rig compatibility, and artist-provided semantic labels that drive an autoregressive generator to synthesize requested category-specific auxiliary bone chains. To robustly assign skinning weights under dense skeletal configurations, we further introduce a hierarchical chain-to-bone formulation: vertices are first associated to a kinematic chain, and bone weights are then refined within that chain, reducing weight bleeding between spatially adjacent but semantically unrelated structures. We evaluate TailorRig on a diverse set of characters and introduce Human-Mix, a large-scale dataset of over 38,000 humanoid models curated for garment-aware rigging. Experiments show that TailorRig improves skeleton accuracy, deformation stability, controllability, and practical usability over existing methods. Project page: https://xxxx0601.github.io/trig/ ; code: https://github.com/Tencent/TailorRig ; dataset: https://huggingface.co/datasets/tencent/TailorRig

版本更新 · v2arXiv:2601.21629

面向量子电路优化通道自适应组合的强化学习

Reinforcement Learning for Adaptive Composition of Quantum Circuit Optimisation Passes

Daniel Mills, Ifan Williams, Jacob Swain, Gabriel Matos, Enrico Rinaldi, Alexander Koziell-Pipe

AI 速读 基于摘要

作者提出并演示用强化学习智能体为量子电路自动组合优化通道序列,以降低双量子比特门数量。

quant-ph · cs.LG

查看原文摘要

Many quantum software development kits provide a suite of circuit optimisation passes. These passes have been highly optimised and tested in isolation. However, the order in which they are applied is left to the user, or else defined in general-purpose default pass sequences. While general-purpose sequences miss opportunities for optimisation which are particular to individual circuits, designing pass sequences bespoke to particular circuits requires exceptional knowledge about quantum circuit design and optimisation. Here we propose and demonstrate training a reinforcement learning agent to compose optimisation-pass sequences. In particular the agent's action space consists of passes for two-qubit gate count reduction used in default PyTKET pass sequences. For the circuits in our diverse test set, the (mean, median) fraction of two-qubit gates removed by the agent is $(57.7\%, \ 56.7 \%)$, compared to $(41.8 \%, \ 50.0 \%)$ for the next best default pass sequence. In doing so our model demonstrates the ability to forgo reward to improve the final two-qubit gate count. Our approach clearly separates the problem of designing circuit optimisation passes from the problem of using them optimally in practice. We expect this separation of concerns to facilitate innovations in both areas.

版本更新 · v2arXiv:2610.12532

LLM驱动音色生成与插值中的结构与表征能动性

Structural and Representational Agency in LLM-Driven Timbre Generation and Interpolation

Joseph Cameron, Alan Blackwell

AI 速读 基于摘要

在一项LLM驱动音色界面的测量研究中,作者报告用户实际能改变的声音空间与界面所报告的位置可以分离:前者由结构精确刻画,后者存在显著误差。

cs.SD · cs.HC · cs.LG · cs.MM · eess.AS

查看原文摘要

Generative creative interfaces routinely report the user's position within a space of possibilities. Such displays imply that the quantities shown were measured from the artefact; in a growing class of systems they are instead predicted by a language model describing its own output, and the two are indistinguishable on screen. Most creative domains lack an agreed measurement against which such a self-report can be checked; audio admits one. In a measurement study of a working LLM-driven timbre interface, what a user can cause proves separable from what the interface reports. The first is exactly characterisable: the reachable set is a zonotope in parameter space whose dimension increases by one per sound placed, so the space the model may act within is demonstrably the user's own. The second is inaccurate. The displayed position differs from measurement by approximately three times the width of that region, and the error has two independent sources: model-predicted coordinates, whose accuracy varies twofold between the two models evaluated, and the interpolation step, which arises from combining synthesiser parameters and is unaffected by model quality. End to end, 27 of 36 refinement instructions produced no perceptible change, and under the weaker model the facility performed no action while reporting convergence. Effective agency (what a user can cause) and epistemic agency (whether they can know it) are therefore separable and independently engineered: structurally enforced constraints survive model failure, whereas guarantees resting on the model's own report do not. The failure mode that remains is silence presented as success. We close with a short protocol for building and auditing such displays.

版本更新 · v2arXiv:2605.13675

跨视觉模型的通用物体表征刻画

Characterizing universal object representations across vision models

Florian P. Mahner, Johannes Roth, Ka Chun Lam, Michael F. Bonner, Francisco Pereira, Martin N. Hebart

AI 速读 基于摘要

作者将162个视觉模型的物体相似性结构分解为非负维度,发现跨模型反复出现的“通用维度”更可解释、更受概念性图像属性驱动,且与生物视觉表征更吻合。

cs.CV · cs.LG · q-bio.NC

查看原文摘要

Deep neural networks trained with different architectures, objectives, and datasets have been reported to converge on similar visual representations. However, what remains unknown is which visual properties models actually converge on and which factors may underlie this convergence. To address this, we decompose the object similarity structure of 162 diverse vision models into a small set of non-negative dimensions. To determine universal versus model-specific dimensions, we then estimate how often each dimension reappears across models. In contrast to model-specific dimensions, universal dimensions are more interpretable and more strongly driven by conceptual image properties, indicating the relevance of interpretability and semantic content as implicit factors driving universality across models. Differences in architecture, objective function, training data, model size, and model performance do not explain the emergence of universal dimensions. However, models with more universal dimensions also better predict macaque IT activity and human similarity judgments, suggesting that universality reflects representations relevant to biological vision. These findings have important implications for understanding the emergent representations underlying deep neural network models and their alignment with biological vision.

版本更新 · v2arXiv:2605.30472

你的多模态语音模型说我长了一张适合电台的脸

Your Multimodal Speech Model Says I Have a Face for Radio

Maya K. Nachesa, Vlad Niculae, Vagrant Gautam

AI 速读 基于摘要

作者提出首个多模态语音识别偏见评估,用唇同步将不同人脸与同一音频配对,测量转录准确率的变化。

cs.CL

查看原文摘要

As large neural models have become better at language tasks, researchers are increasingly building multi- and omnimodal models that handle more than one modality of data. One example is the expansion of speech recognition models to audio-visual data, the goal being noise mitigation and multimodal subtitling. While performance and bias have been studied extensively in the single-modality regime, it is unknown how new modalities affect this, even though they produce biases in humans. We therefore propose the first bias evaluation of multimodal speech recognition, where we create videos pairing different faces with the same audio using lip-sync, and measure changes in speech transcription accuracy. We find large quality-of-service differences across mWhisper-Flamingo and Gemini models, with drops of up to 4.05 word error rate points, across self-declared gender, ethnicity, and their intersection. Our findings point to a priority for developers to evaluate, fix, and communicate such limitations, as providing more signals through additional modalities is not necessarily better, and may even lead to biased outcomes.

版本更新 · v2arXiv:2608.01896

GeoCore-9B:面向地球观测的地理感知生成式基础模型

GeoCore-9B: Towards Geo-Aware Generative Foundation Models in Earth Observation

Jeonghyeok Do, Munchurl Kim

AI 速读 基于摘要

作者提出一个90亿参数、完全基于地球观测数据从零训练的生成式基础模型,用Flow Matching扩散Transformer原生融合文本与地理空间元数据条件。

cs.CV

查看原文摘要

Existing generative models for earth observation (EO) predominantly rely on fine-tuning natural image priors, which limits their scalability and introduces perspective biases that conflict with geospatial constraints. To address this, we introduce GeoCore-9B, a 9-billion-parameter generative foundation model, which is the first of its scale to be trained from scratch exclusively on EO data. Unlike previous EO foundation models, GeoCore-9B is built upon a Flow Matching-based Diffusion Transformer (DiT) and natively conditions generation on text descriptions and continuous geospatial metadata, including ground sample distances, latitudes, and longitudes. To overcome the convergence and spatial disorientation challenges of training at this scale, we propose a Geospatial Semantic Alignment loss. This objective distills structural Earth surface priors (e.g., terrain and urban areas) from a frozen specialist teacher network, constraining the diffusion latent trajectory during training without adding inference overhead. Pre-trained on the global-scale Git-10M dataset, GeoCore-9B demonstrates strong downstream versatility. Beyond standard proxy generative tasks, we show that GeoCore-9B can be effectively adapted for practical EO applications, including highly challenging tasks such as cloud removal and SAR-to-optical cross-modal translation. Extensive evaluations confirm that GeoCore-9B establishes new state-of-the-art performance in both visual fidelity and geographic structural accuracy.

版本更新 · v2arXiv:2607.03012

HyperVAttention: Efficient Sparse Attention with Spatio-Temporal Clustering for Video Diffusion

Dongyeun Lee, Amir Zandieh, Vahab Mirrokni, Junmo Kim, Insu Han

原文摘要

Video Diffusion Transformers (VDiTs) have demonstrated significant capabilities in high-fidelity video generation. However, their ability to produce long-duration videos is fundamentally constrained by the quadratic complexity of the self-attention mechanism. Recent clustering-based sparse attention methods improve the quality-speed trade-off by grouping semantically similar tokens, but their practical efficiency remains limited by two bottlenecks: substantial clustering overhead and low CTA utilization caused by irregular cluster-induced blocks. We propose HyperVAttention (HVA), a training-free sparse attention framework that addresses both bottlenecks jointly. To reduce clustering overhead, we introduce 3D local-window clustering, which exploits the spatio-temporal locality of video tokens to restrict centroid search to fixed local neighborhoods, and implement it with a custom Triton kernel for efficient execution. We further propose a hybrid clustering strategy that performs full clustering only at anchor steps and updates only subset tokens at intermediate steps, leveraging the temporal stability of cluster assignments across denoising steps. To improve CTA utilization, we present hardware-aware cluster merging that minimizes CTA-aligned execution cost through parallel agglomerative merging, improving block density and approximation fidelity by utilizing idle tile capacity. Together, these components reduce clustering overhead, avoid redundant updates, and better align sparse attention with the fixed tile structure of modern GPU kernels. Experiments on video generation show that HVA advances the Pareto frontier of training free sparse attention, achieving higher fidelity than SVG2 at comparable latency and a $2.13\times$ speedup over full attention on HunyuanVideo, or $2.23\times$ with its faster variant HVA-Flash.

cs.CV · cs.AI · cs.LG

版本更新 · v4arXiv:2603.10648

Less is More: Compact-Token Masked Feature Prediction for Skeleton Representation Learning

Jeonghyeok Do, Yun Chen, Geunhyuk Youk, Munchurl Kim

原文摘要

Current skeleton representation learning paradigms face distinct limitations: Contrastive Learning (CL) often overlooks fine-grained motion details, while Masked Auto-Encoders (MAE) rely on coordinate-level reconstruction. This reconstruction inherently demands dense token sequences and heavy decoders, wasting pre-training computation on discarded components and forcing downstream inference to process dense token grids. To resolve these bottlenecks, we propose SLiM (Skeleton Less is More), a compact-token framework that unifies masked feature prediction and contrastive learning via a shared encoder. By shifting the objective from raw coordinate reconstruction to decoder-free, teacher-guided feature prediction, SLiM breaks the reliance on dense tokenization and enables effective learning with a highly compact token grid. Crucially, to prevent trivial shortcut learning arising from strong inter-joint dependencies of human, we introduce Semantic Tube Masking together with Skeleton-Aware Augmentations to enforce deep skeletal-temporal reasoning and anatomical consistency. Extensive experiments across multiple downstream protocols demonstrate that SLiM achieves state-of-the-art performance while structurally reducing inference computation by 7.89$\times$ compared to dense-token MAE baselines.

cs.CV

版本更新 · v2arXiv:2610.25284

Learning Lane-Structured Conditioning Fields for Online HD Mapping

Andrea Bertogalli, Anthony Dell'Eva, Marcello Ceresini, Federico Pirazzoli, Lorenzo Cipelli, Filippo D'Addeo, Nicola Marinello, Massimo Bertozzi

原文摘要

High-definition maps provide critical structural information for downstream autonomous driving tasks. Recent methods increasingly rely on external priors to improve prediction accuracy and robustness. Yet, existing prior representations expose a trade-off: dense rasters are easy to integrate but ignore certain map geometric properties, while sparse vectors preserve geometric structure but require dedicated encoders and are difficult to plug into existing architectures. Moreover, external priors can be inaccurate, costly to acquire, or difficult to maintain, motivating a structural representation that can provide geometric guidance while remaining learnable and broadly applicable. We address this limitation with the Lane Field, a dense BEV-aligned geometric representation that encodes local lane structure at each spatial location and can be either predicted from onboard observations or constructed from external map priors. Building on this representation, we introduce a plug-and-play framework that enhances online HD mapping through two complementary mechanisms: BEV feature modulation, which injects lane-structured cues into latent BEV features, and deformable attention guidance, which steers query sampling toward the underlying lane geometry. We primarily evaluate our method on OpenLane-V2, showing that conditioning BEV features and guiding deformable attention with local lane geometry enables our framework to consistently improve existing methods and achieve state-of-the-art performance on lane segment perception. Project page: https://andberto.github.io/LaneField/

cs.CV

版本更新 · v2arXiv:2610.14629

Executable but Unfaithful: Mock Object, Stub, and Assertion Generation by LLMs and Coding Agents

Hao Ning, Jiahui Dong, Zhen Yang, Guanda Yan, Yichen Sun, Xuanyu Zhu, Zheng Shen, Yifei Pei 等 10 位作者

原文摘要

Mocking is a foundational practice in unit testing, enabling developers to isolate the code under test by replacing real dependencies with simulated alternatives. Automating the generation of mock objects, stubs, and mock assertions has been a long-standing research goal. However, capture-and-replay approaches rely on recording the runtime behavior of actual dependencies. Evolutionary approaches such as StubCoder overcome this requirement but constrain generation to predefined symbol pools. Large language models (LLMs) and coding agents offer an alternative by generating mocking artifacts from source code without requiring recorded dependency executions or fixed symbol pools. Yet, no systematic study has evaluated their capabilities across all three artifact types. To address this gap, we construct MockBench, a benchmark comprising 165 mock artifact generation tasks from real-world Java and Python projects, with validated reference mock objects, stubs, and mock assertions. We evaluate six LLMs and three coding agents in terms of executability, behavior fidelity, and interaction-fault detection capability, complemented by comparisons with existing tools and sensitivity analysis. Experimental results show that LLMs and coding agents indeed have been the state-of-the-art techniques in mock-based testing. More specifically, compared with LLMs, coding agents substantially improve executability, while these gains do not consistently translate into higher behavior fidelity or better interaction-fault detection. Furthermore, we analyze 660 failures from LLMs and coding agents, and identify ten error types across the three artifact categories, revealing behavioral deviations that passing tests can conceal. Our findings highlight the challenges of mock artifact generation and provide practical guidance for both developers and researchers.

cs.SE

版本更新 · v2arXiv:2607.22599

Differencing the Diffusion Trajectory toward Uncertain Components for Time Series Forecasting

Chen Su, Yuanhe Tian, Yan Song

原文摘要

Diffusion models have become a widely used framework for probabilistic time series forecasting, modeling the distribution of future values given an observed history. In time series forecasting, however, the future continues the observed history, creating an asymmetry the standard diffusion process leaves unaddressed, with slowly-varying content largely determined by the observed continuity while higher-frequency dynamics carry most of the residual uncertainty. Existing diffusion-based forecasters decouple this asymmetry through an external rule before generation, leaving the corruption trajectory blind to which parts of the target the history can already anchor. We propose DiffDiff, a diffusion framework that embeds this predictability asymmetry into the diffusion trajectory itself, so that a single end-to-end diffusion process becomes aware of which parts of the target the history can already anchor. DiffDiff makes the forward operator step-dependent so that the noisy intermediate state progressively shifts from the target itself toward its second-order differenced structure, while a conditioning pathway supplies the denoiser with both value-domain and differential history information balanced by a stage-adaptive gate at each diffusion step. The terminal distribution approaches a standard Gaussian, preserving compatibility with existing samplers. On seven benchmarks across four prediction horizons, DiffDiff outperforms six diffusion baselines, while our analyses show that its trajectory reallocates modeling effort according to history predictability and improves distributional fidelity.

cs.AI

版本更新 · v3arXiv:2603.24626

One Method Does Not Fit All: A Large-Scale Benchmark of Imputation Methods for Single-Cell RNA Sequencing Data

Yuichiro Iwashita, Ahtisham Fazeel Abbasi, Koichi Kise, Andreas Dengel, Muhammad Nabeel Asim

原文摘要

Single-cell RNA sequencing (scRNA-seq) enables gene expression profiling at cellular resolution but is inherently affected by sparsity caused by dropout events, where expressed genes are recorded as zeros due to technical limitations. These dropout events distort gene expression distributions and compromise downstream analyses. To address this challenge, numerous traditional and deep learning (DL)-based imputation methods have been developed to detect dropout events and impute their true expression values. However, evaluations of these methods have largely been restricted to specific downstream tasks and limited datasets, and large-scale systematic comparisons under standardized benchmarking conditions are lacking. To fill this gap, we present a comprehensive framework that evaluates 16 imputation methods across 10 experimental protocols, 6 downstream tasks, and 30 benchmark datasets. Through extensive experimentation, we systematically investigate the strengths, limitations, and task-specific performance of these methods under consistent evaluation settings. Our results show that traditional methods perform relatively better across a wide range of downstream tasks, whereas DL-based methods are more variable. In addition, strong performance in numerical gene expression recovery does not necessarily translate into improved biological interpretability in downstream analyses, including cell clustering, differential expression analysis, marker gene analysis, trajectory analysis, and cell type annotation. Furthermore, performance varies substantially across datasets, protocols, and downstream analyses, with no single method consistently outperforming the others. Our findings provide practical guidance for selecting imputation methods tailored to specific analytical objectives and underscore the importance of task-specific evaluation in scRNA-seq data analysis.

q-bio.GN · cs.LG · stat.ML

版本更新 · v2arXiv:2603.27220

Cohesion-Sensitive Power Indices: Representation Results for Banzhaf and Shapley Values

Thomas Pitz, Vinicius Ferraz

原文摘要

In many applications of cooperative game theory -- from corporate governance and cartel formation to parliamentary voting -- not all winning coalitions are feasible. Ideological distances, institutional constraints, or pre-electoral agreements may render certain coalitions implausible. Classical power indices ignore this and weight all winning coalitions equally. We introduce cohesion structures to quantify coalition feasibility and axiomatically characterize two families of cohesion-sensitive power indices, represented as expected marginal contributions under Luce-type distributions. In the Banzhaf branch, coalition weights are a power transformation of cohesion; in the Shapley branch, additional axioms separate size from cohesion, recovering the classical size weights with cohesion acting within each size class. All results have been mechanically verified in Lean 4 with Mathlib. We illustrate the framework on the German Bundestag and the French Assemblée Nationale, where cordon sanitaire and double cordon scenarios produce sharp, interpretable power shifts.

econ.TH · cs.GT

版本更新 · v3arXiv:2608.26831

XION: Interpreting Network Traffic Anomalies via Dependency Violations

Federica Uccello, Simin Nadjm-Tehrani

原文摘要

Current research on security monitoring is increasingly focusing on machine-learning-based approaches, but caveats remain. In addition to huge computational overhead, one concern is the lack of insights into ``why" alerts are raised. Existing interpretability approaches rely on feature attribution methods that ignore dependencies among features or on causal modeling that requires extensive domain knowledge or computational resources. This work proposes XION, a method for modeling relationships among network-flow features based on benign traffic only. During detection, anomalies are identified through violations of expected feature dependencies. Further, XION supports post-alert analysis by identifying which feature relationships break, when they break along the attack timeline, and how dependency violations evolve relatively to other identified violations. XION is evaluated on standard IDS datasets and compared against an Isolation Forest (IF) baseline across multiple attack scenarios, including both volumetric and stealthier attacks. Results show that XION matches or exceeds IF recall in all evaluated scenarios, while requiring up to 7$\times$ less inference time. At the post-alert stage, the dependency violation analysis reveals temporal and structural patterns consistent with known attack behaviors, and is further evaluated against SHAP-based feature attribution of IF alerts. Together, these findings confirm that attacks indeed disrupt feature dependencies learned from benign traffic, and that these disruptions provide additional information for understanding an alert.

cs.CR

版本更新 · v2arXiv:2610.22915

Reward Engine: Reward-Centric Self-Evolution for Open-Ended Tool Agents

Litong Feng, Ji Qi, Yiran Cheng, Shaowu Pan, Xinyu Zhang, Qijun Li

原文摘要

Open-ended tool agents can complete the same objective through different tool paths, and their quality depends jointly on the interaction trace, environment state, and final artifacts, so such tasks rarely admit an executable ground-truth reward. The central challenge is therefore to construct a reward from existing experience that exposes the current policy's failures while remaining stationary during optimization. We introduce Reward Engine, which, like a self-motivated student, first works out what success requires and then optimizes toward it. From the task and initial trajectories, it builds a RewardPack of testcases for required correctness and rubrics for graded quality, each executed as a check that returns a judgment with evidence. The frozen RewardPack then guides the agent to revise and select its outputs, and can also serve as a reward for reinforcement learning. On WorkBuddy Bench Office and Claw-Eval, the constructed reward largely agrees with the benchmark graders, not reversing their ordering on 82.3\% and 80.5\% of candidate pairs. Used to revise executions, it raises three-run consistency ($\mathrm{Pass}^{3}$) over Baseline from 45.0\% to 54.0\% and from 78.4\% to 83.7\%.

cs.AI

版本更新 · v2arXiv:2610.01279

PickMoment: Continuous-Time Single-Image-to-Video via Learning Deblurring and Blur-to-Video

Junseong Shin, Hyeonsu Jo, Daehyun Kim, Tae Hyun Kim

原文摘要

Motion blur arises from the temporal integration of a continuous sharp signal over a finite exposure window, yet existing learning-based methods sidestep this physical model and predict only the sharp signal itself: most single-image deblurring methods recover a single frame at the exposure center, while blur-to-video methods predict a fixed set of frames. We introduce PickMoment, a continuous-time reformulation that directly learns the interval-mean blur over arbitrary sub-intervals of the exposure with a single deterministic model. Drawing an analogy to MeanFlow's average-velocity formulation, we train the model with three supervisions derived from the blur integral: an empirical reconstruction loss from available subframes, an additivity loss that enforces self-consistency across overlapping sub-intervals, and a sharp-frame loss anchored at the zero-interval limit. A single trained model unifies single-image deblurring, blur-to-video generation, and continuous-time pick-a-moment recovery as different queries to the same network, with no separate training for each task. Our PickMoment achieves state-of-the-art performance among generative-based deblurring methods on GoPro and HIDE while competitive against restoration-based methods on RealBlur, and the highest per-frame fidelity on GoPro-7 blur-to-video, all in a single forward pass without iterative sampling.

cs.CV · cs.AI

版本更新 · v2arXiv:2605.03158

HackerSignal: A Multi-Source, CVE-Linked Corpus and Prospective Benchmark for Cyber Threat Intelligence

Benjamin M. Ampel, Sagar Samtani

原文摘要

Cyber threat intelligence (CTI) systems must link exploit code, advisories, and fix commits to the vulnerabilities they concern, and they must do so for vulnerabilities disclosed after training. HackerSignal is a public corpus and benchmark built for this prospective setting. It aggregates 7.45 million exact-deduplicated documents from 64 public sources across eight source layers (hacker forums, exploit archives, exploit question-and-answer sites, vulnerability references, advisories, bug-bounty disclosures, fix commits, and known-exploited-vulnerability records), spanning 1988--2026. All layers are keyed to a shared CVE identifier space with source provenance and release-mode governance. We define three benchmark tasks with temporal splits: (1) CVE linkage retrieval over 340,536 National Vulnerability Database descriptions, (2) exploit type classification into eight classes, and (3) temporal generalization, where training and test CVE sets do not overlap ($C_{\rm train} \cap C_{\rm test} = \emptyset$). Zero-shot dense and hybrid retrievers reach a mean reciprocal rank (MRR) of 0.71 on Task 1, and the top models are statistically tied under paired bootstrap tests. Cybersecurity-pretrained encoders lead exploit type classification (macro-F1 0.90). A two-annotator audit of 240 linkage labels finds 79.5% precision (Cohen's \k{appa} 0.94), with per-source precision from 20% to 100%. The hacker-community layer contributes 6.97 million posts but only about 2,000 CVE-linked records. A masked forum-post grounding pilot over these records reaches MRR 0.08, about nine times lower than the structured-source tasks. Grounding informal hacker discourse to vulnerabilities therefore remains an open problem that the release is designed to support. Code, data, and Croissant metadata are available at https://huggingface.co/datasets/BenAmpel/HackerSignal and https://github.com/BenAmpel/hackersignal.

cs.CR

版本更新 · v2arXiv:2607.00894

Space-Optimal Sensitivity Oracles for Single-Source Mincuts

Koustav Bhanja, Merav Parter, Asaf Petruschka

原文摘要

How much information is needed to describe the effect of any single edge failure on all mincuts? Classical mincut representations, such as Gomory-Hu trees and cactus-like structures, show that mincuts often admit unexpectedly compact static certificates. However, much less is known about whether such compactness survives failures. We study the single-source setting. Given an undirected multigraph $G$ and source $s$, our oracle reports, for any failed edge $e$, all vertices $t$ whose $s,t$-mincut value decreases in $G\setminus{e}$; equivalently, whether $e$ belongs to some $s,t$-mincut. This is the single-source analogue of Picard-Queyranne's fixed-pair [MPS '80] and Baswana-Pandey's all-pairs sensitivity results [SODA '22], and the failure counterpart of Baswana-Gupta-Knollmann's insertion oracle [Algorithmica '22]. No subquadratic-space solution was known to date. We give an optimal $O(n)$-space oracle with $O(n)$ reporting time, matching the insertion bounds. We also give $O(n^{1.5})$-space oracles for insertions and failures, with $O(\log n)$ membership queries or amortized $O(\log^2 n)$ reporting time per affected vertex. Subquadratic-space oracles with such fast query times were unknown, even for insertions. Our main technical contribution connects farthest mincut DAGs, central to the insertion oracle, with Dinitz-Vainshtein's Connectivity Carcass for Steiner mincuts [STOC '94], which generalizes global-mincut cactus representations. These novel, intricate connections reveal the carcass's potential beyond Steiner mincuts.

cs.DS

版本更新 · v4arXiv:2604.03420

Zero-Shot Quantization via Weight-Space Arithmetic

Daniele Solombrino, Antonio Andrea Gargiulo, Alessandro Zirilli, Luca Zhou, Adrian Robert Minut, Emanuele Rodolà

原文摘要

We show that robustness to post-training quantization (PTQ) is a transferable direction in weight space. We call this direction the quantization vector: extracted from a donor task by simple weight-space arithmetic, it can be used to patch a receiver model and improve PTQ Top-1 accuracy by up to 60% in a 3-bit setting, without receiver-side quantization-aware training (QAT). When using no receiver validation data, the method provides a zero-shot, low-cost alternative to QAT for extremely low-bit deployment. Across multiple vision and language models and more than 30 tasks, donor quantization vectors often yield substantial gains even when donor and receiver tasks differ markedly. We further prove rigorously that quantization vectors are well-defined and do not suffer from reparameterization symmetries, and provide a local geometric account of their effect. Together, these results suggest that quantization robustness can be partially isolated, reused, and transferred through simple weight-space algebra.

cs.CV · cs.AI · cs.LG

版本更新 · v2arXiv:2606.17985

Gaussian Light Field Splatting: A Physical Prior-Driven Vision Transformer for Low-Light Image Enhancement

Yuhan Chen, Wenxuan Yu, Guofa Li, Fuchen Li, Kunyang Huang, Yicui Shi, Ying Fang, Wenbo Chu 等 9 位作者

原文摘要

Existing low-light image enhancement methods often suffer from local exposure imbalance and color distortion under complex, non-uniform illumination, while mainstream Vision Transformers lack explicit modeling of physics-based priors for illumination degradation. To address these limitations, we propose GLFS, a Gaussian Light Field Splatting-based Vision Transformer that incorporates continuous light-field priors from 2D Gaussian Splatting (2DGS) as an explicit physical bias into the Transformer self-attention mechanism to guide spatial gain estimation under complex non-uniform illumination. Specifically, GLFS represents scene illumination as a superposition of anisotropic Gaussian basis functions and introduces a physics-informed bias into the self-attention mechanism to adaptively infer a spatial gain field, enabling accurate and spatially uniform restoration under complex illumination conditions. To mitigate color shifts and structural degradation during enhancement, we further introduce a color vector angular loss and a luminance edge loss, with the former enforcing hue consistency and the latter improving the structural fidelity of local details. Extensive ablation studies and quantitative evaluations demonstrate that GLFS achieves competitive overall performance in illumination correction, structure preservation, and perceptual quality, validating the effectiveness of the continuous Gaussian light field prior for low-light image enhancement.

cs.CV

版本更新 · v2arXiv:2610.01291

ODDR: One-Step Deshadow Diffusion via Reward Guidance

Junseong Shin, Kijun Kim, Minseong Kim, Dongjin Kim, Tae Hyun Kim

原文摘要

Recent advances in deep learning for shadow removal have significantly enhanced image quality and realism. However, most approaches rely on real-world paired datasets, which are costly to collect and often limited in scene diversity, leading to limited generalization. To address these limitations, we propose One-step Deshadow Diffusion via Reward guidance (ODDR), a new framework that achieves efficient and high-fidelity shadow removal without relying on real-world paired supervision. Our method begins with One-step Deshadow Diffusion (ODD), a baseline model trained on synthetic shadow data for efficient one-step shadow-free reconstruction. We further adapt ODD into ODDR using ShadowReward. In contrast to traditional, annotation-heavy approaches, ShadowReward is the first reward model for shadow removal trained entirely without human annotation. It learns to mimic human perceptual judgments by ranking synthetically generated images with controlled degradations, such as texture distortion and boundary artifacts. This reward-guided fine-tuning enables ODDR to close the synthetic-to-real domain gap. Extensive experiments show that ODD achieves strong performance without relying on real-world paired supervision, and ODDR further improves the results, narrowing the gap to fully supervised methods trained on real-world paired data while maintaining higher computational efficiency as a single-step model.

cs.CV

版本更新 · v2arXiv:2512.14056

FacEDiT: Talking Head Video Editing via Facial Motion Infilling

Kim Sung-Bin, Joohyun Chang, David Harwath, Tae-Hyun Oh

原文摘要

Editing a local segment of a talking head video without reshooting the scene remains challenging and underexplored. Local edits, such as insertion, deletion, or substitution, require synthesizing a replacement segment while leaving the rest of the video unchanged. This involves variable-duration local rewriting while preserving identity, unedited regions, and boundary continuity, which standard speech-driven generation and lip synchronization methods do not directly address. Moreover, direct supervision is impractical, as it would require paired videos of the same person and scene that differ only in a local spoken segment. We instead formulate talking head video editing as facial motion infilling, a self-supervised pretext task that recovers masked facial motion from speech and surrounding motion context using ordinary video--speech pairs. Our key insight is that local video editing can be simulated during training by masking a motion span and reconstructing it from speech and visible motion context. Building on this formulation, we introduce FacEDiT, a mask-controlled talking head model with a local temporal attention bias and temporal smoothness regularization to improve lip--speech alignment and transition continuity. We also introduce FacEDiTBench, the first benchmark for talking head video editing, covering diverse edit types and lengths with dedicated evaluation metrics. Extensive experiments show that FacEDiT produces accurate, speech-aligned edits with strong identity preservation and seamless boundary transitions. Beyond editing, the same facial motion infilling model extends to portrait animation and lip synchronization by simply changing the mask pattern, establishing FacEDiT as a unified framework for talking head video editing and generation.

cs.CV · cs.AI

版本更新 · v2arXiv:2610.02801

VIGOR: Zero-Shot Visual Generalization via Latent-Space Consistency in Model-Based Reinforcement Learning

Mingyu Park, Samyeul Noh, Hyun Myung, Donghwan Lee

原文摘要

Model-based reinforcement learning (MBRL) achieves strong sample efficiency by planning in learned latent dynamics but degrades substantially under unseen visual distractions such as background, lighting, or camera changes. Unlike model-free RL, where encoder perturbations affect only single-step predictions, MBRL suffers from a two-level vulnerability: distractions push encoder outputs out of distribution, and these errors compound through recursive latent rollouts. We propose visual generalization via latent-space consistency in model-based RL (VIGOR), which enables zero-shot generalization to unseen distractions while retaining its backbone's sample efficiency. VIGOR comprises three interdependent components: (i) asymmetric weak-to-strong augmentation, which pairs weak-only and weak-to-strong latent views within a batch; (ii) dynamics-level consistency, which enforces augmentation-invariant transition predictions via direct latent regression; and (iii) encoder-level stabilization, which prevents encoder drift under the cross-augmentation supervision of (ii). On the DeepMind Control Suite (DMC) and Robosuite, VIGOR instantiated on TD-MPC2 outperforms state-of-the-art model-free and model-based baselines in zero-shot generalization, surpassing the second-best baseline by 3.4% on DMC and 43.6% on Robosuite, with sample efficiency comparable to its backbone and fewer environment interactions than the other baselines. Ablations show that VIGOR is augmentation-agnostic: augmentations from distinct perturbation families preserve strong generalization, confirming that latent-space consistency, not the augmentation choice, drives robustness. A second instantiation on DreamerV3, a structurally distinct recurrent state-space model, also improves zero-shot generalization on both benchmarks, indicating backbone-agnostic benefits. Code and videos: https://vigor-gen.github.io/.

cs.AI · cs.LG · cs.RO