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

+1112新增论文
+875其中 AI 相关
+133已有论文修订
+94新增 AI 导读

arXiv 最近完成扫描:2026-11-19 20:16 · 全站最近入库:2026-11-19 19:09各信源进度 →

统计口径与数据新鲜度

统计 2026-11-18 20:23 — 2026-11-19 20:23(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:2026-11-19 20:16最近发现新增版本:2026-11-19 19:09库内论文最新更新时间:2026-11-19 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。

27,982 已收录论文23,249 AI 相关8,559 有修订112 HF 社区发现

浏览论文库

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

找到 20,663 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-19 20:10(北京时间)
版本更新 · v3arXiv:2609.11588

Quasilinear multiplication in the Cayley-Dickson tower

Harrison Lemley

原文摘要

Direct evaluation of the defining product of the Cayley-Dickson algebra $A_n$ of rank $N:=2^n$ over a unital commutative ring $R$ requires a quadratic number of arithmetic operations. This paper gives a uniform bilinear algorithm for multiplication in this algebra when $2 \in R^\times$, costing $O(N \log N)$ arithmetic operations in $R$ and $O(N)$ auxiliary storage when the doubling parameters of $A_n$ are units whose square roots lie in a fixed commutative coefficient algebra $S$ finitely generated as an $R$-module and admitting an $R$-linear map $π\colon S \to R$ with $π(1) = 1$. In particular, the algorithm applies to every Cayley-Dickson tower over $\mathbb{R}$ and finite fields $\mathbb{F}_q$ of odd characteristic, and the tensor rank of multiplication in $A_n$ over these fields is at most $(36n-69)2^{n-3}+10$ when $n\ge4$. To the author's knowledge, the algorithm presented in this paper is the first uniform subquadratic algorithm for multiplication in $A_n$. An implementation for the standard tower over $\mathbb{R}$, $\mathbb{C}$, and $\mathbb{F}_q$ for $q = p^k$ with $p \le 2^{31}-1$ an odd prime and $k \le 12$ is available in the MIT-licensed C11 library fastCD, with a NumPy-backed Python interface pyfastcd, available on PyPI. In single-core benchmarks against direct multiplication and the uniform Cariow-Cariowa method over $\mathbb{R}$, the quasilinear implementation had the lowest mean time of the three at every tested dimension $N\ge32$ for both single and batched products and was over $17$ times faster than direct multiplication.

math.RA · cs.SC · math.NA

版本更新 · v2arXiv:2611.13370

AHEADance: Anticipatory Humanoid Dance from Streaming Audio

Tianhu Peng, Yukun Wang, Lingfan Bao, Yiyang Jia, Chengxu Zhou

原文摘要

A humanoid that dances to streamed music must commit each motion chunk before it has heard the music that chunk accompanies, and must continue from motion it has already committed. We present AHEADance, a robot-native streaming dance generator executed on a humanoid robot by a whole-body tracker. ListenAhead conditions each chunk on heard music and a frozen music model's forecasts of the coming music, aligned to their target times and weighted by reliability. The forecasts carry the timing of upcoming beats almost as well as the actual future; with them, the generator is better synchronized with the music and concentrates more motion energy at its song's beat frequency. Commit Forcing trains continuation from the model's own committed chunks, updating both the latent history and the decoded boundary of the next chunk; it reduces low activity from 83.5% to 10.3%, the level of real dance. Three music evaluators, each validated on real G1 dance, test synchronization, song and style. On a physical Unitree G1, it dances while the music plays and comes to rest when it stops, without a fall. Project page: https://ihcr.github.io/AHEADance

cs.RO · eess.SY

版本更新 · v4arXiv:2510.06844

Oops!... I did it again. Analysing and Handling Conclusion (In-)Stability in Socio-Technical Software Engineering

Nicole Hoess, Carlos Paradis, Rick Kazman, Wolfgang Mauerer

原文摘要

Context: Mining software repositories is a popular means to gain insights into a software project's evolution, monitor project health, support decisions and derive best practices. Tools supporting the mining process are commonly applied by researchers and practitioners, but their limitations and agreement are often not well understood. Objective: This study investigates some threats to validity in complex tool pipelines for evolutionary socio-technical software analyses. We evaluate the tools' agreement in terms of data, study outcomes and conclusions for the same research questions to derive actionable advice for researchers and practitioners. Method: We conduct a lightweight literature review to select \emph{three} studies on collaboration and coordination, software maintenance and software quality from high-ranked venues, which we formally replicate with \emph{four} independent, systematically selected mining tools to quantitatively and qualitatively compare the extracted data, analysis results and conclusions. Results: We find and summarise numerous technical details in tool design and implementation which accumulate along the complex mining pipelines and can cause substantial differences in the extracted baseline data, its derivatives, subsequent results of statistical analyses and, under specific circumstances, conclusions. Conclusions: Users should evaluate mining tools and their limitations carefully to scope the validity of their conclusions, for instance with the checklist we provide. Researchers and tool authors can further reduce uncertainty through reproduction packages and comparative studies in other MSR fields following our methodology.

cs.SE

版本更新 · v2arXiv:2610.12737

Kernelized Parameter Sensitivity Atlases for Amortized Conditional Interpretability

Davide Wiest, Ahanaf Hasan Ariq, Abdul Monaf Chowdhury

原文摘要

Gradient-based attribution provides input-specific information about model parameters, but computing it via backpropagation for many samples is expensive, while static importance scores discard input dependence. We derive KPSA (Kernelized Parameter Sensitivity Atlas) from a block decomposition of the empirical neural tangent kernel. It measures normalized parameter-group sensitivity, embeds those sources in an explicit query space, and uses a lightweight kernel to estimate sensitivities for unseen inputs. We derive bounds that separate finite-source from irreducible representation error and give margin-based top-$k$ recovery. This paper studies two interpretability applications. Firstly, KPSA identifies parameter groups with high susceptibility to perturbations across vision, forecasting, and protein-modeling setups. Across a fivefold range of parameter budgets ($0.02$-$0.1\%$), its selected groups attain $39$-$85\%$ of the perturbation susceptibility of a first-order backpropagation oracle. In vision and forecasting, these parameter groups form sparse sensitivity circuits that extend across depth and vary systematically with the input. Secondly, we use KPSA on an activation-attribution problem for MLP features in transformers. It retains $81\%$ of the gap between a backpropagation oracle and a constant-sensitivity baseline on vision while increasing batched query-time throughput from 551 to 2,041 images/s and reducing peak GPU memory from 4.55 to 0.44GB. On the protein-modeling task, it recovers $68\%$ of this gap. An ablation of KPSA's kernel aggregation shows that nearest-neighbor retrieval in KPSA space can remain competitive. The empirics show that compact kernel atlases can store useful input-dependent parameter sensitivity while amortizing repeated gradient attribution, a step toward scalable conditional interpretability. Code can be found at https://github.com/Axym-Labs/kpsa .

cs.LG · cs.CV · stat.ML

版本更新 · v5arXiv:2606.00419

Parameter-Free and Group Conditional Online Conformal Prediction

Beepul Bharti, Ambar Pal, Jacopo Teneggi, Jeremias Sulam

原文摘要

Quantifying the uncertainty in predictions made by machine learning models is challenging in online settings where the characteristics of the data and labels change over time. Online conformal prediction (OCP) methods tackle this problem by producing intervals for each observation and guaranteeing that a proportion of them contain their labels, a property known as marginal coverage. OCP methods, however, may systematically miscover observations belonging to groups defined by attributes like age or race, raising fairness concerns. Group-conditional OCP (G-OCP) methods address this by producing intervals that provide coverage for multiple groups. Existing approaches rely on learning rates, which control how they react to changes in the data stream, but can be hard to tune in online settings. Leveraging advancements in learning-rate-free methods, we develop a learning-rate-free G-OCP algorithm with strong group-coverage guarantees that hold under weaker assumptions than those required in prior work. Empirical results show that our method provides group coverage, produces interval lengths competitive with those of standard methods, and reacts more effectively to changes in the data than fixed-learning-rate methods.

stat.ML · cs.LG

新论文 · v1arXiv:2611.21096

Inference Equivalence and Quotient Coding

Zibo Meng, Andrew C. M. Austin, Alin Achim

原文摘要

Incomplete observations can admit multiple physical interpretations without making every task conclusion ambiguous. Recovering all compatible scenes is often unnecessary, whereas indiscriminate compression may change which claims are justified. We ask which distinctions among observation-compatible interpretation sets must be retained to preserve every valid conclusion of a specified task. We first define inference validity as truth across all physically compatible states and characterize it, under a closedness condition, by distances to physically realizable falsifying measurements. For a finite semantic claim family ordered by specificity, the Exact Inference Frontier forms the unique minimal antichain generating all valid claims. We then define inference equivalence between compatibility sets and construct a canonical quotient code from their frontiers. Among deterministic codes admitting exact inference decoding, it induces the coarsest partition, minimizing the number of distinguishable code values rather than storage or computation. Finally, we characterize inference errors from learned compatibility sets, derive conditions for exact inference preservation despite set errors, and evaluate an asymmetric learning surrogate under distribution shifts. On a registered finite bank, 145 compatibility sets yield 31 quotient codes with unchanged valid-claim sets. Across geometry, mechanism, and joint shifts, learned estimates exhibit different accuracy-specificity trade-offs. On 1,962 official HRSID test images, 1,913 distinct candidate sets induce only 36 seven-claim inference signatures, while a visual 3-NN baseline achieves 87.31% precision on issued claims.

cs.AI · cs.LG

新论文 · v1arXiv:2611.21095

SWE-WhatIf: Simulating Real Users of Coding Agents from Recorded Session

Tae Soo Kim, Tongshuang Wu, Juho Kim

原文摘要

Recorded sessions between users and coding agents offer a way to evaluate how agents complete real-world tasks and interact with users. However, these sessions cannot simply be replayed with new agents, because recorded user replies may no longer apply when an agent takes a different trajectory. We introduce SWE-WhatIf, which replays recorded sessions with new agents by modeling the user's intents, the criteria for satisfying them, and the conditions under which they arise. Using this model, a user simulator generates replies that adapt to unseen agent turns while remaining grounded in the recording. Across 89 real-world sessions, SWE-WhatIf generated the replies most similar to those of real users and matched the strongest baseline when sessions were rewritten. To test adaptation, we replay sessions with agents instructed to avoid or repeat actions that the recorded user pushed back on. SWE-WhatIf produced the highest win rate for avoiding these actions (66.3%), significantly above chance. We further demonstrate its utility by ranking five LLMs as coding agents, where the highest-ranked model depended on how users specified and developed their intents, and a study with 16 users, where replaying their own sessions changed 14 participants' rankings and informed their future agent use.

cs.CL · cs.AI · cs.HC

版本更新 · v2arXiv:2610.23356

STC-MGNet: Spatial-Temporal-Channel Attention-Enhanced Multimodal Graph Network for Pedestrian Intention Prediction

Yuanzhe Li, Meng Wang

原文摘要

Accurate prediction of pedestrian crossing intentions plays a critical role in enhancing the safety of pedestrian-vehicle interactions in autonomous driving. However, effectively capturing subtle spatiotemporal patterns in pedestrian pose sequences and integrating complementary visual and ego-vehicle motion information remain challenging. This paper proposes a spatial-temporal-channel attention-enhanced multimodal graph network (STC-MGNet) for pedestrian crossing intention prediction. STC-MGNet introduces a spatial-temporal-channel (STC) attention-enhanced pose graph encoder to learn discriminative representations from pedestrian pose sequences. Three parallel attention branches selectively emphasize informative skeletal joints, observation frames, and feature channels, with their outputs integrated through learnable fusion weights. To leverage complementary visual and ego-vehicle motion information, a multi-stage cross-modal channel modulation mechanism is introduced to adaptively refine pose representations through modality-specific channel gating at successive graph encoding stages. Experimental results demonstrate that STC-MGNet achieves accuracy rates of 70%, 87%, and 88% on the JAADbeh, JAADall, and PIE datasets, respectively, outperforming several representative baseline methods. Ablation experiments further highlight the complementary contributions of visual and ego-vehicle motion information to prediction performance.

cs.CV

新论文 · v1arXiv:2611.21094

Conflict-Aware Multi-Teacher On-Policy Distillation in Aligned Action Spaces for Cross-Game Generalization

Jing Yi, Cehao Yang, Hualei Zhang, Haitao Li

原文摘要

Zero-shot cross-game generalization aims to consolidate accessible source-game experts into a single vision-language model (VLM) that controls unseen games from raw pixels without target-game parameter updates. While multi-teacher on-policy distillation (MOPD) mitigates offline exposure bias and single-game reinforcement learning (RL) overfitting, scaling it to autoregressive VLMs faces two bottlenecks. First, token-to-action misalignment from subword length bias distorts discrete action probabilities and uncertainty margins. Second, multi-teacher policy conflict arises as source teachers under incompatible dynamics issue contradictory closed-loop corrections, causing negative transfer. We propose CAMOPD (Conflict-Aware Multi-Teacher On-Policy Distillation), a post-training framework that resolves multi-teacher policy conflict within an aligned semantic action space. CAMOPD first uses complete-action alignment to normalize command completion likelihoods over each game's legal action set, forming a calibrated categorical policy that provides reliable uncertainty margins for selective on-policy retention and visual specialist probes that can be evaluated across source games. To resolve policy conflict without target-game data, conflict-aware multi-teacher consolidation cross-evaluates these specialists across source games to capture both visual identifiability and inter-teacher compatibility, upweighting compatible teachers while suppressing conflicting corrections. On GameWorld's temporal and directional transfer suites with Qwen3.5-4B, CAMOPD achieves 26/30 (86.7%) and 24/30 (80.0%) Task-1 zero-shot success on unseen GeoDash and Doodle Jump, outperforming prompting, weight-merging, and multi-task baselines.

cs.AI

新论文 · v1arXiv:2611.21088

AudioTrace: File-Level Membership Detection for Audio Large Language Models

Guangzhi Sun, Xiao Zhan, Shuai Wang, Jose Such, Phil Woodland

原文摘要

Significant privacy and copyright concerns are raised by the use of publicly available audio recordings to train audio large language models (LLMs), yet effective auditing methods remain limited. Existing membership inference attacks typically assume knowledge of complete training examples, including their associated textual supervision. In practice, however, audio recordings may be paired with privately generated transcripts, captions, or question-answer pairs that are unavailable to owners of the recordings. This motivates audio-file-level membership inference, which determines whether a recording was used for training without assuming knowledge of its training tasks and annotations. This paper provides the first systematic investigation of this challenge and introduces AudioTraceBench, a controlled benchmark comprising 1,285 audio recordings with explicit fine-tuning membership labels. Evaluations across three audio LLMs show that existing text- and video-based membership inference methods adapted to audio perform close to random guessing. To address this challenge, AudioTrace is proposed, where audio-conditioned proxy questions are generated using the target model, and membership evidence is extracted from its responses and aggregated at the recording level. Consistent improvements over existing baselines are observed, with AUROC reaching 0.60 and increasing further under increased fine-tuning exposure or adaptation capacity. Further experiments were conducted to examine the factors influencing membership signals for audio-file-level detection. These findings establish audio-file-level membership inference as a distinct and practically important auditing challenge, with AudioTrace providing an effective initial approach to this challenge.

cs.CR · cs.SD

版本更新 · v4arXiv:2504.02293

Breaking the Silence: A Dataset and Benchmark for Bangla Text-to-Gloss Translation

Sharif Mohammad Abdullah, Abhijit Paul, Shubhashis Roy Dipta, Zarif Masud, Shebuti Rayana, Ahmedul Kabir

原文摘要

Gloss is a written approximation that bridges Sign Language (SL) and its corresponding spoken language. Despite a deaf and hard-of-hearing population of at least 3 million in Bangladesh, Bangla Sign Language (BdSL) remains largely understudied, with very limited prior work on Bangla text-to-gloss translation and no publicly available parallel dataset at the time of this work. To address this gap, we construct the first publicly available Bangla text-to-gloss dataset, consisting of 1,000 manually annotated and 4,000 synthetically generated Bangla sentence-gloss pairs, along with 159 expert human-annotated pairs used as a test set. Our experimental framework performs a comparative analysis between several fine-tuned open-source models and a leading closed-source LLM to evaluate their performance in low-resource BdSL translation. GPT-5.4 achieves the best automatic metric scores, while open-source models perform competitively despite being much smaller. This work introduces the first publicly available expert-annotated Bangla text-to-gloss dataset, benchmark and trained model for Bangla text-to-gloss translation. It also demonstrates the effectiveness of systematically generated synthetic data for addressing challenges in low-resource sign language translation.

cs.CL · cs.AI

版本更新 · v2arXiv:2605.10668

A Spectral Framework for Closed-Form Relative Density Estimation

Francis Bach

原文摘要

We propose a closed-form spectral framework for relative log-density estimation in linearly parameterized probabilistic models, including unnormalized and conditional models. This is achieved by representing the Kullback-Leibler (KL) divergence as an integral of weighted chi-squared divergences, converting KL estimation into a family of least-squares problems. We derive an explicit spectral formula based only on first- and second-order feature moments, yielding closed-form estimators of both divergences and log-density potentials for fixed features. The framework extends to a broad class of $f$-divergences and can be combined with kernelization or feature learning with neural networks (with then only a closed form for the last layer). We prove convergence guarantees for the resulting estimators and empirically compare them on synthetic data with optimization-based variational formulations, including logistic and softmax regression for normalized conditional models.

cs.LG · math.OC · math.ST

版本更新 · v2arXiv:2604.09220

TinyNeRV: Compact Neural Video Representations via Capacity Scaling, Distillation, and Low-Precision Inference

Muhammad Hannan Akhtar, Ihab Amer, Tamer Shanableh

原文摘要

Neural Representations for Videos (NeRV) encode video sequences within neural decoder parameters, enabling direct frame reconstruction from temporal indices. Deploying NeRV in resource-constrained settings remains challenging because reconstruction quality, model capacity, decoding cost, and numerical precision are tightly coupled. This paper presents TinyNeRV, a deployment-oriented framework for compact neural video representations based on capacity scaling, teacher-guided quality recovery, and low-precision inference. We introduce NeRV-T (0.80M parameters, 22.62 GFLOPs) and NeRV-T+ (1.68M parameters, 89.80 GFLOPs) as minimum-compute and balanced low-capacity configurations. To recover detail without increasing inference cost, frequency-focal knowledge distillation emphasizes high-frequency teacher-student discrepancies, improving PSNR on Big Buck Bunny by 0.84 dB for NeRV-T and 0.88 dB for NeRV-T+. We also evaluate weights-only INT8, INT6, and INT4 quantization using post-training quantization (PTQ), quantization-aware training (QAT), and distillation-assisted QAT. Across the seven-sequence UVG benchmark, frequency-focal distillation improves baseline PSNR by an average of 0.70 dB for NeRV-T and 0.78 dB for NeRV-T+. Across UVG and Big Buck Bunny, INT8 is nearly lossless; INT6 causes modest PTQ degradation largely recovered by QAT; and INT4 benefits from retraining-aware adaptation and teacher supervision. The results establish practical capacity-supervision-precision trade-offs for real-time, resource-constrained neural video reconstruction.

cs.CV

新论文 · v1arXiv:2611.21085

State-Message Interaction Adapter for Cross-Domain Few-Shot Segmentation

Hang Xing, Yan Xia, Yongjiang Liu, Siqi Zhang, Xueming Ding

原文摘要

Cross-domain few-shot segmentation adapts natural-image pretrained encoders to medical or industrial domains from a few pixel-annotated support masks. Within frozen DINOv3, intermediate per-head attention messages already carry target-background distinctions that baseline final-layer similarities and predictions do not consistently reflect. We propose the State-Message Interaction Adapter (SMIA) to learn how per-head messages should update each receiving token representation while keeping pretrained attention and FFN parameters frozen. Inserted right after per-head aggregation, SMIA conditions the transformation of each head's aggregated output (message) on each patch's pre-attention feature (state), allowing similar contextual messages to produce distinct or opposing updates depending on the receiving position. Because scarce support masks cover only limited state-message combinations, a shared low-rank bilinear map confines the learned interactions to low-dimensional state, message, and output subspaces. Added before each adapted block's frozen FFN, these updates yield adapted features that are refined by frozen FFNs and deeper blocks for linear readout. With only 0.51M trainable parameters and single-pass, support-free inference, SMIA achieves six-dataset mean mIoUs of 77.40% at 5 shots and 80.99% at 10 shots across 20 medical and industrial tasks, surpassing the strongest prior methods by +3.61 and +3.93 percentage points. In controlled 10-shot comparisons across three benchmarks on the same DINOv3 backbone, SMIA also outperforms parameter-matched AdaptFormer, LoRA, and ParaX with a better precision-recall balance.

cs.CV

新论文 · v1arXiv:2611.21084

CompCert4Lean: Porting a Verified C Compiler to Lean in Six Weeks with Parallel Coding Agents

Cong Wang, Zhiyang Chen, Fan Long, Martin Rinard

原文摘要

We present CompCert4Lean, a large-scale port of CompCert 3.17's verified compiler core from Rocq to Lean, completed in approximately six weeks by one expert directing parallel LLM agents. Our migration workflow partitions the development along its dependencies, audits translated definitions and theorem statements before large-scale proof reconstruction, and then reconstructs proofs in parallel against stable interfaces. The resulting compiler covers typed CompCert C through x86-64 assembly, discharges all 4,908 recorded proof obligations without project-specific axioms, and produces executable Linux binaries. The workflow enabled the expert to concentrate on organizing the work rather than supplying technical solutions. Of 1,071 human interactions, 88.8% concerned workflow and coordination, while only three supplied technical direction. Optimization campaigns further demonstrate the benefit of the Lean port for agent-driven verified development. Under matched budgets of 24 active hours, GPT-6-astra produced a Lean candidate with 11.7% lower generated-program runtime across 29 small-tests programs, compared with 2.3% lower for its Rocq candidate, both relative to original CompCert 3.17 and while preserving the end-to-end correctness theorem. These results show that parallel LLM agents can support large-scale migration with limited human technical direction and subsequently improve the migrated compiler while retaining its formal guarantees.

cs.PL · cs.LO

新论文 · v1arXiv:2611.21083

Think2Listen: Evidence-Centric Learning For Agentic Audio Understanding With LALMS

Jin Wang, Liang Lin, Kaiwen Luo, Weiliu Wang, Yuxuan Li, Zengxin Han, Gan Zhang, Jingtong Wu 等 10 位作者

原文摘要

Recent advances in large audio-language models (LALMs) have broadened audio understanding across speech, sound, and music, with tool-based agents further supporting these tasks by invoking predefined audio tools. However, a persistent gap remains between perceiving acoustic content and using it as evidence for reasoning: recognizing what is audible does not ensure that a model can identify which cues resolve an uncertain decision. To bridge this gap, we propose Think2Listen, an evidence-centric learning framework for agentic audio understanding. Guided by evidence before commitment, Think2Listen couples iterative reasoning with a targeted LALM interaction to actively seek acoustic cues that distinguish competing hypotheses and revise candidate judgments based on the observations returned. It learns what to ask, how to use evidence, and when to stop, using observation-induced changes in candidate-state to inform training credit. Extensive evaluations on two benchmarks spanning seven audio categories demonstrate that Think2Listen achieves (I) enhanced multi-step audio reasoning, with gains in overall accuracy of 5.08%--17.67% over competing audio agents; and (II) improved audio understanding and reasoning across speech, sound, and music, boosting overall accuracy by 2.29%--13.10% over the corresponding baselines. The source code is available at https://github.com/JinWang79/Think2Listen.

cs.SD · cs.AI

版本更新 · v3arXiv:2605.14598

DSSP: Diffusion State Space Policy with Full-History Encoding

Zhiyuan Guan, Jianshu Hu, Han Fang, Yunpeng Jiang, Yize Huang, Shujia Li, Xiao Li, Yutong Ban

原文摘要

Diffusion-based imitation learning has shown strong promise for robot manipulation. However, most existing policies condition only on the current observation or a short window of recent observations, limiting their ability to resolve history-dependent ambiguities in long-horizon tasks. To address this, we introduce \ours, a history-conditioned Diffusion State Space Policy that enables efficient, full-history conditioning for robot manipulation. Leveraging the continuous sequence modeling properties of State Space Models (SSMs), our history encoder effectively compresses the entire observation stream into a compact context representation. To ensure this context preserves critical information regarding future state evolution, the encoder is optimized with a dynamics-aware auxiliary training objective. This high-level context representation is then seamlessly fused with recent state observations to form a hierarchical conditioning mechanism for action generation. Furthermore, to maintain architectural consistency and minimize GPU memory overhead, we also instantiate the diffusion backbone itself using an SSM. Extensive experiments across simulation benchmarks and real-world manipulation tasks show that \ours achieves state-of-the-art performance with a significantly smaller model size, demonstrating superior efficiency of the hierarchical conditioning in capturing crucial information as the history length increases. Project page: https://yuaniw.github.io/DSSP-Page/

cs.RO

版本更新 · v2arXiv:2611.17505

Adversarial Tracking: Attacks on Image Sensors for Object Tracking in Autonomous Vehicles

Han Wu, Alex Chan, Ekin Can Erkus, Tousif Rahman, Rishad Shafik

原文摘要

Object Tracking systems estimate the motion of surrounding vehicles by linking detected objects across video frames. In particular, Tracking-By-Detection (TBD) systems combine object detectors with motion prediction and data association to maintain object identities over time. However, neural network-based detectors are vulnerable to adversarial attacks that perturb input images, and their impact on downstream trajectory estimation is not fully characterized by detection accuracy alone. Therefore, this paper evaluates image-specific and image-agnostic attacks on YOLOv3 and YOLOv4 detectors paired with SORT, Deep SORT, Strong SORT, and OC-SORT tracking systems. The evaluation uses a CARLA autonomous driving simulator and measures tracking quality through Multiple Object Tracking Accuracy (MOTA), Higher Order Tracking Accuracy (HOTA) and Identification F1 (IDF1). Under image-specific attacks, HOTA for YOLOv3-based trackers declines from 53.45-56.19% to 14.85-17.03%, while MOTA reaches -772.60% in the worst case. YOLOv4-based trackers exhibit greater robustness, while HOTA still declines to 25.06-26.86%. In addition, image-agnostic attacks can reduce the tracking accuracy using a single Universal Adversarial Perturbation (UAP), without generating a separate perturbation for individual frames.

cs.CV · cs.AI

版本更新 · v2arXiv:2511.16349

LoC-IN: Real-Time Localization of Cameras in Neural Rendered Prior LiDAR Maps

Joni Vanherck, Steven Moonen, Brent Zoomers, Kobe Werner, Jeroen Put, Lode Jorissen, Nick Michiels

原文摘要

Accurate camera localization is crucial for robotics and Extended Reality (XR), enabling reliable navigation and alignment of virtual and real content. Existing visual methods often suffer from drift, scale ambiguity, and depend on fiducials or loop closure. This work introduces a real-time method for localizing a camera within a pre-captured, highly accurate colored LiDAR point cloud. By rendering synthetic views from this cloud, 2D-3D correspondences are established between live frames and the point cloud. A neural rendering technique narrows the domain gap between synthetic and real images, reducing occlusion and background artifacts to improve feature matching. The result is drift-free camera tracking with correct metric scale in the global LiDAR coordinate system. Two real-time variants are presented: Online Render and Match (R&M), and Prebuild and Localize (P&L). We demonstrate improved results on the ScanNet++ dataset and outperform existing SLAM pipelines.

cs.CV · cs.GR

新论文 · v1arXiv:2611.21079

SemiNovo: Learning Beyond Search-Identified Spectra for De Novo Peptide Sequencing

Pan Liu, Dongxin Lyu, Yusen Tan, Yifan Li, Hai-tao Yu, Yunhua Zhong, Zikun Wang, Tianze Ling 等 9 位作者

原文摘要

Sensitive peptide identification is fundamental to protein characterization and novel protein discovery, for which de novo peptide sequencing (DNPS) from tandem mass spectra provides an important database-independent approach. Although deep learning has substantially advanced de novo sequencing, existing models are predominantly trained on high-confidence peptide-spectrum matches obtained through database search. This search-conditioned supervision excludes a large fraction of unidentified spectra, creating a distribution mismatch between the selectively labeled training data and the complete spectrum stream encountered at inference time, thereby limiting peptide identification performance. To address this challenge, we propose SemiNovo, a semi-supervised framework that learns jointly from identified and unlabeled spectra. SemiNovo first constructs DarkSpec, a large-scale quality-controlled corpus of unlabeled spectra collected from diverse public proteomics projects, thereby extending training beyond search-identified spectra and mitigating the resulting shift in DNPS supervision. It then combines stepwise mass-constrained autoregressive decoding with cumulative path confidence to derive reliable supervision from unlabeled spectra. Moreover, SemiNovo introduces head-wise chemical-evidence gating to improve spectral evidence aggregation under noisy and incomplete fragmentation. Experiments show that SemiNovo performs favorably against state-of-the-art methods on multiple benchmarks. Code and dataset are publicly available.

cs.LG · cs.AI

版本更新 · v2arXiv:2607.18765

High-Level Synthesis of Efficient Pipelines with Visibility Control

Jungin Rhee, Minseong Jang, Jaewoo Kim, Jeehoon Kang

原文摘要

High-level synthesis (HLS) raises the abstraction of hardware design from concurrent register-transfer level (RTL). Among the forms of parallelism HLS exploits, pipelining demands fine-grained control over pipeline structure and hazard resolution to achieve competitive power, performance, and area (PPA). However, existing tools either lack such control or sacrifice sequential semantics to provide it. We present an HLS tool that embeds fine-grained pipeline control in a sequential programming model while generating synchronization logic only where needed to resolve the pipeline's potential hazards. The tool builds on visibility control, a novel programming abstraction that unifies hazard resolution techniques including stalling, bypassing, speculation, deferred commit, and register renaming. We evaluate on in-order RISC-V cores, histograms, and an AES accelerator. On RISC-V cores, we implement stall, bypass, speculation, and register renaming; on histograms, we implement scheduling strategies that previously required RTL or concurrent programming models. Compiled pipelines outperform HLS tools with sequential semantics and achieve PPA comparable to hand-written RTL.

cs.PL · cs.AR

新论文 · v1arXiv:2611.21077

MORE: Merge Once, Route Early for Training-Free Multi-Subject Composition from Large LoRA Pools

Tian Xia

原文摘要

Public hubs offer many personalized LoRAs per text-to-image backbone, yet composing two of them in one image remains fragile. In FLUX, keeping a pool of $K=10$ or $20$ subject LoRAs loaded makes their deltas interfere, and even loading only the two requested LoRAs keeps both subjects in fewer than 4% of images: one subject vanishes, the two fuse, or the image splits into a collage. We present MORE (Merge Once, Route Early), a training-free two-stage pipeline for FLUX. Merge once: the whole pool is folded into a single fixed-rank adapter by the subspace-regression router of SSR-Merge, made reliable by ridge, rank and calibration choices, and factored exactly into a shared part and per-subject parts. Route early: during the first 30% of the denoising steps each subject's part is gated to its layout region; a decaying soft attention bias ties every region to its own subject tokens, and a cyclic swap guidance extrapolates away from the prediction with the subjects' regions permuted, which counteracts region homogenization. On held-out seeds, MORE keeps both subjects in 94.2% ($K=10$) and 78.3% ($K=20$) of images, against 50.8% / 35.8% for FreeFuse and 25.0% / 24.2% for LoRAShop, and reaches an ImageReward of 1.32 (best baseline 0.07), while all $K$ subjects stay in one adapter. With layouts inferred automatically it still keeps both subjects in 70.0% / 55.0% of images.

cs.CV

新论文 · v1arXiv:2611.21076

Counting What Matters: Reducing Redundant Object Detections in Overlapping UAV Mapping Imagery for Disaster Response

Max Schulte, Hartmut Surmann, Julien Meine

原文摘要

UAV mapping flights enable rapid aerial reconnaissance of large-scale incident sites in civil protection. However, the image overlap required for orthophoto generation causes every physical object to be detected multiple times across images, while dynamic objects may be omitted from or duplicated in the resulting orthomosaic. Analysis based on the orthophoto alone is therefore structurally insufficient. We address this by detecting objects on the individual images and merging redundant detections across them. This task lies between multi-object tracking and re-identification, as mapping flights form temporally ordered but sparsely sampled sequences with object reappearances across flight strips. We fine-tune YOLO detectors on a new domain-specific nadir dataset from real civil protection operations, roughly doubling mAP50 over the previous system, and propose re-identification approaches based on 3D positions from sparse reconstruction, DINOv3 appearance embeddings, and their combination via constrained greedy clustering. %On four manually annotated real-world mapping flights, the combined weighted multi-pass approach achieves the best trade-off between association accuracy and count error. On four manually annotated real-world mapping flights, an association-only evaluation using ground-truth detections shows that the combined weighted multi-pass approach provides the best trade-off between association quality and count error. It is integrated into the open-source web application ARGUS and visualized as verifiable, correctable object groups for first responders. https://github.com/RoblabWh/argus

cs.AI

新论文 · v1arXiv:2611.21073

Semantic Validity vs. Graphical Arrangement in Visualization Grammars

Cynthia A. Huang, Matthew Kay

原文摘要

Systems for modular, declarative and compositional specification for data visualizations, otherwise known as visualization grammars, often claim Wilkinson's Grammar of Graphics as motivation, inspiration or foundation. However, despite the shared intellectual lineage, visualization grammars can vary greatly in design scope and focus. In particular, systems that strongly adhere to domain-specific notions of correctness must trade-off expressivity for validity. This trade-off between semantics and graphical flexibility is currently under-articulated, leading to difficulty integrating various GoG-inspired visualization contributions, especially between InfoVis and Statistical Graphics. We propose three new types of visualization contributions to support such integration: Semantic Validity Theories, Semantic Visualization Systems and Graphical Arrangement Systems. The first two prioritize the adherence of data visualizations to correctness conditions, such as statistical or mathematical transformations and data semantics, whilst the third largely leave validity concerns to the user. We illustrate these new contribution types using examples from two recent grammar of graphics inspired papers, and show how they supported more nuanced notions of expressivity and usability in evaluation and design of visualization systems.

cs.HC

新论文 · v1arXiv:2611.21069

CoopDistill: Equilibrium-Guided Multi-Agent Policy Distillation for Cooperative Driving

Shiyu Fang, Sheng Zhao, Jiarui Zhang, Chengkai Xu, Xiaocong Zhao, Peng Hang, Jian Sun

原文摘要

Cooperative driving extends autonomous driving from individual vehicle decisions to joint maneuvers that improve traffic safety and efficiency. Optimization-based methods impose safety constraints but often fix paths to simplify multi-vehicle planning, limiting spatial flexibility. Multi-agent reinforcement learning (MARL) explores alternatives through interaction, but coordinated learning and safe execution remain challenging. Their integration nevertheless faces challenges in exploration efficiency, experience allocation, and execution reliability. To address these challenges, this paper proposes CoopDistill, an equilibrium-guided multi-agent policy distillation framework. A potential-game (PG) formulation provides a fixed-path longitudinal reference, around which a residual MARL policy learns longitudinal acceleration, conflict-approach timing, and lateral path corrections. A hard-case curriculum concentrates experience on interactions that PG leaves unresolved or substantially delayed. A comparative rollout selector evaluates the joint consequences of learned corrections against PG before choosing the executed action. Experiments in merging, roundabout, and intersection scenarios compare CoopDistill with six baselines at four connected automated vehicle penetration levels. Relative to PG, success improves by 2.1--27.1 percentage points and delay decreases by 7.3--27.3\%, with no observed collisions. Code is available at https://github.com/FanGShiYuu/CoopDistill

cs.RO · cs.MA

版本更新 · v2arXiv:2609.33496

Chameleon: Dynamic Format Adapter for Efficient Diffusion

Arnab Sanyal, Sandeep Chinchali

原文摘要

Post-training quantization (PTQ) is the standard way to run modern diffusion models on memory-constrained accelerators, yet every existing diffusion PTQ scheme fixes the $\mathit{number\ format}$ in advance and only tunes the scale, zero point, or per-layer bit-width. At a fixed bit-width the best format depends on the distribution being encoded, and that distribution differs across weight channels, across layers, and along the diffusion timestep, where activation distributions slide from heavy-tailed and noise-dominated to tightly clustered and structured. We propose Chameleon, a PTQ framework that holds the bit-width fixed and treats the format itself as a discrete variable, chosen per weight channel and per activation layer, optionally refined per timestep bucket. Activation formats come from {INT8, FP8 E4M3, FP8 E5M2, MXFP8, MXINT8}, selected ahead of time from two cheap statistics (empirical kurtosis and the closed-form diffusion SNR) and stored in a lookup table; weight formats come from {INT8, MXINT8} at 8 bits or {INT4, NF4, FP4 E2M1, MXINT4, MXFP4} at 4 bits, selected offline by reconstruction error. An architectural fork adapts the same selection layer to multi-step UNets, single-step distilled models, and Diffusion Transformers. Most of the gain comes from the per-layer choice; the timestep refinement helps mainly at 6-bit activations. Across SDXL, SDXL-Turbo, and PixArt-$α$ on COCO-2014, Chameleon achieves the best FID in all six backbone $\times$ bit-width settings, with CLIP within 0.24 of the FP16 reference and the best of all quantized methods at $W_{4}A_{8}$. On PixArt-$α$, a real mixed-format kernel runs 1.09$\times$ faster than FP16 end-to-end with 3$\times$ smaller weights.

cs.LG · cs.AI · cs.CV

版本更新 · v3arXiv:2609.18682

Rank and computation of the pathlifting Jacobian of a DAG ReLU network

Manon Verbockhaven

原文摘要

This paper provides a self-contained proof of the rank of the pathlifting Jacobian of a DAG ReLU network by performing an induction on the network's number of hidden nodes. In fact, the induction is elementary, and the key recipe is to consider the skeleton matrix of the network, a sparse matrix encoding the network paths, and transform the representation of one of its hidden neurons into an output node. The proof relies on intermediate propositions which link the pathlifting, its Jacobian, the network parameters, and its skeleton matrix, which, on top of permitting to conclude on the rank of the pathlifting Jacobian, also provide a way to compute it without backpropagation and whose computation cost is super efficient in practice compare to usual backpropagation. The paper is provided with a Python module that implements the different propositions of the paper for feed forward networks and is used to experimentally quantifies the computational gain of computing the pathlifting Jacobian with the proposed theory.

stat.ML · cs.LG

版本更新 · v2arXiv:2607.13941

Peak-End-Net: A Peak-End Rule Inspired Framework for Generalizable Video Aesthetic Assessment

Geng Li, Haiwen Li, Rui Chen, Jing Tang, Lei Sun, Xiangxiang Chu

原文摘要

Video aesthetic assessment (VAA) aims to predict how aesthetically pleasing a video is, yet remains far less explored than other visual assessment tasks. Its progress is hindered not only by the scarcity of large-scale benchmarks, but also by the intrinsic subjectivity of aesthetic judgment, which is shaped by human perception. In this paper, we revisit VAA from a psychological perspective and propose \textit{Peak-End-Net}, a lightweight and interpretable framework inspired by the \textit{peak-end rule}, which suggests that people tend to judge a temporal experience mainly according to its salient moments and the ending. Building on this intuition, we first transfer knowledge from image aesthetic assessment (IAA) to VAA by introducing a pretrained IAA head to produce frame-wise aesthetic priors, which serve as surrogate signals for identifying aesthetically salient moments and guiding \textit{peak-end rule}-based temporal aggregation. To further capture how a video evolves aesthetically over time, we design an aesthetic rhythm encoder that models temporal progression beyond isolated moments. Additionally, we refine the overall assessment through a dynamic gated fusion mechanism to improve robustness under distribution shift. Our method is built on a frozen vision transformer (ViT) and requires only a small number of trainable parameters, making it scalable and parameter-efficient. Extensive experiments on two existing VAA benchmarks, including in-domain evaluation on VADB and cross-domain testing on DIVIDE-3K, demonstrate that our approach achieves state-of-the-art performance, affirming the value of psychologically grounded modeling for VAA. Our code and models are available at https://github.com/AMAP-ML/Peak-End-Net.

cs.CV

版本更新 · v3arXiv:2609.02486

ViSAR: Training-Free Adaptive-$k$ Retrieval for Document Visual Question Answering

Adrien Mialland, Marc Plantevit, Julien Gallois, Céline Robardet

原文摘要

Document Visual Question Answering (DocVQA) often leverages Retrieval-Augmented Generation (RAG), where late-interaction encoders are commonly used to identify document pages relevant to a user query, before answer generation by a Large Vision-Language Model (LVLM). Existing approaches typically retrieve a fixed top-$k$ number of pages regardless of query complexity, which increases LVLM latency and may degrade answer accuracy when fewer pages would be enough. We introduce ViSAR (Visual Semantic Activation Retrieval), a training-free, budget-aware adaptive-$k$ retrieval method for late-interaction visual document retrieval. ViSAR operates directly in the embedding space to construct a query-conditioned page-level similarity matrix that highlights query-relevant semantics and dynamically determines the number of pages to retrieve, within a given page budget. We evaluate our method across multiple datasets, encoders, and LVLMs. ViSAR retrieves compact, query-adapted page sets that reduce end-to-end RAG latency by up to 58.9\%, with no statistically significant accuracy loss and significant gains in several settings, compared with fixed top-$k$ and adaptive retrieval heuristics. Furthermore, we show that the similarity matrix structure correlates with answer accuracy, suggesting future directions for retrieval quality-aware document understanding.

cs.IR · cs.AI · cs.CL · cs.CV

版本更新 · v2arXiv:2605.05781

Steering Visual Generation in Unified Multimodal Models with Understanding Supervision

Zeyu Liu, Zanlin Ni, Yang Yue, Cheng Da, Huan Yang, Di Zhang, Kun Gai, Gao Huang

原文摘要

Unified multimodal models are envisioned to bridge the gap between understanding and generation. Yet, to achieve competitive performance, state-of-the-art models adopt largely decoupled understanding and generation components. This design, while effective for individual tasks, weakens the connection required for mutual enhancement, leaving the potential synergy empirically uncertain. We propose to explicitly restore this synergy by introducing Understanding-Oriented Post-Training (UNO), a lightweight framework that treats understanding not only as a distinct task, but also a direct supervisory signal to steer generative representations. By incorporating objectives that encode semantic abstraction (captioning) and structural details (visual regression), we enable effective gradient flow from understanding to generation. Extensive experiments on image generation and editing demonstrate that understanding can serve as an effective catalyst for generation.

cs.CV · cs.AI