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

+479新增论文
+361其中 AI 相关
+64已有论文修订
+93新增 AI 导读

arXiv 最近完成扫描:2026-11-16 01:34 · 全站最近入库:2026-11-16 21:56各信源进度 →

统计口径与数据新鲜度

统计 2026-11-15 22:31 — 2026-11-16 22:31(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:2026-11-16 01:34最近发现新增版本:2026-11-16 21:56库内论文最新更新时间:2026-11-14 02:59

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

23,827 已收录论文19,941 AI 相关6,922 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 17,708 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-16 01:33(北京时间)
新论文 · v1arXiv:2611.15381

Low-Cost Mobile Bimanual Manipulation for Long-Horizon Toilet Cleaning

Sven Schreiber, Bernard Burdiek, Lasse Huber-Saffer, Yung-Hui Li, Donglin Chen, Shang-Ching Liu, Jianwei Zhang

原文摘要

Vision-Language-Action (VLA) models have shown promising capabilities for robot manipulation, but their performance on accessible mobile bimanual platforms remains underexplored. We present a low-cost and reproducible mobile bimanual platform integrating two LeRobot SO-101 manipulators with a TurtleBot 4 and a LeRobot-ROS 2 software stack. Using long-horizon toilet cleaning as a case study, we collect 250 real-world demonstrations and evaluate $π_0$, $π_{0.5}$, SmolVLA, and GR00T under the same embodiment and task conditions. We further evaluate variants using SARM-derived task-progress signals and short-term MEM observation history. Among the evaluated models, $π_0$ performs best, achieving 60% strict and 70% relaxed autonomous success, increasing to 90% with limited human intervention. In contrast, the evaluated progress- and memory-augmented variants do not improve their corresponding baselines. Our results demonstrate the feasibility of long-horizon mobile bimanual manipulation on accessible hardware while highlighting remaining challenges in reliable VLA deployment. Project website: https://so101-turtlebot4.github.io/mobile-bimanual-cleaning/

cs.RO

版本更新 · v2arXiv:2603.01514

Training Dynamics of Softmax Self-Attention: Fast Global Convergence via Preconditioning

Gautam Goel, Mahdi Soltanolkotabi, Peter Bartlett

原文摘要

We study the training dynamics of gradient descent in a multilayer softmax self-attention network trained to perform linear regression and show that in the infinite-data limit this regression problem is equivalent to a matrix factorization problem. This connection motivates the design of Hera, a new first-order optimization algorithm for training self-attention networks. Key features of Hera include a reparameterization of the model arising from the matrix factorization geometry and a rescaling of the parameters which ensures that the empirical loss gradients remain close to the auxiliary matrix factorization gradients throughout training. Hera updates in the reparameterized model are equivalent to a certain preconditioned update in the original parameterization. We also propose a variant of Hera called Hera-PSD; our main theoretical result is that under small Gaussian initialization the excess risk of the model trained by Hera-PSD decreases exponentially fast in the number of gradient descent iterations, up to an additional error term that decreases as $1/n$, up to logarithmic factors, in the size of the training set. A particularly appealing aspect of our excess risk bound is that it does not degrade as the number of self-attention layers increases. In a stylized time-series prediction experiment in a nonlinear dynamical system, we find that a model trained by Hera achieves $98.1\%$ lower mean-squared test error than a model trained by AdamW and $97.7\%$ lower than a model trained by Muon at the same iteration budget.

cs.LG · stat.ML

新论文 · v1arXiv:2611.15378

A Modified Glicko Model for Rating Soccer Clubs and National Teams Worldwide

Andrei Shelopugin, Alexander Sirotkin

原文摘要

Comparing soccer teams that rarely or never meet is a fundamental challenge in sports analytics: clubs play mostly within their national leagues, and national teams within their confederations. We propose a rating framework that places clubs and national teams worldwide on a common probabilistic scale. Building on a recent extension of Glicko with order effects and strength-dependent draw probabilities, we adapt the model to soccer with goal-scoring adjustments, a decomposition of home advantage, extra time, league-aware rating initialization and rules of promotion and relegation that do not inflate the levels of leagues; national teams start from the levels of their confederations. The models are evaluated on about 1.2 million club matches of 252 leagues from 166 countries (2000-2026) and 12.6 thousand national team matches (1998-2026). On held-out seasons, the modified Glicko model is clearly more accurate than Elo and as accurate as score-matrix models built on the goal expectations of a gradient boosting model, including the Dixon-Coles and Markov chain models, while it is substantially more accurate in international club matches. Averaging the predictions of the rating model and the Markov chain model is more accurate than either model alone. For national teams, the model is significantly more accurate than all baselines, especially in the final tournaments, where teams of different confederations meet. The resulting ratings are interpretable and can be used for player evaluation, transfer analysis and match forecasting. The match data, the ratings and the code are publicly available at https://github.com/andreyshelopugin/SportsRatings

cs.LG · stat.AP

版本更新 · v2arXiv:2611.13028

GTAGym: AAA Game Tasks Environment for Evaluation and Reinforcement Learning

Jiajun Chen, Shicheng Fang, Jiahao Lu, Yuxin Wang, Xipeng Qiu

原文摘要

AAA games offer a rich testbed for evaluating agent perception, reasoning, and control. However, evaluating whether agents can complete tasks in AAA games is challenging and often human-annotated because task outcomes are often difficult to verify automatically, let alone applying it for Reinforcement Learning (RL). We introduce GTAGym, an environment and framework for evaluation and RL on GTA V tasks. Its evaluation benchmark, GTAVBench, derived from GTA V main-story missions, contains 103 online tasks reconstructed across nine task families, alongside 47 image-based shooting and sniping tasks using GTA V imagery. A web server provides access to the online environment and evaluation interface, supporting parallel testing through screenshots and keyboard-mouse actions with automatic engine-state outcome checks. We evaluate six frontier LLMs on GTAVBench with textual task instructions and on its GTAVBench-AGI branch, where instructions appear only in the first screenshot and a harness automatically compresses interaction history. The best configuration achieves 70.9% overall accuracy in GTAVBench-AGI, yet aerial navigation, evasion and interception remain difficult. GTAGym also provides a scalable generator that samples task parameters within user-specified ranges and option sets under fixed templates, creating batches of 3D and image-based shooting environments for RL with automatic reward and termination signals. LoRA RL of Qwen3.8-27B on generated on-foot navigation tasks improves the in-domain performance and transfers to out-of-domain task categories, and native GTA V tasks from 15% to 23%.

cs.CV · cs.AI · cs.CL

新论文 · v1arXiv:2611.15373

SecDVA: A Maliciously Secure Multi-Party Framework for Data Value Assessment in Modern Machine Learning Data Markets

Lixu Wang, Xinpeng Yang, Meng Hao, Guangyi Zhang, Tianwei Zhang, Wei Dong, Chenkai Weng

原文摘要

Data Value Assessment (DVA) plays a critical role in modern data markets by enabling buyers to evaluate how a seller's data would improve model performance before purchase. However, existing DVA approaches require one party to reveal the data or model, exposing participants to privacy leakage, strategic manipulation, and high deployment costs. These limitations motivate a DVA protocol that is secure, reliable, widely applicable, and efficient enough for practical deployment in modern data markets. In this paper, we present SecDVA, the first maliciously secure multi-party computation (MPC) framework for reliable and practical DVA in modern machine learning data markets. SecDVA ensures that neither the seller's data nor the buyer's model and test data are disclosed beyond the final DVA outcome, while remaining robust against malicious behavior. To address the efficiency and applicability challenges of DVA for large models, we use the neural tangent kernel theory to linearize model updates and introduce lightweight mini-batch approximations to significantly reduce computation overhead. We further design a (2+1)-party MPC protocol based on function secret sharing, together with optimized non-linear primitives. In addition, we optimize the underlying MPC backends, the distributed point function, and the distributed comparison function, reducing the number of Advanced Encryption Standard calls to improve efficiency. Extensive experiments demonstrate that SecDVA outperforms existing baselines in the plaintext setting and enables DVA for large-scale models in a reasonable time under MPC.

cs.CR

新论文 · v1arXiv:2611.15372

LLMRunner: Closed Workload Model Performance Testing Framework for LLM Servings

Arush Upadhyaya, Debadatta Mishra

原文摘要

Performance testing large language model (LLM) serving engines is a crucial requirement considering the criticality of the infrastructure for deploying conversational AI at scale. Most of the benchmarking tools used to carryout performance evaluation of LLM serving do not capture realistic workload orchestration unlike closed workload performance testing infrastructure used in other domains such as web and database engines. For example, state-of-the-art LLM benchmarks for performance testing model traffic as streams of independent single-turn and stand-alone requests omitting the end-user behavior such as think-time between conversational turns. We present LLMRunner, a performance tester for LLM serving that addresses this gap by providing a framework to generate realistic load consisting of multi-turn conversations with configurable end-user parameters such as number of users in the workload loop, think-time distributions etc. Moreover, LLMRunner provides declarative matrix-sweep experiment plans, integrated GPU and KV-cache monitoring, and an automated artifact collection and plotting subsystem. We use LLMRunner to comparatively analyze the performance of LLM serving when NVIDIA CUDA Unified Virtual Memory (UVM) is used as a drop-in replacement memory allocator for the Key-Value cache (KV-Cache). Our test demonstrates that UVM incurs no measurable throughput or latency penalty compared to the baseline memory allocation technique for non-oversubscription scenarios.

cs.PF

新论文 · v1arXiv:2611.15370

Supervision-Aware Evaluation of MedGemma for Reference-Assisted Radiotherapy Contour Ranking

Jaden Moon, Caiwen Jiang, Songlin Zhao, Wei Liu

原文摘要

Radiotherapy contour ranking can support quality assurance, but vision-language model (VLM) gains may reflect supervision, score construction, available inputs, or representation access. We evaluated frozen MedGemma 1.5 in a single-institution study, a 12-patient held-out HaN-Seg test, five-fold cross-fitting on 30 non-test HaN-Seg patients, and transfer without refitting to 120 SegRap2023 patients. On held-out HaN-Seg, a supervised readout of the final prompt state exceeded a supervised readout of the model's yes/no output probabilities by +0.3047 pooled average precision (AP; 95% patient-bootstrap interval [+0.2822, +0.3269]); the advantage persisted against a parameter-count-matched nonlinear control. A readout of projected visual tokens exceeded the final-state readout in cross-fitting and transfer (+0.0706 AP averaged equally across patients on SegRap; 95% interval [+0.0607, +0.0804]), while a structured panel-feature readout remained competitive with visual tokens. The anatomy-conditioned readout did not demonstrate an advantage over the native-feature comparator. These results motivate visual-token and structured-feature controls in medical-imaging VLM evaluations; outcomes were computational or synthetic, not clinical acceptability.

eess.IV · cs.CV · cs.LG

新论文 · v1arXiv:2611.15368

Bind-WAM: Binding-Uncertainty-Guided Selective Future Imagination in World Action Models

Chenhao Qiu, Shuyi Chen, Songen Gu, Shufeng Nan, Tianyu Wang, Siyu Lin, Yifan Li, Bingwen Zhu 等 9 位作者

原文摘要

World Action Models (WAMs) can improve robot learning by jointly predicting future visual states and actions, but generating a visual future at every replanning step is expensive and sometimes unnecessary. This raises a critical question: when should a robot policy imagine the future? In this paper, we propose Bind-WAM, a binding-uncertainty-guided WAM for selective future imagination. Bind-WAM introduces language-object binding uncertainty, selectively invoking joint future-action generation when the current observation does not support stable task-relevant action generation. An offline labeling pipeline uses complete demonstrations and action context to identify task-relevant objects and detect reference mismatch, invisibility, and identity drift. The resulting pseudo-labels supervise a lightweight router that predicts binding uncertainty from the current observation and language instruction. At inference, a shared WAM policy invokes selective future imagination according to the predicted uncertainty and its change between consecutive replanning steps. Simulation and real-world experiments demonstrate that Bind-WAM can establish binding uncertainty as a strong, interpretable signal for selective future imagination.

cs.RO

版本更新 · v3arXiv:2611.03043

VideoEditor-MoE: Scaling Complex Video Editing with Mixture of Experts

Xiao Cao, Fuchen Long, Yansong Qu, Wen Xiao, Xin Zhang, Xinlong Sun, Xi Chen, Zhiyong Huang

原文摘要

Complex video editing requires understanding complex instructions while jointly fulfilling multiple editing requests. Meeting these demands through dense model scaling substantially increases computational cost. We introduce VideoEditor-MoE, the first open-source mixture-of-experts framework for video editing. Beyond sparse MLP-MoE layers, MoE-Attention expands attention capacity through routed key--value experts while preserving global interactions among video tokens. Initialized from a 5B video editing model, VideoEditor-MoE scales to 14B and 56B parameters, with approximately 8B parameters activated per token in both variants. Comparisons with existing methods and two complementary baselines, Explicit-MoE and MLP-MoE, assess its benefits over explicit expert guidance and MLP-only expert architectures. Further analyses demonstrate that all experts remain active and reveal that learned routing responds more strongly to sample content and diffusion noise levels than to human-defined editing categories. Spatial analysis further indicates sensitivity to object-level semantic features. We also identify decoupled weight decay as a strategy for stabilizing expert scaling.

cs.CV · cs.AI

版本更新 · v2arXiv:2604.16667

Model-Predictive Spill-Free Emergency Stopping for Liquid-Manipulating Robots

Samuli Hynninen, Ville Kyrki

原文摘要

Manipulating open liquid containers is challenging because liquids are highly sensitive to vessel accelerations and jerks. Although spill-free liquid manipulation has been widely studied, emergency stopping under unexpected hazards has received little attention, despite the fact that abrupt braking may cause hazardous spills. This letter presents an emergency stop system for robots manipulating liquids in open containers. We formulate emergency stopping as an optimal control problem and solve it in a model predictive control framework to generate time-optimal, spill-free stopping trajectories. The method operates as a plug-and-play safety layer on top of existing slosh-free motion planning methods, enabling immediate reaction to detected hazards while accounting for nonlinear liquid dynamics. We demonstrate, through simulation and on a 7-DoF Franka Emika Panda robot, that the proposed approach achieves fast emergency stopping without spilling.

cs.RO

新论文 · v1arXiv:2611.15357

Multimodal Semantic Communication using Shared and Unique Representation Decomposition

Peng Yi, Ying-Chang Liang

原文摘要

Semantic communication (SemCom) has emerged as a promising paradigm for bandwidth-constrained mobile and vehicular networks. This paper proposes a unified representation learning framework for multimodal SemCom (MSC), which encodes heterogeneous data into a compact semantic space for efficient transmission and source recovery. The framework addresses the multimodal semantic representation gap, which arises when at least one modality contains semantic information that cannot be fully inferred from the other. To preserve this information for reconstruction, the framework decomposes each representation into shared and modality-specific components. We formulate a mutual information objective for source reconstruction, cross-modal alignment of the shared components, and decoupling of the shared and modality-specific components. To enhance transmission efficiency, we introduce a shared knowledge base implemented as a compact codebook of semantic vectors learned during training. The shared components are transmitted digitally as codebook indices, while the modality-specific components are sent in analog form. We apply knowledge distillation during training, using a pretrained vision foundation model to guide the learning of quantized shared features. On image--text pairs over additive white Gaussian noise (AWGN) and Extended Vehicular A (EVA) channels, the proposed framework improves image reconstruction over conventional separate source-channel coding (SSCC) and representative multimodal and unimodal SemCom systems. Further evaluations show gradual degradation under channel-estimation mismatch, complementary effects of alignment, decoupling, and knowledge distillation on the shared representation, and alignment of shared features across modalities with separation of the modality-specific features.

cs.IT · eess.SP

版本更新 · v3arXiv:2605.00950

Equation-Free Digital Twins for Nonlinear Structural Dynamics

Mohammad Mahdi Abaei, Ahmad BahooToroody, Arttu Polojärvi, Heikki Remes, Ulf Tyge Tygesen, Mikko Suominen, Michael Beer

原文摘要

Monitoring floating offshore wind turbines (FOWTs) under stochastic excitation and sensor failure requires computationally efficient and physically interpretable estimators. This paper introduces an equation-free, output-only digital twin framework combining Hankel dynamic mode decomposition (Hankel-DMD), sparse-sensor reconstruction, and a stochastic Kalman extension. A sufficient identifiability condition is derived for modal amplitudes recovered from delay-embedded measurements, while Tikhonov regularization addresses ill-conditioned sensor layouts. Correlation-domain identification improves damping estimates under stochastic forcing and distinguishes structural modes from rotor harmonics. The framework is validated using OpenFAST simulations of an NREL 5-MW spar turbine and experimental basin-test measurements from a 1:70-scale IEA 15-MW semi-submersible. In simulations, the identified tower modes agree with established operational modal analysis methods (MAC at least 0.99), while the Kalman estimator achieves R-squared values of 0.91 to 0.99 under severe sensor sparsity. Experimental results identify tower modes at 0.430 and 0.415 Hz with MAC of 1.00 against SSI-COV and reconstruct tower-base bending moments on unseen runs with R-squared values of 0.83 to 0.89 using three of four sensors. The framework provides a computationally efficient, sensor-failure-resilient structural-state estimation layer for digital twins.

eess.SP · cs.LG · eess.SY

版本更新 · v2arXiv:2509.24921

CineWild: Balancing Art and Robotics for Ethical Wildlife Documentary Filmmaking

Pablo Pueyo, Fernando Caballero, Ana Cristina Murillo, Eduardo Montijano

原文摘要

Unmanned aerial vehicles (UAVs) have become powerful tools across domains from industry to the arts. In documentary filmmaking, they offer dynamic, otherwise unreachable perspectives, transforming how stories are told. Wildlife documentaries especially benefit, yet UAVs also raise ethical concerns like the risk of disturbing the animals they aim to capture. This paper introduces CineWild, a model predictive control (MPC) formulation for autonomous aerial cinematography that combines robotics, cinematography, and ethics. Building on an MPC solution that jointly controls drone motion and camera focal length, CineWild dynamically adjusts flight paths and camera settings to balance cinematic quality with animal welfare. Key features include adaptive zoom for filming from acoustic and visual safe distances, path-planning that avoids an animal's field of view, and smooth, low-noise maneuvers. CineWild exemplifies interdisciplinary innovation, bridging engineering, visual storytelling, and environmental ethics. We evaluate the formulation in simulation, with ablation studies and a comparison against a cinematographic baseline and a welfare-aware baseline under identical weights, and we demonstrate the perception and camera-actuation pipeline in real-world proof-of-concept trials. We will release the code.

cs.RO · cs.MM

版本更新 · v2arXiv:2608.03664

Morphology-Aware Implicit Super-Resolution Network for Pathological Images

Jiaming Liang, QiHui Han, Haolin Chen, Chengxin Ye, Jiawen Liu, Jiazhou Chen, Xiaoqi Sheng, Hongmin Cai

原文摘要

Accurate diagnosis in Digital Pathology (DP) relies on high-resolution whole-slide images, yet clinical deployment is often limited by hardware costs. Super-Resolution (SR) offers a promising alternative by computationally enhancing low-resolution acquisitions. However, existing SR methods frequently struggle to preserve fine-grained cellular morphology, leading to texture oversmoothing and blurred structural boundaries under complex tissue variability. To address this issue, we propose Morph-ISR, a morphology-aware implicit super-resolution framework for DP that restores diagnostically relevant details with sub-pixel precision. Morph-ISR reformulates SR as a continuous coordinate-based reconstruction problem and integrates an Implicit Position-aware Kernel Generator (IPKG) to adaptively model spatially varying tissue morphology. To further enhance structural fidelity, a Morphological Fidelity Prior (MFP) is introduced, leveraging semantic guidance from a pre-trained cell segmentation network to enforce boundary-preserving and region-aware reconstruction, thereby improving the representation of critical cellular boundaries and nuclear textures. Experiments on TCGA and SurGen datasets show that Morph-ISR achieves the best LPIPS and ST-LPIPS among the evaluated methods, reducing them by up to 38.37% and 39.55%, respectively, over the second-best methods while maintaining strong PSNR and SSIM. These results demonstrate superior preservation of diagnostically relevant cellular boundaries and nuclear textures, while compact parameterization and high throughput support efficient edge deployment. Code and trained models will be released upon publication.

cs.CV

版本更新 · v2arXiv:2611.13711

Hint, Not Hole: Localization-Robust Video Overlay Removal

Dongyang Zhang, Wei Zhang, Renjie Zou, Zhiwei Huang

原文摘要

Video overlay removal recovers the scene beneath burned-in subtitles, karaoke lyrics, titles, credits, logos and watermarks. Generative video models now fill a known region convincingly, but in practice the region comes from a detector or a segmenter, and it is wrong in both directions: it misses strokes, halos, whole frames and whole objects, and it covers background between characters, around boxes and on genuine scene text. Erasers that treat the mask as a hole, regenerating inside and copying outside, inherit every one of these errors. We present HintErase, which treats localization as a hint, not a hole. The model sees the full, unmasked video together with a hint channel; during training the hint is corrupted the way real localizers fail, while supervision follows the true overlay. The result is a single model that removes overlay the hint missed, keeps background the hint wrongly covered, and is trained to keep genuine scene text even when the hint lands on it. For full-length releases, a streaming pipeline handles interlaced sources, shot cuts, missed and jittering hints, paste-back and audio, with memory independent of the length of the video. This report describes the approach, the pipeline that turns it into full-length deliveries, and qualitative results; video results, with input, hint and output side by side, are on the project page.

cs.CV

新论文 · v1arXiv:2611.15348

Breath-Dx: A Time Series Classification Benchmark for Disease Screening with eNose

Ivan Poddiakov, Ivan Kruzhilov, Galina Zubkova, Svetlana Erofeeva, Andrey Savchenko, Pavel Blinov

原文摘要

Electronic nose (eNose) technology offers a portable, non-invasive route for disease screening via exhaled breath analysis. However, progress in the field is hindered by the lack of systematic benchmarks that compare machine learning methods under realistic conditions, particularly sensor drift, a known but rarely quantified failure mode. Most existing studies rely on small, private datasets with inconsistent protocols, rendering results non-comparable and slowing progress. We introduce Breath-Dx, the first comprehensive benchmark of time series classification methods for breath-based disease screening. Our work leverages the Scent of Health (S-OH) dataset, the largest public clinical eNose collection to date, comprising 1,234 patients across nine diagnostic groups, each providing a 17-channel multivariate time series with explicit temporal train/test splits that account for sensor drift. We systematically evaluate 10 methods spanning four families: feature-based tabular learners, convolution-based models, graph-based, and time series architectures. All methods degrade significantly under temporal validation (0.05-0.12 AUC drop compared to random cross-validation), confirming that drift embeds a strong temporal signal. Under this realistic evaluation, the best methods achieve ROC AUC of 0.74-0.76 for some lung conditions, though others remain near chance. Recurrent architectures show the most consistent performance across diseases, while specialized time series methods achieve the highest peak results. Demographic confounders matter: breath signals provide independent diagnostic value for some conditions but add little beyond age for others. We release the S-OH dataset, all temporal splits, evaluation protocols, and code to enable reproducible research and accelerate progress in clinical breathomics.

cs.LG · cs.AI

新论文 · v1arXiv:2611.15346

Towards Better Legal Reasoning LLMs: Signal Balancing and Reward Scheduling

Yinghao Hu, Qian Xiao, Yuting Huang, Kaisong Song, Yiquan Wu, Ang Li, Junjie Cao, Jun Lin 等 10 位作者

原文摘要

In the legal domain, LLMs can improve reliability and interpretability by generating explicit reasoning traces. However, training LLMs to produce high-quality reasoning traces remains challenging. Existing supervised fine-tuning methods struggle when reasoning traces are significantly longer than final answers, as the learning signal for the answer becomes diluted. Meanwhile, reinforcement learning methods such as Group Relative Policy Optimization also have several limitations, including costly reward design, performance plateaus, and reward hacking. We therefore propose a two-stage training framework. In Stage I, JurisCoT-SFT employs a length-normalized dual objective to balance learning signals between reasoning traces and final answers. In Stage II, Lifecycle-Aware Backtrackable Policy Optimization dynamically evaluates and retires auxiliary reward signals based on their impact on final evaluation objectives, enabling efficient reward utilization. Our framework substantially improves legal reasoning performance on relevant tasks selected from multiple legal benchmarks, including ChJGBench, LawBench, and LegalBench. Experimental results demonstrate consistent improvements in both reasoning trace quality and final answer accuracy, while our models remain competitive with larger models.

cs.CL

版本更新 · v2arXiv:2508.14821

The C-index Multiverse

Begoña B. Sierra, Colin McLean, Peter S. Hall, Catalina A. Vallejos

原文摘要

Quantifying out-of-sample discrimination performance for time-to-event outcomes is a fundamental step for model evaluation and selection in the context of predictive modelling. The concordance index, or C-index, is a widely used metric for this purpose, particularly with the growing development of machine learning methods. Beyond differences between proposed C-index estimators (e.g. Harrell's, Uno's and Antolini's), we demonstrate the existence of a C-index multiverse among available R and python software, where seemingly equal implementations can yield different results. This can undermine reproducibility and complicate fair comparisons across models and studies. Key variation sources include tie handling and adjustment to censoring. Additionally, the absence of a standardised approach to summarise risk from survival distributions, result in another source of variation dependent on input types. We demonstrate the consequences of the C-index multiverse when quantifying predictive performance for several survival models (from Cox proportional hazards to recent deep learning approaches) on publicly available breast cancer data, and semi-synthetic examples. Our work emphasises the need for better reporting to improve transparency and reproducibility. This article aims to be a useful guideline, helping analysts when navigating the multiverse, providing unified documentation and highlighting potential pitfalls of existing software. All code is publicly available at: www.github.com/BBolosSierra/CindexMultiverse.

stat.ML · cs.LG · stat.AP

新论文 · v1arXiv:2611.15343

Hybrid Classical-Quantum Optimization for CVRPTW: A Systems Framework with Provable Feasibility and Empirical Hardware Limits

Syamasundar Gopasana, Swatantra Singh, Rohan Joydhar, Shubhashis Sengupta

原文摘要

Solving large-scale Capacitated Vehicle Routing Problems with Time Windows (CVRPTW) remains challenging because strict operational constraints and scale limit both exact and heuristic methods. We present a hybrid classical-quantum architecture combining Mixed Integer Linear Programming (MILP) with vehicle-level Quadratic Unconstrained Binary Optimization (QUBO), solved using Toshiba SQBM+ and QAOA on Rigetti Ankaa-3 and IQM Emerald quantum processors. The decomposition avoids the variable growth and constraint-enforcement difficulties of monolithic CVRPTW QUBO formulations. We evaluate the framework on five realistic benchmarks derived from London traffic and location data, spanning 10--70 customers, using a validation pipeline enforcing feasibility and monotone improvement over classical baselines. SQBM+ achieves verified fleet-distance reductions of 2.80% (7,927 m) for the 25-customer instance and 0.22% (1,186 m) for the 55-customer instance. QAOA experiments establish a current hardware boundary for reliable constraint satisfaction at subproblems of up to 4 customers (16 qubits). Diagnostic results identify assignment-level time-window tightness, rather than local route optimization, as the dominant scalability bottleneck. The results demonstrate a practical near-term role for hybrid classical-quantum optimization while establishing empirical limits for current quantum hardware.

quant-ph · cs.ET

新论文 · v1arXiv:2611.15342

MechBench: A Domain-Specific LLM Benchmark for Parallel Mechanisms

Zhihao Xia, Tingting Wang, Hui Li, Yongheng Xing, Chunxu Tian

原文摘要

Parallel mechanisms pose distinctive challenges to large language models (LLMs), as valid reasoning depends jointly on mechanism topology, loop closure, geometry, joint types, and modeling assumptions. Yet existing LLM benchmarks provide little systematic coverage of such domain-specific knowledge, constraint-sensitive analysis, and constructive design. To fill this gap, we introduce MECHBENCH, a 2,541-question benchmark for evaluating LLMs in parallel mechanisms across professional knowledge, mechanism analysis, and design. MECHBENCH contains standard and challenge tiers spanning six capability categories and nine executable analysis and design tasks. It is constructed through MECHFORGE, a domain-grounded framework that combines expert-organized knowledge, LLM-assisted question generation, programmatic instance generation, two-stage expert review, and task-aware scoring with exact matching, rubric-based evaluation, and program verification. We further develop MECHGPT as a diagnostic domain-adapted model. Experiments show that strong knowledge performance does not directly translate into executable mechanism reasoning: leading closed-service models achieve over 95 on the standard tier but only 47-60 on the challenge tier, while domain adaptation substantially improves MechGPT over its Qwen3-8B baselines but leaves a large gap in constrained analysis and design. These results establish MECHBENCH as a systematic testbed for measuring domain-specific LLM capabilities in mechanism science.

cs.LG

版本更新 · v2arXiv:2605.02319

Optimal Privacy-Utility Trade-Offs in LDP: Functional and Geometric Perspectives

Seung-Hyun Nam, Hyun-Young Park, Si-Hyeon Lee

原文摘要

Local differential privacy (LDP) has emerged as a gold-standard framework for privacy-preserving data analysis. However, characterizing the optimal privacy-utility trade-off (PUT) and the corresponding optimal LDP channels remains largely fragmented, relying on problem-specific, case-by-case analyses. In this work, we develop a unified theoretical framework that systematically characterizes the optimal PUT and optimal LDP channels for general privacy-preserving statistical decision-making problems. We first identify key functional properties of Bayesian and minimax risks as functions of the LDP channel, including the data processing inequality (DPI), direct-sum quasi-convexity (or additivity), concavity, and symmetry invariance. Leveraging these properties, we reduce the optimization domain required to compute the optimal PUT. To this end, we reinterpret the existing result implying the one-to-one correspondence between maximal LDP channels under the Blackwell order and a finite-dimensional polytope, through an alternative convex geometric derivation. This result renders the optimal PUT computationally tractable via vertex enumeration or linear programming. Furthermore, when the underlying problem exhibits symmetries characterized by a transitive group action, we show that the optimal PUT can be computed by comparing finitely many values, rather than optimizing over a high dimensional polytope. Our framework applies broadly beyond risk minimization, encompassing the maximization of information-theoretic measures. We demonstrate the efficacy of our theoretical framework by recovering or strengthening several known results, and deriving exact analytic expressions for the optimal PUTs in specific tasks that were previously unaddressed.

cs.CR · cs.IT

新论文 · v1arXiv:2611.15341

Janus-MOPD: Tracing the Student Learning Dynamics in Multi-Teacher On-Policy Distillation

Zongji Yu, Ming Zhang, Dongcheng Zhang, Jiawei Zhou, Haotian Pang, Yiliu Sun, Xiang Wang, Youbang Sun 等 10 位作者

原文摘要

Multi-domain capabilities are critical for practically applicable Large Reasoning Models (LRMs). Therefore, Multi-teacher On-Policy Distillation (MOPD) has become a crucial part of the post-training process, integrating the capabilities of multiple domain-specific teacher models into a single student model. However, the student usually fails to acquire multi-domain capabilities simultaneously due to undesirable domain heterogeneity. Existing works have address this problem in view of temporal uneven efficiency and spatial hard allocation in external multi-teacher supervision. While in broader multi-domain training, varying degrees of performance degradation is still largely unexplained. In this paper, we shift the focus on how the student learns and integrates multi-domain capabilities internally, and then propose \textbf{\textsc{Janus-MOPD}}, a general learning dynamics framework of geometric representations for LRMs. Specifically, our main focus is on the internal performance of multi-domain capabilities on both temporal and spatial dimensions within the student. Temporally, our study shows that the student exhibits a Expansion-Consolidation-Optimization (ECO) three-phase learning dynamics, together with inherent limitations in learning efficiency and scaling. Spatially, we summarize two types of subspace locking within and across domains by tracing the representation space of the student, while explaining its clear learning directions but inherent cross-domain conflict. Therefore, \textsc{Janus-MOPD} yields a unified learning dynamics formulation, a geometric tracing methodology, and systematic findings of how the student learns and integrates multi-domain capabilities. Overall, we provide a \textsc{Janus} perspective to internally trace the continual evolution of the multi-domain capabilities within the model itself.

cs.CL

版本更新 · v2arXiv:2601.16764

ReLU Networks for Model Predictive Control: Network Complexity and Performance Guarantees

Xingchen Li, Keyou You

原文摘要

Recent years have witnessed a resurgence in using ReLU neural networks (NNs) to represent model predictive control (MPC) policies. However, determining the sufficient network complexity to ensure closed-loop guarantees remains a fundamental open problem. This involves a critical precision-complexity trade-off: undersized networks may fail to capture the MPC policy, while oversized ones may outweigh the benefits of ReLU network approximation. In this work, we project the MPC actions onto a tightened constraint set to satisfy the hard constraints, and establish a state-dependent Lipschitz continuity property for the optimal MPC cost function, which enables sharp convergence analysis of the closed-loop system. For the first time, we derive explicit bounds on ReLU network width and depth for approximating MPC policies with closed-loop guarantees. To reduce network complexity for the convergence to user-specified target sets, we propose a non-uniform error framework with a state-aware scaling function to adaptively adjust both the input and output of the ReLU network. Our contributions provide a foundational step toward certifiable ReLU NN-based MPC.

eess.SY · cs.LG · math.OC

版本更新 · v4arXiv:2510.01720

Constructions of Efficiently Implementable Boolean Functions with Provable Nonlinearity/Resiliency/Algebraic Immunity Trade-Offs

Palash Sarkar

原文摘要

We describe new families of efficiently implementable Boolean functions achieving provable trade-offs between resiliency, nonlinearity, and algebraic immunity. Compared to one direction of research on balanced and 1-resilient functions with optimal algebraic immunity, the new functions provide lower algebraic immunity, but much lower -- in fact, almost optimal -- linear bias and importantly are much more efficient to implement. Compared to another direction of research based on constructions using direct sum, the new functions while achieving similar efficiency, has less number of variables for the same order of resiliency, same lower bound on algebraic immunity, and similar nonlinearity.

cs.CR

新论文 · v1arXiv:2611.15339

Interactive Perception of Granular Materials via Robotic Scooping and Force-Torque Sensing

Samuli Hynninen, Mohammad Amman, Meri Mäkelä, Anirvan Dutta, Ville Kyrki

原文摘要

Granular materials exhibit complex and highly variable interaction dynamics, making their reliable manipulation challenging without knowledge of their physical properties. We present an interactive perception framework that infers granular material properties directly from force-torque measurements acquired during scooping. The proposed Latent Filter++ combines a bio-inspired frequency-temporal filter-bank representation with a causal latent dynamical model, allowing the representation to evolve as interaction observations become available while avoiding hand-crafted signal preprocessing. The learned latent representation captures physically meaningful variation in granular materials and supports estimation of properties including grain size and bulk density. Evaluated on 26 diverse granular materials, the approach generalizes to materials excluded from training and enables few-shot adaptation of property-specific mappings using a small number of additional interaction samples. Finally, we demonstrate the relevance of the inferred properties to downstream manipulation by examining their correlation with the maximum acceleration the materials can withstand without spilling from the spoon during transport.

cs.RO

版本更新 · v2arXiv:2607.05243

GUSH3R: Everyone Everywhere All at Once as Gaussians

Keito Abe, Kaede Shiohara, Takashi Otonari, Toshihiko Yamasaki

原文摘要

Reconstructing dynamic human-scene environments from monocular videos is a challenging problem that requires jointly modeling scene geometry, camera motion, and non-rigid human dynamics while enabling photorealistic rendering. Recent feed-forward methods can efficiently predict geometry, but they are often limited to non-photorealistic representations such as point clouds and meshes, or they fail to handle non-rigid objects, particularly dynamic humans. To fill this gap, we present GUSH3R (Gaussian-Unified Scene Human 3D Reconstruction), a feed-forward framework for online dynamic human-scene reconstruction. From a monocular human-scene video, our method reconstructs dynamic humans (everyone) and static scenes (everywhere) in a single forward pass (all at once) as 3D Gaussian Splatting (3DGS) primitives (as gaussians), which are geometrically consistent and capable of novel view synthesis. Experiments on monocular human-scene datasets demonstrate that our approach achieves competitive novel view synthesis quality while significantly improving inference efficiency compared to optimization-based methods.

cs.CV

版本更新 · v3arXiv:2607.11465

What Score Distillation Transfers to Compact Dense Retrievers

Kirill Dubovikov, Martin Takac, Salem Lahlou

原文摘要

Compact retrievers are often specialized by distilling a large embedding model's similarity scores on hard negatives mined by the teacher model. We argue that this common setup calls for a different loss than the scale-dependent ones usually used, such as MarginMSE and PairMSE. When the teacher discards candidates scored too close to the labeled positive, the negatives it keeps get nearly identical scores. These losses then mostly ask the student to reproduce this tiny score range rather than the teacher's ranking. A per-row Pearson correlation loss, invariant to the shift and scale of the scores in each row (a query and its 32 candidates), avoids this failure, and removing PairMSE's scale term alone recovers most of Pearson's lead. We propose row scale dispersion, computed from the untrained student before training, to tell when this matters. With this loss, on rows mined around a labeled document, the teacher's scores substitute for relevance labels rather than add to them. Training on the binary label alone works as well, with the same loss or, for one student, a contrastive loss: the teacher's value lies in choosing the negatives. Without labels, the teacher's scores on its own top-ranked documents recover most of the labeled gain, largely through its top document as a pseudo-label. Experiments with two compact students on three BEIR datasets support both conclusions, including, for one student, comparisons with MarginMSE and a tuned listwise KL. We suggest a simple practical rule: distill teacher-mined negatives with a scale-free loss and, on such rows, use the teacher's scores where labels are missing rather than on top of them.

cs.IR

新论文 · v1arXiv:2611.15329

TabOmics: A Small, Open Tabular Foundation Model for Wide Omics Data

Andrea López-García, Luis Bote-Curiel, Francisco Javier Gimeno-Blanes, José Luis Rojo-Álvarez, Sergio Muñoz-Romero

原文摘要

Omics studies often produce tables with thousands to tens of thousands of molecular features and few samples. Tabular foundation models use diverse strategies to predict labels for such tables, with some relying on in-context learning, while approaches to wide tables may involve subsampling, grouping, or compressing features. We present TabOmics, a small, open family of tabular foundation models with 7.3M parameters, pretrained exclusively on synthetic tables. TabOmics processes all features retained after preprocessing jointly and derives predictions, attention-derived feature-importance scores, and test sample embeddings from the same forward pass. On 46 classification tasks from three benchmarks of high-dimensional, low-sample-size tables, TabOmics v2.5 with eight ensemble members achieves the highest mean area under the receiver operating characteristic curve (AUROC) among the evaluated models (0.943), outperforming TabPFN-3.5 with eight members on 32 tasks despite having about 30 times fewer parameters. With a single member, TabOmics v2.5 reaches a mean AUROC of 0.941, versus 0.942 for TabPFN-3.5 with eight members, with no significant paired difference under the reported task-level analysis. On synthetic tables, its feature-importance scores achieve feature recovery similar to that of TabPFN-Wide, and its predictions remain above chance at 30,000 features. On breast cancer tasks, the top ranked genes include PAM50 genes, which define the subtype labels, indicating concordance with the label definition rather than independent biological validation. The code, model weights, training recipes, and experimental artifacts are publicly available.

cs.LG

新论文 · v1arXiv:2611.15328

Vision-Grounded Semantic Learning and Alignment for Scalable Multilingual Text-Based Person Retrieval

Xinyu Zhou, Fan Li, Shasha Chen, Min Cao, Ahtamjan Ahmat, Qiaoming Zhu, Mang Ye

原文摘要

Multilingual Text-Based Person Retrieval (TBPR) aims to retrieve individuals with texts in multiple languages. Despite its practical importance, it has been overlooked by existing research, which mainly focuses on English-only retrieval. The only prior attempt at multilingual TBPR suffers from (1) the absence of a large-scale, multilingual pretraining dataset and (2) a suboptimal model architecture for multilingual text encoding and cross-modal alignment, which employs a shared encoder that overlooks linguistic heterogeneity and relies on English as a pivot for cross-modal alignment, introducing English-centric bias. For this, we contribute MultiPedes, the first large-scale, pedestrian-centric multilingual pretraining dataset. It contains 10 million image-text pairs across four languages, constructed by a multilingual quality-aware data construction pipeline. We also propose ViSeLA (Vision-grounded Semantic Learning and Alignment), a novel framework featuring: (1) a semantic-aware multilingual learning module that extends the mixture-of-experts architecture by routing language-invariant semantics to specialized experts for improved multilingual feature learning; and (2) a vision-grounded cross-modal alignment module that employs a semantic probe to capture fine-grained semantics across modalities, enabling direct cross-modal alignment without relying on a pivot language. Extensive experiments on both multilingual and English-only TBPR datasets demonstrate the effectiveness of ViSeLA. The dataset and code are available at https://github.com/Flame-Chasers/ViSeLA

cs.CV · cs.CL · cs.IR

版本更新 · v2arXiv:2611.14244

An AI Assistant for Automating Anti-Corruption Expertise of Draft Legislation Based on the TRACE Framework

Serhii Zabolotnii, Alina Tarasova

原文摘要

We present an AI assistant for automating anti-corruption expertise of draft legislation based on the TRACE architecture, designed to screen bills for corruption risks: pinpoint provisions an expert would criticise and name the regulatory defect in each. The pipeline combines a deterministic parsing core, model-guided single-provision qualification, rule-based factor assignment, and external human oversight. To evaluate the assistant, we constructed an open benchmark from the published conclusions of Ukraine's National Agency on Corruption Prevention (NAZK), comprising 132 draft laws with 474 model-extracted findings. A system flag agrees with a finding when it targets a compatible provision and an LLM judge rates the risk as the same (strict) or partly the same (lenient). On 100 draft laws, the pipeline matches 0.28 of the findings within its top ten flags using the local Gemma 4 31B model, 0.30 with Sonnet 5.5, and 0.43-0.46 across all flags (non-blind; the frozen blind baseline achieves 0.21, 95% CI 0.16-0.26). Across all runs, most misses occur on provisions that the system located but qualified as a different risk, and none of the 32 legal-collision findings was matched. On 21 acts held out from tuning, the pipeline outperforms a monolithic prompt containing the full methodology under the strict measure, though not under the lenient one, and Opus 5.5 adds nothing over Sonnet. The reported scores reflect agreement with model-extracted findings rated by an LLM judge whose two access routes agree on 73 of 74 labels, without human expert validation.

cs.CL · cs.AI