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
+63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 13:44各信源进度 →

统计口径与数据新鲜度

统计 2026-11-15 17:48 — 2026-11-16 17:48(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 13:44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6,921 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 23,827 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:2611.13229

Incorporating World Models in Decision Making for Next Best View Selection in Active Sensing

Nikita Sarawgi, Omey M. Manyar, Satyandra K. Gupta

原文摘要

Efficient robotic 3D reconstruction requires selecting views that contribute new surface measurements with few physical captures. Pretrained world models can guide this choice, but a predicted surface may not produce valid depth measurements under the operating conditions of the deployed sensor. We present a deployment adaptation methodology for next-best-view selection that learns a sensor-return adapter from limited local calibration data while keeping the pretrained predictor fixed. The adapter estimates measurement dropout over each predicted surface, allowing the selection method to favor views likely to fill gaps in the measured reconstruction. In a real-robot experiment with a structured-light sensor, the adapted method achieves higher measured surface coverage with 60% fewer captures than nominal selection. The adapted method recovers a surface region that the nominal method repeatedly targets but fails to measure. These results suggest that robots can reuse pretrained predictive models while adapting their decisions to local operating conditions in data-scarce environments.

cs.RO

新论文 · v1arXiv:2611.13228

Continuous Horizon Forecasting

Malcolm Wolff, Elliott Zackrone, Souhaib Ben Taieb, Michael W. Mahoney, Kin G. Olivares

原文摘要

Multi-horizon forecasting has been dominated by discrete forecast strategies that treat model outputs as fixed-vectors or recurrently create them. In this paper, we introduce the Continuous Horizon (CH) strategy that generalizes upon recursive, direct, and multi-step strategies by redefining forecasts as a continuous operator and the forecast horizon as a query variable rather than an index. We present the theoretical foundations for the CH strategy, a Universal Operator Approximation Theorem, and a continuous-horizon bias-variance decomposition that formally proves the estimation variance improvements of CH over the discrete strategies. We validate the CH strategy empirically through a simulation study, and on 16 large-scale datasets from the M-series benchmark, showing how CH achieves competitive RMSE accuracy, while uniquely enabling parallel inference, targeted horizon querying, and interpolation and extrapolation across unseen horizons.

stat.ML · cs.LG · math.ST

新论文 · v1arXiv:2611.13227

Latent Barrier Steering: Hierarchical Safety for Generative Planning

Renhao Zhang, Mingzhe Li, Bruno Castro da Silva

原文摘要

We introduce Latent Barrier Steering (LBS), a hierarchical semantic-to-path framework for safe flow-matching-based generative planning. Existing control-barrier-function (CBF) safety filters for diffusion and flow-matching planners act locally on generated samples, either during sampling or after prediction. Such repair is effective for small violations, but can struggle when a new constraint blocks the whole selected behavior mode rather than merely perturbing it locally. LBS decouples behavior selection from path certification through two coordinated interventions: it first steers the generator's behavior latent toward rollouts with larger safety margin, then decodes the steered latent and applies a path-space CBF quadratic program (QP) only for residual violations. We show that latent steering increases the margin available to the path-space corrector, yielding finite-time recovery guarantees and reduced correction burden. Experiments across navigation and robot end-effector planning tasks show that LBS preserves safety while improving goal reaching and reducing correction burden compared with path-level safety filters.

cs.RO · cs.CR

新论文 · v1arXiv:2611.13226

Unilluminable rooms: minimal examples, classifications and exceptions

J. C. F. de Winter, George W. Tokarsky

原文摘要

A room with mirrored walls can contain two points that are mutually invisible: no light ray from one reaches the other. Tokarsky built the first polygonal examples by tiling a room with reflected copies of a triangle and making every tile vertex of certain kinds a corner of the room. We study the smallest rooms of this kind, which we call lattice rooms. Their interior tile vertices of the remaining kind, the sources, are invisible and pairwise mutually invisible. On the tiling by isosceles right triangles, a hole-free lattice room with a source has at least 12 sides, attained only by the pinwheel. On the square tiling the least number is 16, again attained by one room. A hole-free lattice room with two sources has at least 24 sides on the first tiling and 34 on the square tiling. Up to congruence there are 12 such rooms with 24 sides, 89 with 25 sides and 14 with 34 sides. One absorbing post, one partition, or one hole that touches the outer boundary at one point lowers the least number of sides for two sources to 18, 20 and 22 (21 if sides are counted as maximal straight segments). Two and three posts give 17 and 16 sides, and a hole that does not touch the outer boundary gives 26. We list all smallest rooms. Rooms assembled from n modules have exactly 12n sides and n sources, and a hole-free lattice room with three or more sources has at least 36 sides. We also give rooms in which some tile vertices need not be corners: 15 types of 27-sided rooms on the tiling by triangles with angles 30, 60 and 90 degrees, which are smallest in their class, a nonagon with three mutually invisible corners, and a 42-sided room in which a rotation through 180 degrees takes the place of a corner. In a ring of three trapezoids a light ray returns to a corner that is invisible in one trapezoid. Several classifications rely on exhaustive computer searches made finite by the bounds proved here.

math.CO · cs.CG

版本更新 · v4arXiv:2410.07787

Learning Beyond Rigid Manipulation with Hybrid Soft-Rigid Robots

Mariano Ramírez Montero, Ebrahim Shahabi, Giovanni Franzese, Jens Kober, Barbara Mazzolai, Cosimo Della Santina

原文摘要

Rigid-soft robot hybrids promise to unite the precision and repeatability of rigid manipulators with the compliance and adaptability of soft structures. Yet, full-fledged rigid-soft manipulators remain largely unexplored, in part because the intelligence required to exploit this combination is still missing. Here, we introduce a hybrid robotic system obtained by the series interconnection of a rigid arm and a soft manipulator, and systematically examine its ability to perform both established and previously inaccessible manipulation tasks. We show that a tailored imitation learning algorithm -- generally applicable to this class of hybrid systems -- leverages the complementary roles of rigid precision and soft embodiment to acquire manipulation skills and generalize them in a single shot across task variations. As a result, the system matches or exceeds the performance of rigid robots in classical tasks such as stacking, while enabling qualitatively new behaviors, including grasping objects through narrow openings.

cs.RO · cs.AI

新论文 · v1arXiv:2611.13224

An Improved $(1+2e)$-Approximation Algorithm for Metric Data Placement

S. Rasoul Etesami

原文摘要

We study the metric data placement problem with site-capacity constraints and fixed installation costs, aiming to minimize the total cost of placing data (resources) and accessing their nearest copies. We develop a deterministic polynomial-time approximation algorithm based on a natural linear programming relaxation of the problem. Starting from an optimal fractional solution, we construct multiscale metric-ball partitions independently for each resource and connect overlapping balls at consecutive levels to form a flow network. The resource layers are then coupled through shared site-capacity arcs, allowing us to obtain an integral placement by minimizing a surrogate cost over a single min-cost circulation. By optimizing the choice of ball radii, we establish an approximation factor of $1+2e\approx 6.44$. To our knowledge, this is the best approximation guarantee for the metric data placement problem since its introduction by Baev and Rajaraman in 2001.

cs.DS · cs.DM · cs.GT · cs.NI · math.OC

版本更新 · v7arXiv:2604.11581

Hidden Measurement Error in LLM Pipelines Distorts Annotation, Evaluation, and Benchmarking

Solomon Messing

原文摘要

LLM evaluations drive model deployment, safety standards, research conclusions, and future projections, for example, about AI's impact on the labor market. Yet the way we often estimate noise from autoraters (LLM-as-judge) ignores variability from factors like judge model choice and prompt phrasing, which means we're underestimating error in a way that worsens with more data. This unmeasured variance can affect results enough to reverse conclusions. And judge-pipelines that fail to average this variation are more vulnerable to "benchmark hacking". This paper introduces a framework we call "total evaluation error (TEE)," which decomposes LLM pipeline uncertainty and uses design-study projections to reduce total error. In our safety and ideology applications, naive standard errors are 53-70% smaller than the corrected SE under our TEE framework. Using Chatbot Arena data, we show naive 95% CI coverage drops as n grows, while TEE-corrected coverage holds at 95%, and TEE-guided pipelines restrict the potential range of benchmark gaming by roughly 43%. We show further that a small pilot allows for the estimation of proper CIs, with the added benefit of projecting which design changes most improve precision. Using those projections to allocate API calls cut MMLU estimation error roughly in half against the answer key at the same budget, and increased per-match agreement with human voters on Chatbot Arena by 7.9 percentage points.

cs.CL

版本更新 · v2arXiv:2502.13297

Understanding and Tackling Label Errors in Individual-Level Nature Language Understanding

Yunpeng Xiao, Kyrie Zhixuan Zhou, Youpeng Zhao, Kai Shu

原文摘要

Natural language understanding (NLU) is a task that enables machines to understand human language. Some tasks, such as stance detection and sentiment analysis, are closely related to individual subjective perspectives, thus termed individual-level NLU. Previously, these tasks are often simplified to text-level NLU tasks, ignoring individual factors. This not only makes inference difficult and unexplainable but often results in a large number of label errors when creating datasets. To address the above limitations, we propose a new NLU annotation guideline based on individual-level factors. Specifically, we incorporate other posts by the same individual and then annotate individual subjective perspectives after considering all individual posts. We use this guideline to expand and re-annotate the stance detection and topic-based sentiment analysis datasets. We find that error rates in the samples were as high as 31.7\% and 23.3\%. We further use large language models to conduct experiments on the re-annotation datasets and find that the large language models perform well on both datasets after adding individual factors. Both GPT-4o and Llama3-70B can achieve an accuracy greater than 87\% on the re-annotation datasets. We also verify the effectiveness of individual factors through ablation studies. We call on future researchers to add individual factors when creating such datasets. Our re-annotation dataset can be found at https://github.com/24yearsoldstudent/Individual-NLU

cs.CL · cs.AI

版本更新 · v2arXiv:2610.17326

Evaluating Long-Horizon Robot Planning under Uncertainty

Harsh Muriki, Gayathri Aravindan, Yuchen Zheng, Sonia Chernova

原文摘要

Household robots must successfully navigate im-perfect circumstances, both in their internal models and the external world. We introduce a benchmark of seven long-horizon household tasks in Habitat that vary initial robot memory and environment configurations. We evaluate adaptations of ReAct, VLM-TAMP, and Tru-POMDP within a shared execution framework across 90 tasks-environment-memory combinations, totaling 1,350 episodes. Our evaluation measures task success, partial completion, and action efficiency, along with rule-based analysis of execution traces. ReAct achieves the highest overall task success (45.3%) and completion (71.7%), while having accurate initial memory does not consistently yield the highest performance for every planner. We identify recurring placement errors, precondition errors, and redundant actions through trace analysis. These findings highlight differences in how the evaluated planners use initial knowledge and respond to uncertainty.

cs.RO

版本更新 · v2arXiv:2506.09195

Graph Attention-based Decentralized Actor-Critic for Dual-Objective Control of Multi-UAV Networks

Haoran Peng, Ying-Jun Angela Zhang

原文摘要

This research focuses on optimizing multi-UAV systems with dual objectives: maximizing service coverage as the primary goal while extending battery lifetime as the secondary objective. We propose a Graph Attention-based Decentralized Actor-Critic (GADC) to optimize the dual objectives. The proposed approach leverages a graph attention network to process UAVs' limited local observation and reduce the dimension of the environment states. Subsequently, an actor-double-critic network is developed to manage dual policies for joint objective optimization. The proposed GADC uses a Kullback-Leibler (KL) divergence factor to balance the tradeoff between coverage performance and battery lifetime in the multi-UAV system. We assess the scalability and efficiency of GADC through comprehensive benchmarking against state-of-the-art methods, considering both theory and experimental aspects. Extensive testing in both ideal settings and NVIDIA Sionna's realistic ray tracing environment demonstrates GADC's superior performance.

eess.SP · cs.AI · cs.MA

版本更新 · v3arXiv:2602.22847

Decentralized Ranking Aggregation via Gossip: Convergence and Robustness

Kerrian Le Caillec, Anna Van Elst, Igor Colin, Stephan Clémençon

原文摘要

The concept of ranking aggregation plays a central role in preference analysis, and numerous algorithms for calculating median rankings, often originating in social choice theory, have been documented in the literature, offering theoretical guarantees in a centralized setting, \textit{i.e.}, when all the ranking data to be aggregated can be brought together in a single computing unit. For many technologies (\textit{e.g.} peer-to-peer networks, IoT, multi-agent systems), extending the ability to calculate consensus rankings with guarantees of convergence and resilience to potential contamination in a decentralized setting, when preference data is initially distributed across a communicating network, remains a major methodological challenge. Indeed, in recent years, the literature on decentralized computation has mainly focused on computing or optimizing statistics such as arithmetic means using gossip algorithms. The purpose of this article is precisely to study how to achieve reliable and resilient consensus on collective rankings in a decentralized setting, thereby raising new questions, robustness to corrupted nodes, and scalability through reduced communication costs in particular. The approach proposed and analyzed here relies on the robustness guarantees offered by random gossip communication, which allows autonomous agents to compute a global ranking consensus using local interactions only, without coordination or a central authority.

cs.LG · cs.AI · stat.ML

新论文 · v1arXiv:2611.13218

Wrong the Same Way: Cross-Prompt Sharing Both Suppresses and Amplifies Reward Hacking in RLVR

Michael Li, Rohan Pandey, Ray Chen

原文摘要

Two accounts of how reinforcement learning with verifiable rewards (RLVR) interacts with an imperfect verifier coexist: in one, reward hacking is inherited from the base model; in the other, RL amplifies it. We study both in a controlled environment small enough to compute the entire response distribution exactly. We vary the universality of a leak, or the fraction of prompts that accept the same wrong response, and the verifier's static false-positive rate. At its KL-regularized optimum, RLVR acts as a filter that reweights the accepted set without altering its composition, so the hack rate is inherited exactly. Taking that as a null model, a known decomposition writes the drift from it as a sum over prompts of covariances between reward and the empirical tangent kernel. The terms from other prompts vanish without parameter sharing. With sharing, every prompt that accepts a wrong answer pushes it up and every prompt that rejects it pushes it down through the baseline, in proportion to the model's probability of producing it there. A shared-parameter policy thus drives a leak that few prompts accept below the base rate and amplifies one that most prompts accept. At every convergent learning rate under Adam, SGD, and sampled GRPO, hacking ends higher at high universality than at low, but whether either end crosses the base rate depends on the step size. Independent per-prompt copies remove the dependence. The direction of this push depends on neither the verifier's error rate nor universality alone, but on how much of the model's kernel-weighted probability of a wrong answer falls on prompts that accept it, which frozen base model controls support. Verifier fixes should target the mistakes the model repeats across prompts: two patches that remove equally many errors differ by up to 38 points of true accuracy. We propose this alignment as a hypothesis for why earlier studies disagree.

cs.LG · cs.AI

新论文 · v1arXiv:2611.13216

Latin American Studies since the 1980s: From Apocalyptic Dreams to Assimilationist Consolidation

Thiago Dumont Oliveira, Juan Grigera, Pedro Mendes Loureiro

原文摘要

We investigate how Latin American Studies (LAS) has evolved since 1982 in three leading LAS journals. Using network analysis, text mining, and bibliometrics, we trace transformations in disciplinary positioning, authorship, and citation patterns. Our main claim is that LAS has undergone an assimilationist consolidation. The field is now institutionally recognisable, with core LAS journals occupying central positions in disciplinary networks and acting as bridges across fields. Yet this consolidation has reinforced hierarchies privileging Anglophone academic norms. Journals based in Latin America, once central, have been marginalised. The share of Latin American authors has grown, but their profiles converge towards Northern standards.

cs.DL · cs.CL

版本更新 · v2arXiv:2611.08100

An Integrated Open-Source Stack for Vision-Based Autonomous Navigation of Wheeled Rovers

Iaroslav Kolomiets, Iaroslav Sheipak, Mikhail Reppo, Vsevolod Egorov, Viktor Chertopolokhov, Isaac Chairez

原文摘要

Vision-based rover autonomy requires integrating perception, localization, planning and platform-specific control under limited sensing and computing resources. We present an open-source ROS 2 stack for following directional arrows, reaching a terminal cone and returning autonomously to the start. It connects YOLOv8n detection and geometric direction, range and bearing estimation with RGB-D SLAM, wheel-inertial state estimation, kinematics-aware planning, behavior-tree mission management and STM32 actuation. Standard interfaces separate reusable autonomy from the hardware layer; a proportional path-following plugin supports low-speed rover motion. Public resources include software, configuration, a labeled dataset, trained weights and evaluation logs. On 478 held-out images, the detector achieves mAP@0.5 of 0.720 under the Ultralytics evaluator and outperforms a classical OpenCV baseline under a shared evaluation protocol. The reported perception latency is 25 ms on a Jetson Orin NX. Three missions, one indoor and two outdoor, demonstrate marker-guided navigation and autonomous return without operator motion commands or reported collisions. Field results document integration failures and corrective choices; repeated instrumented trials remain necessary to quantify physical navigation errors and reliability.

cs.RO

新论文 · v1arXiv:2611.13215

Counting Under Disagreement: Dollar-Weighted Measurement, Selective Machine Labelling and Budgeted Human Review from Disagreeing AI Readers of Administrative Text

George Koomullil

原文摘要

We study AI-assisted measurement when one text label applies to many administrative records, the target is a monetary total, and the human readings that anchor the machines disagree. The testbed is 121,070 grant-purpose texts from US tax filings (799,675 records, \$126.4B), labelled by a rule-based classifier and three language models, with a designed sample of 651 texts read independently by three people under one codebook. We state exact conditions for a fixed-predictor difference estimator and for a cross-fitted predictor whose folds are assigned to the whole frame before sampling, under Poisson sampling, and assess the implemented fixed-size procedures by simulation: 8 predictors whose own totals span \$11.2-15.0B agree to within \$0.4B once rectified. Two rules that accept machine labels at a stated dollar-weighted error, weighted conformal p-values under a dollar measure and Learn-then-Test on the design estimate of the error, are evaluated on nested held-out folds and on re-draws of the design with synthetic truths; their error control in deployment is not established, and the count-based rule accepts no text under a dollar-drawn sample. In a replay that charges every human label, a two-phase review policy whose calibrator is fitted on its own first phase reduces the RMSE of the total by 28% at a budget of 100 texts relative to dollar-proportional sampling, while pure Neyman allocation fails with an uncalibrated predictor. Independent readers shift the estimated total by \$3.93B, 2.2 times the within-reader sampling interval, and institution-level intervals under-cover at every size. Every headline result is re-run under each reader. The results identify both the benefits and the limits of using learned predictions to decide what a person should verify.

cs.AI · cs.LG · stat.ME

新论文 · v1arXiv:2611.13214

Collision Avoidance and Collision Detection Strategies to Counteract Large-Scale FMCW Automotive Radar Interference

Francesco Miccoli, Vittorio Todisco, Fabrizio Cuccoli, Matteo Passeri, Vincent Martinez, Alessandro Bazzi

原文摘要

Automotive radars operating in dense traffic environments may experience significant mutual interference, potentially degrading sensing reliability for advanced driver assistance systems (ADAS). This issue motivates the development of techniques to mitigate interference among multiple radar-equipped vehicles sharing the same spectrum. This paper investigates interference mitigation strategies for uncoordinated radar networks through system-level simulation. Two approaches are considered: frame-by-frame collision avoidance (FbF-CA), which exploits idle intervals to measure the perceived interference and possibly adapt the operating frequency, and frame-by-frame collision detection (FbF-CD), which estimates interference during transmission based on observed chirp collisions. An event-driven radar network simulator is used to model realistic highway traffic and radar interactions. Results show that collision avoidance provides higher robustness when a larger bandwidth is available, while collision detection is more effective under bandwidth-limited conditions. The analysis highlights the trade-off between collision avoidance and collision detection mitigation strategies and provides design insights for scalable automotive radar deployments.

cs.NI

新论文 · v1arXiv:2611.13213

Before Words, Beyond Speech: Evaluating Nonverbal Social Reasoning in Early Childhood

Marie Amale Huynh, Laura Bravo-Sánchez, Lauren Klein, Nick Haber, Philip Fisher, Serena Yeung-Levy

原文摘要

Children's earliest years are a period of intense social learning, during which meaning is built through gaze, gesture, and physical contact. Yet, this critical aspect of social interaction remains under-evaluated in multimodal video understanding, as existing benchmarks prioritize the conversation-driven interactions of adults. We address this gap with \bench ($\textbf{N}$aturalistic $\textbf{E}$arly-childhood $\textbf{S}$ocial in$\textbf{T}$eractions), the first child-centric benchmark for nonverbal social reasoning. \bench comprises \totalvids manually annotated clips ($\sim$12s) curated for diversity across naturalistic settings, cultural contexts, and developmental stages. Beyond a lack of data, targeted model improvement in this domain is hindered by coarse evaluation schemas that conflate basic perception with higher-order reasoning. To resolve this, \bench introduces a compositional schema that builds atomic interaction cues into progressive levels of social reasoning: contextual, behavioral, interpersonal, and open-ended descriptions. Extensive analysis of state-of-the-art Vision-Language Models (VLMs) reveals a stark performance collapse: despite strong contextual recognition, models exhibit substantial gaps in decoding precise behavioral and interpersonal cues. We trace these failures to specific biases, including poor consequence grounding and temporal sensitivity. Finally, we ground \bench in real-world applications by evaluating VLMs using domain adaptation strategies. Ultimately, \bench provides a rigorous testbed for advancing socially grounded artificial intelligence in early childhood. Project page: https://laubravo.github.io/nest_website/.

cs.CV · cs.AI · cs.LG

新论文 · v1arXiv:2611.13211

Designing UIs with GenAI: Understanding the Use, Benefits, and Limitations of Prompt-to-Design Generators

Jingzhou Ye, Yao Li, Xueqiang Wang

原文摘要

The integration of generative AI (GenAI) into user interface (UI) design has led to prompt-to-design generators (P2dGens), tools that enable designers to create UI prototypes from textual prompts efficiently. Despite their growing adoption, little is known about how designers use these tools in everyday practice, their benefits, and limitations. We conducted semi-structured interviews with 17 UI designers to investigate P2dGens use, focusing on workflow integration, attitudes toward generator-assisted design, and the organizational climate. Findings show that designers primarily use P2dGens as ideation accelerators rather than end-to-end solutions, and they adopt a variety of individual and collaborative prompting strategies. Designers expressed complex attitudes, noting gains in efficiency and inspiration while also highlighting concerns about loss of personal creativity and design quality. We also observed emerging governance practices and varying attitudes within organizations toward P2dGens. These insights illuminate how P2dGens reshape UI design workflows and contribute to understanding human-AI collaboration in UI design.

cs.HC

新论文 · v1arXiv:2611.13210

Near-Optimal Sample Complexity of Deterministic Agnostic PAC Learning

Shai Ben-David, Steve Hanneke, Farnam Mansouri, Amirreza Shaeiri

原文摘要

We study the problem of agnostic PAC learning under deterministic labels, where the examples are labeled by an arbitrary deterministic function, which need not belong to the concept class $C$. This model captures misspecification without stochastic label noise and lies between the realizable and fully agnostic settings. Prior work showed that, when $C$ has finite diameter $K$ and VC dimension $d$, deterministic labels can improve the sample complexity upper bound from the classical agnostic guarantee $O(d/\varepsilon^2)$ to $\widetilde{O}(K/\varepsilon)$, but left a large gap between upper and lower bounds. Our main result sharpens this picture. We prove that every class of VC dimension $d$ and finite diameter $K$ is learnable with sample complexity $\widetilde{O}\left( \frac{K^{1/3} \, d^{2/3}}{\varepsilon} \right)$, and we construct classes for which this bound is tight up to logarithmic factors. We also study the online analogue of this problem and show that the minimax regret is again partly controlled by the diameter of the class. Technically, our upper bounds are obtained via memorization-based learning algorithms that exploit the deterministic label structure.

stat.ML · cs.LG

版本更新 · v3arXiv:2604.10819

Differentially Private Verification of Distribution Properties

Elbert Du, Cynthia Dwork, Pranay Tankala, Linjun Zhang

原文摘要

A recent line of work initiated by Chiesa and Gur and further developed by Herman and Rothblum investigates the sample and communication complexity of verifying properties of distributions with the assistance of a powerful, knowledgeable, but untrusted prover. In this work, we initiate the study of differentially private distribution property verification. After all, if we do not trust the prover to help us with verification, why should we trust it with our sensitive sample? We map a landscape of differentially private verification of properties of distributions. In the non-private case it is known that one-round private-coin protocols can have substantially lower complexity than public-coin (AM) protocols. In contrast, the possibility for improvement in differentially private interactive proofs depends on the privacy parameter regime and model. Drawing on connections between privacy and replicability and privacy amplification techniques in the literature we show: 1. There exists a reduction from any one-round $(\varepsilon,δ)$-differentially private private-coin protocol to a differentially private AM protocol for the parameter regime $\varepsilon = O(1/\sqrt{s})$ and $δ= O(1/s^{5/2})$ with approximately the same privacy and sample and communication complexities. In the local model, this is relaxed to $\varepsilon = O(1/\sqrt{\log s})$ 2. However, when the privacy guarantee is very relaxed ($\varepsilon \in Ω(\log s)$), private coins indeed reduce sample and communication complexities. We also obtain a computationally efficient Merlin-Arthur proof for privately testing whether samples are drawn from a product distribution and prove that its sample complexity is optimal up to a $polylog N$ factor by reducing uniformity testing to independence testing with Boolean attributes and appealing to known lower bounds on sample complexity for private uniformity testing.

cs.DS · cs.CC · cs.LG

版本更新 · v6arXiv:2511.15238

A Quantitative Approach to Reach-Avoid Reachability Analysis for Discrete-Time Nonlinear Control Systems

Ihab Tabbara, Eliya Badr, Hussein Sibai

原文摘要

A reach-avoid specification requires a dynamical system to reach a target set of states while avoiding a failure set. Reachability analysis partitions the state space into parts starting from which the specification can be satisfied by following some control signal and those from which it cannot despite the best control effort. In this paper, we present a quantitative approach for reachability analysis to analyze the satisfaction of avoid and reach-avoid specifications for discrete-time nonlinear control systems. In such an approach, a value function assigns a value to each state that determines how robustly the system satisfies or violates the specification starting from it and following the optimal control signal. However, computing such a value function is difficult. Instead, our approach constructs discrete abstractions on which it performs value iteration to compute sound bounds on the desired value function. The resulting lower bound enables a sound under-approximation of the reach-avoid set (RAS) and, in the avoid case, a sound over-approximation of the backward reachable set (BRS). When the control set is finite, the upper bound additionally enables a sound over-approximation of the RAS and, in the avoid case, a sound under-approximation of the BRS. Moreover, the obtained bounds provide quantitative guarantees on safety margins and, in discounted reach-avoid problems, on safe target-reaching times. When the control set is finite, they also bound failure-reaching times. Our approach further refines the abstraction by splitting the grid cells that could not be classified as within or outside the BRS (or RAS) to improve the bounds. We evaluate our approach on five systems with up to five-dimensional state spaces.

eess.SY · cs.FL · cs.SC

新论文 · v1arXiv:2611.13209

ARoute: Task-Aligned Batch and Streaming Model Routing

Daofu Zhang, Qingcheng Zeng, Chunwei Liu, Fuheng Zhao

原文摘要

Language models increasingly power data-intensive applications, from semantic data processing to question answering and agentic workflows. Using frontier models for every request is prohibitively expensive, while relying only on smaller models reduces accuracy. We study model routing under a user-specified cost budget, aiming to maximize accuracy under task-specific correctness criteria. Existing routers fall short in two ways: pretrained routers rely on generic preference signals that may not reflect a task's own correctness criteria or a new model pair, and model cascades run the weaker model first on every request, adding sequential latency. We present ARoute, a task-aligned router that learns cost and accuracy signals to select a model for each request before inference. ARoute has two variants: ARoute-Batch operates on a batch of requests known in advance, routing requests with the greatest expected benefit to the stronger model subject to the budget, whereas ARoute-Stream processes requests online, using a threshold calibrated on a small validation set to satisfy the cost quota and dynamically adjusting it as the budget is consumed. For ARoute-Stream, we provide a finite-sample guarantee on expected normalized spending. We evaluate ARoute on MMLU-Pro, LiveCodeBench, SuperGPQA, and two value-filling tasks built from SWAN semantic query benchmark against other state-of-the-art approaches. At matched cost, ARoute-Batch achieves the highest mean accuracy in all 15 task-cost settings, improving over the strongest baseline by 4.1 percentage points on average and by up to 11.9 points. ARoute also completes a batch up to 2.8x faster than cascades.

cs.DB

新论文 · v1arXiv:2611.13207

HiCAP: Hierarchical Cultural Adaptation via Personality for Dialogue Generation

Lingzhi Shen, Xiaohao Cai, Yunfei Long, Imran Razzak, Shoaib Jameel

原文摘要

Cultural adaptation in dialogue systems refers to the ability to generate responses that conform to linguistic norms and social expectations across different cultures. However, existing dialogue generation approaches pay limited attention to the role of personality in shaping cultural expression and communication style. We therefore propose HiCAP, a novel framework that integrates personality traits into culturally adaptive dialogue generation. HiCAP employs a personality-conditioned control mechanism that hierarchically modulates response style by first aligning it with the cultural tone and then refining linguistic expression and content according to individual personality attributes. To embed these factors, the model combines supervised fine-tuning and instruction-based tuning, supported by dynamic instruction templates and multi-agent demonstrations to capture diverse combinations of culture and personality. Experiments on CulturalPersonas and CPED show that HiCAP achieves the strongest overall performance across various dialogue generation metrics, demonstrating effective cultural adaptability and personality modelling.

cs.CL · cs.LG

版本更新 · v2arXiv:2610.37534

Computing GR(1) Assume-Guarantee Contracts

Yigit Narter, Emily Zhao, Inigo Incer

原文摘要

Assume-guarantee contracts provide a set of algebraic operations that enable the efficient compositional design and analysis of complex engineering systems. The explicit computation of these operations at component interfaces is currently supported only for some logics lacking a notion of time. This paper develops theory and algorithms to compute these operations when the contracts are expressed in the GR(1) modulo theory. The key challenge is the computation of two quantifier elimination problems for temporal formulas. We develop context-aided variable elimination procedures on GR(1)-modulo formulas, establish their soundness, characterize subproblem-wise optimal solutions within an admissible class of formulas, and use the procedures to compute contract operations, including composition and quotient. We implement the resulting framework in Pacti, automating compositional reasoning over GR(1) contracts. Finally, we demonstrate the framework on an automated emergency-braking case study, where we compute system-level contracts and expose compatibility conditions arising from the interaction of component specifications. Experiments on synthetic families show that the cost is governed by the number of eliminated variables and liveness conjuncts, and a comparison with exact automata-based elimination shows complementary scaling.

cs.LO · eess.SY

新论文 · v1arXiv:2611.13204

Ground Reaction Force-Aware Model Predictive Control for Underactuated Upright Physical Assistance

Darren Biskup, Jeonghan Yu, Zibo Yu, Sunil K. Agrawal

原文摘要

Many rehabilitation robots execute prescribed motions using reactive tracking controllers that do not explicitly incorporate the user's instantaneous human-ground interaction. This paper presents Ground Reaction Force-Aware Model Predictive Control (GRF-MPC), a framework for upright physical assistance. By integrating live GRF and center of pressure (CoP) measurements directly into a human centroidal dynamics model, GRF-MPC formulates a real-time quadratic program (QP) that optimizes assistive forces and moments over a receding horizon. We deployed GRF-MPC on the Robotic Upright Stand Trainer (RobUST) in an underactuated 4-cable configuration and evaluated it against a fully actuated 8-cable Cartesian impedance baseline in nearly 300 trials across six healthy participants. Across tasks spanning impulsive perturbations, lunge trajectories, and voluntary limits of stability, GRF-MPC reduced peak planar center-of-mass (CoM) error by 19.6% during perturbations, lowered trajectory tracking error by 24.1%, required 28.7% less net mechanical work, and permitted a 45.9% greater normalized voluntary excursion area. The underlying QP solves in under 1 ms, comfortably enabling 100 Hz replanning and demonstrating that incorporating live human-ground interaction into predictive control yields stronger, more transparent physical assistance with half the actuators.

cs.RO · eess.SY

新论文 · v1arXiv:2611.13203

Audited Gradients, Anchored Policies: Model-Based Reinforcement Learning for Efficient Flow Control

Ruige Kong, Stefania Fresca

原文摘要

Feedback controllers for flows are expensive to design, since each candidate must be evaluated with a high-fidelity solver. Differentiable surrogates reduce that cost, but they also introduce a failure mode that prediction error cannot detect. Policy optimization uses the derivative of the objective with respect to the action, which we call the action gradient and which can be computed either through the surrogate or through the high-fidelity solver. Optimization through the surrogate uses only its action gradient, and a surrogate can follow the flow trajectory closely while misestimating this derivative. The surrogate action gradient is compared against a central finite-difference estimate from the high-fidelity solver on held-out states, under acceptance criteria fixed before the comparison, and we call that comparison an audit. A low-dimensional feedback law is then fitted to the surrogate action gradient when the surrogate passes the audit, and to the solver estimate when it does not. This law initializes a neural policy, and a quadratic penalty holds the policy near it during optimization. The proposed approach is validated on four test cases. The resulting controllers outperform or match a model-free baseline, while using 3 to 156 times fewer high-fidelity solver interactions across the four test cases. One surrogate predicts the objective at correlation 0.9996 but its action gradient at only 0.459, so prediction accuracy does not identify a usable surrogate. Another surrogate, whose action gradient agrees with the solver estimate at correlation 0.996, makes the objective 0.52% worse when the policy is optimized through it without the anchor penalty, so an accurate gradient alone does not provide a working controller. Finally, a surrogate that passes the audit at one fixed action fails it around the actions the deployed controller applies, so a pass is local.

cs.LG

新论文 · v1arXiv:2611.13200

Image Quality and the Reliability of Fabric-Density Measurement: Counting Threads from Smartphone Photos Through a Pick Glass

Anoubhav Agarwaal

原文摘要

Thread density, the threads per inch along and across a woven fabric, is a basic specification that buyers check whenever fabric is sold. Without laboratory instruments, inspectors count it slowly, by eye, through a pick glass, a folding 10$\times$ magnifier with a one-inch window. A smartphone photo through that glass could make the check faster and leave a record. We judge such a photo by what it is used for, as in task-based image quality: it is good if the thread count from it is within $\pm$5% of the true count, a common tolerance in fabric specifications. We introduce PickGlass-TPI: 76 factory scenes of 44 fabric patches, each counted in an open annotation tool by a textile-trained annotator, whose counts are the reference. On it we benchmark frontier and open vision-language models, CNNs, counting specialists and a training-free spectral counter (PG-Lattice). No method agrees with the reference within the tolerance. The best methods are off by about a quarter: a small open model that we tuned only on synthetic images (Molmo2-4B, 25%) is level with the best frontier model, GPT-6.1-sol (27% when asked directly, 25% given four magnified close-ups of the window). Both err less than a non-expert annotator (37%), but only the tuned model also beats, on both error measures, a baseline that guesses each scene's density as the average of the others. PG-Lattice most often repeats its count within 5% across two photos of the same patch, but its misses, like GPT-6.1-sol's errors, are often a factor of two. Even the reference repeats a count across such re-shoots only to a median 9%, so one photo cannot yet be shown to meet $\pm$5%; the image quality that matters is reliability across re-shoots. Generic no-reference quality scores chosen before labeling do not rank PG-Lattice's error, and generative enhancement raises perceptual scores but does not reliably improve the count.

cs.CV · eess.IV

新论文 · v1arXiv:2611.13199

XAI-Driven Confidence Estimation for OOD-Resilient Runtime Prediction

Mahmoud Sanad, Mohammed Bakr Sikal, Osama Abboud, Heba Khdr, Jörg Henkel

原文摘要

Machine learning-based resource management (RM) models degrade under out-of-distribution (OOD) inputs at runtime, yet existing approaches in the literature rely on a single global error margin that is overly conservative and fails to distinguish reliablepredictions from catastrophic failures. We propose Lightweight Runtime EXplainability (LEX), the first post-hoc XAI framework for input-specific runtime confidence estimation in ML-based RM. LEX clusters the embeddings of the training input space using a Gaussian Mixture Model and associates per-cluster prediction error bounds to input samples, enabling fine-grained runtime confidence signals without retraining. Evaluated on a state-of-the-art LSTM-based IPS forecasting model on Intel Alder Lake, across 34 benchmark applications, LEX achieves ~94% containment on unseen workloads and produces ~17% tighter per-input confidence bounds than global baselines. With its negligible runtime overhead, LEX advances the field by providing explainable and reliable input-specific confidence signals of ML predictions, thereby making ML-based resource management resilient to OOD at runtime.

cs.LG · cs.AI

新论文 · v1arXiv:2611.13198

Structured Kernel Surrogate Models for Multimodal Attribution

Ziniu Zhang, Zhenshuo Zhang, Jianglin Lu, Ruoxuan Xiong, Yun Fu, Hongyang R. Zhang

原文摘要

We study the problem of cross-sample modality attribution, which involves measuring how one modality in one sample affects the learning outcome in other multimodal samples. This problem is crucial in a multimodal system because a data modality (such as a text description) may be redundant for one sample (a self-explanatory image), but the same data modality would otherwise be uniquely informative for another sample, or informative when combined with other modalities. Prior work focuses on influence estimation at the level of one sample, and it remains an open question how one should capture the interactions across both samples and modalities. To capture scenarios like the above, we propose structured kernel surrogate models, which are surrogate mappings from both samples and modalities to the prediction outcome when they are trained together. We design an algorithm, SKS, which applies kernel ridge regression with a structured, compositional kernel on the selection states of individual samples and modalities. Specifically, the kernel function multiplies kernel-based similarity scores between every pair of data modalities for one sample and a pair of two samples. Additionally, we design a gradient-based estimation algorithm to efficiently estimate the structured kernel surrogate model. We evaluate our approach on two multimodal benchmarks, spanning a range of data modalities including satellite images, road networks, text, audio, and tabular data. We find that SKS achieves $57.8$% higher correlation with the true prediction outcomes than existing methods. In downstream data selection, SKS improves the average accuracy by $1.93$% over baselines. Models with data selected by SKS have lower Hessian traces, indicating a flatter loss surface. Using SKS to select satellite images identifies road segments, where accident scenes commonly occur, as opposed to background images.

cs.LG · cs.AI · cs.CV

版本更新 · v2arXiv:2608.28618

Predicting Student Attrition in Competitive Programming: A Large-Scale Study Integrating Survey Insights and Global Behavioral Logs

Azuad Islam Ruhan, Md. Rejaul Korim Sadi, Rakibul Islam Rafi, Dewan Fahad Chowdhury, Nazira Bani Opi, Sherin Afrin Mim, Golam Mostofa Naeem

原文摘要

We compared the recorded behavior of competitive programmers who became inactive on Codeforces with that of active users of similar skill. From 6,927 accounts collected through the public Codeforces application programming interface in June 2026, we removed 888 accounts whose retrieved submission histories were incomplete and restricted both groups to a peak rating below 1500. This left 3,421 active accounts and 908 accounts with no submission for at least three months, labeled attrition. The differences between the groups were small. Accounts labeled attrition accepted smaller shares of their attempts, overall and on mathematics and data-structure problems (rank-biserial r from -0.16 to -0.12), and solved fewer problems per rated contest (median 5.95 versus 7.80; r = -0.13); struggle time, upsolving and recovery after rating drops did not differ after correction for multiple testing. Two pre-specified models using eight features that do not depend on the length of the observation window reached cross-validated areas under the receiver operating characteristic curve of 0.596 (logistic regression) and 0.611 (random forest), with probability estimates barely better than the base rate. Across random samples, stricter cleaning rules, other definitions of inactivity and matching on account tenure, the value ranged from 0.58 to 0.66. Because the features summarize each user's whole recorded history, these results describe differences between groups and do not show that inactivity can be predicted for individual users. The data and code are publicly archived, and one command regenerates every reported number.

cs.CY