跳到正文
研踪PaperTrail

YOUR RESEARCH, IN SIGHT

让下一篇重要论文，不再错过。

覆盖计算机全学科，以 AI 为快捷入口。关注方向、作者和关键词，把发现变成持续的研究积累。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+0新增论文
+0其中 AI 相关
+2已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-08 08:12 · 全站最近入库：2026-11-08 06:02各信源进度 →

统计口径与数据新鲜度

统计 2026-11-07 08:16 — 2026-11-08 08:16（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：2026-11-08 08:12最近发现新增版本：2026-11-08 06:02库内论文最新更新时间：2026-11-06 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。

17,100 已收录论文14,661 AI 相关4,057 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

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 17,100 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-08 08:11（北京时间）
版本更新 · v2arXiv:2610.36460

Full Collapse Is Suboptimal: Controlled Partial Neural Collapse for Long-Tailed Recognition

Hyeontaek Hwang, Cheonghwan Hur, Daeyoung Kim

原文摘要

Recent long-tailed recognition (LTR) methods incorporate neural collapse (NC) and align features to a fixed Equiangular Tight Frame (ETF) as a uniform target for all classes. However, forcing tail classes to fully collapse over-contracts their samples and weakens generalization. To address this, we propose a controlled partial neural collapse, realized through two coordinated controls on a backbone: the Class-Adaptive Sparse Activation (CASA) classifier and the ETF-constrained adaptive margin (ETF-AM) loss. Rather than forcing a uniform full collapse, this regime allows head classes to collapse along the ETF while tail classes retain within-class diversity. This prevents the tail from being forced into full collapse, which is suboptimal under imbalance. To set the dimensionality of collapse, CASA induces class-related node specialization through feature-conditioned gating so that tail classes occupy their own feature directions. For the clearance axis, ETF-AM applies a frequency-adaptive margin, with smaller margins for head classes and larger for tail classes. On CIFAR100-LT, ImageNet-LT, Places-LT, and iNaturalist-2018, our method achieves the best overall accuracy over prior methods. In the empirical analysis, within-class spread increases monotonically from head to tail classes, consistent with tail classes retaining variability rather than fully collapsing.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2610.11152

Imitation Dominates Reinforcement: Direct In-Context RL Is Closer to ICL Than RL

Minchan Kwon, Seunghee Koh, Sunghyun Baek, Minsung Bae, Junmo Kim

原文摘要

LLM agents increasingly improve at inference time by accumulating experience in context rather than by updating parameters. This process is often described as in-context reinforcement learning (ICRL). Whether in-context learning (ICL) can actually play the role of RL, however, has not been tested. We study this question in its simplest form, direct ICRL, where the model conditions directly on raw trajectory-reward pairs, and ask whether the reward acts as a learning signal. Through controlled experiments on four benchmarks across six models, we find that the reward is read, but its effect is small: flipping, randomizing, or removing the reward leaves the improvement curve almost unchanged, and this holds even under meta-prompts that explicitly instruct the model to explore, exploit, or reason over rewards. Trajectories drive improvement, but not through their semantic content: shuffled or corrupted trajectories work as well as real ones. These patterns closely mirror those known in ICL, suggesting that direct ICRL is better understood as a special case of ICL than as inference-time RL. This reframing has implications for agent memory design: ICL factors such as input distribution and demonstrations may matter more than RL elements such as reward shaping and exploration.

cs.LG · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04123

An automated photometric pipeline for measuring the corotation radius of spiral galaxies

Kimia Yazdani, Sahel Mesforoush, Wayne B. Hayes

原文摘要

The corotation radius R_CR of a spiral galaxy marks the radius where the angular speed of the spiral pattern matches the orbital speed of the disk material. Despite its importance for understanding spiral structure and galaxy evolution, very few galaxies have corotation measurements, often derived using heterogeneous techniques. We present an automated photometric pipeline that identifies corotation radii by tracking the azimuthal phase offset between young and old stellar populations. Building on the color-gradient framework of \citet{Puerari1997}, we extract azimuthal brightness profiles from deprojected dual-band images, fit periodic models to the profiles, and identify corotation candidates at radii where the phase difference between the two stellar tracers changes sign. Each candidate is assigned an adaptive uncertainty interval and reported with a goodness-of-fit metric. We benchmark the method against a cross-method literature compilation, scoring every technique -- including our own -- against a reference built from the published measurements alone and evaluated leave-one-out. Over the $90$ galaxies for which such a reference exists, the pipeline places a correct corotation radius in $80.0\%$ of them, recovers $63.6\%$ of their known corotation radii, and reaches a harmonic mean of recall and purity of $0.58$, the highest values of any method in the comparison. Unlike the published methods, each of which was applied to a small, curated set of galaxies, our pipeline is run on the entire reference sample without any pre-selection, a sample two to six times larger than existing methods. The pipeline is scalable to large surveys, flexible with respect to spiral mode, and naturally recovers multiple corotation radii in galaxies with complex spiral structure, providing a uniform framework for studying disk resonances across diverse galaxy populations.

astro-ph.GA · cs.CV · physics.data-an

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04122

ScrambleMark: Zero-Query Watermark-Agnostic Black-Box Evasion of Modern Speech Watermarking

Binhao Ma, Weikang Ding, Hanqing Guo, Wenwei Zhao, Xiaowen Li, Leah Ding, Zhuo Lu, Rui Duan

原文摘要

Speech watermarking is becoming a key defense for digital copyright and content authentication. Existing attacks treat evasion as a removal problem: they erase the watermark, relying on detector access or on speech data that is hard to collect, such as paired clean-watermarked or clean target speaker speech. Commercial platforms are moving to close off these assumptions. TrustedAudio, for instance, restricts detection to authenticated users and denies meaningful detection feedback to unauthorized parties. Adversaries thus cannot tell whether a sample is watermarked or which watermark it carries, nor reliably remove it without degrading speech quality. In this work, we show that evading a speech watermark does not require erasing it: detection relies on the structural consistency of the embedded pattern, and disrupting this consistency is enough to defeat it. Building on this observation, we identify harmonic structure as an effective attack surface, as perturbing it breaks watermark consistency while preserving perceptual quality. We therefore propose ScrambleMark, a zero-query, watermark-agnostic, black-box attack that evades detection through harmonic scrambling. ScrambleMark combines a deterministic engine for discovering transferable harmonic perturbations with a generative model that learns randomized scrambling patterns, avoiding reliance on fixed attack rules. Across three commercial platforms (Google SynthID, Resemble AI, and TrustedAudio), ScrambleMark is the only attack to both evade detection and keep speech natural, achieving nearly 97% evasion and a 4.2/5 human rating, far above the 1.1 to 1.7 of prior attacks. Beyond evasion, ScrambleMark withstands adaptive recovery designed to reverse it, evidence that structural evasion is hard to mitigate in practice.

cs.CR · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2602.06400

TFusionOcc: T-Primitive Based Object-Centric Multi-Sensor Fusion Framework for 3D Occupancy Prediction

Zhenxing Ming, Yaoqi Huang, Julie Stephany Berrio, Mao Shan, Stewart Worrall

原文摘要

The prediction of 3D semantic occupancy enables autonomous vehicles (AVs) to perceive the fine-grained geometric and semantic scene structure for safe navigation and decision-making. Existing methods mainly rely on either voxel-based representations, which incur redundant computation over empty regions, or object-centric Gaussian primitives, whose ellipsoidal per-primitive geometry can limit representation efficiency for complex, non-convex, and asymmetric structures under a fixed primitive and computational budget. In this paper, we present TFusionOcc, a T-primitive-based object-centric multi-sensor fusion framework for 3D semantic occupancy prediction. Specifically, we introduce a family of Student's t-distribution-based T-primitives, including the plain T-primitive, T-Superquadric, and deformable T-Superquadric with inverse warping, where the deformable T-Superquadric serves as the key geometry-enhancing primitive. We further develop a unified probabilistic formulation based on the Student's t-distribution and the T-mixture model (TMM) to jointly model occupancy and semantics, and design a tightly coupled multi-stage fusion architecture to integrate camera and LiDAR cues effectively. Extensive experiments on nuScenes demonstrate competitive performance against recent state-of-the-art methods and strong performance among object-centric multi-sensor approaches, while additional evaluations on nuScenes-C show robust performance across most corruption scenarios. Further evaluation on the separate SemanticKITTI benchmark demonstrates cross-dataset applicability. The code will be available at: https://github.com/DanielMing123/TFusionOcc

cs.CV · cs.AI · cs.RO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04116

Decomposing Rank Reversals under Temporal Nonstationarity: Alignment, Aggressiveness, and Instability

Deying Song, Xueying Ding

原文摘要

Under temporal nonstationarity, model performance rankings reverse across training horizons due to heterogeneous sensitivities to memory. While existing literature addresses adaptive lookbacks, effective model complexity, and empirical forecast evaluation in isolation, no unified framework explains the structural mechanics driving these rank reversals. We bridge this gap by deriving a model-agnostic decomposition of out-of-sample $R^2$ into target alignment, miscalibrated aggressiveness, and predictive instability. Within this framework, we mathematically characterize complexity phase boundaries, establish closed-form optimal-memory equations for ridge regression under local drift, and extend these guarantees to complex model architectures via surrogate bounds. We validate these theoretical dynamics by benchmarking 14 linear, tree-based, and neural architectures across large-scale financial panels. In doing so, our framework demonstrates that empirical model rank reversals are driven by predictable surges in miscalibrated aggressiveness, providing a rigorous mechanistic diagnosis of generalization under temporal shift.

cs.LG · q-fin.ST

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.12357

Improved Local Leakage Resilience of Shamir Secret Sharing and Worst-Case Optimal Polynomial Intersection

Yihang Sun, Mary Wootters

原文摘要

We study two problems: Local Leakage Resilience (LLR) for Shamir secret sharing, and worst-case Optimal Polynomial Intersection (OPI), over a prime field $\mathbb{F}_p$. For a polynomial $Q(X)$ of degree less than $k$, LLR for Shamir secret sharing asks what a few leaked bits from each of the $n$ shares $Q(α_1), \ldots, Q(α_n)$ reveal about $Q(0)$. Given (worst-case) input lists $S_1, \ldots, S_n \subset \mathbb{F}_p$, OPI seeks a polynomial $Q(X)$ so that $Q(α_i) \in S_i$ for as many $i$ as possible. Leveraging recent connections between these two problems due to (Sun, Wootters 2026), we improve the state-of-the-art for both problems. For LLR, we show that there is some constant $δ> 0$ so that Shamir secret-sharing is one-bit LLR (only a negligible amount about $Q(0)$ is learned) whenever $R := k/n \geq 1/2 - δ$. This is the first result to break the so-called ``one-half barrier'' for LLR, and improves over the previous best known result, requiring $R \geq 0.668$ (Kasser, 2025). For OPI, we give a quantum algorithm that finds a polynomial $Q(X)$ that agrees with at least a $\mathsf{SCL}_ρ(R)-\varepsilon$ fraction of the lists, for every fixed $\varepsilon>0$, where $\mathsf{SCL}_ρ$ is the \emph{semicircle law} of (Jordan et al., 2025). This improves previous algorithmic (and existential) results of (Jo, 2026) and (Horinaga, Yamakawa, 2026). We also give further improved existential results for all OPI instances; and when the evaluation points $α_i$ are chosen at random, for every rate $R > 0$ we show that, with high probability, for worst-case $S_i$ there exists \emph{perfect} solution. Finally, we adapt the hardness result of (Yamakawa, Zhandry, 2024) to apply to OPI (rather than a folded version); this gives an unconditional separation between the quantum and classical hardness of OPI relative to a membership oracle.

quant-ph · cs.CR · cs.DM · cs.IT

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2608.06412

Beyond Visibility: Real-Time Surface Accessibility Fields from Sparse LiDAR

Bradley Scott, Sam Schofield, Richard Green

原文摘要

Understanding which surfaces a given tool can physically access is fundamental to robotic interaction, yet 3D perception typically stops at geometric reconstruction or visibility estimation. Visibility does not imply accessibility: a visible point may still be blocked by tool geometry or approach-corridor clearance. We propose the Accessibility Field: a per-point surface accessibility labelling, computed from streaming sparse LiDAR and updated at sensor rate as the platform moves. Our GPU-based method evaluates surface points against precomputed tool-geometry kernels across rotated approach orientations, underpinned by a scan-centric TSDF integration scheme suited to non-repetitive sensors such as the Livox Mid-360. The system is tool-agnostic, requires no prior scene model, and runs in real time on both workstation and Jetson Orin edge hardware. On synthetic primitives, our method achieves F1=90.8%, against 69.8% for a Hidden Point Removal baseline; ablations within our own pipeline give 81.5% replacing the TSDF query with a point-cloud collision check, and 33.9% replacing scan-centric integration with projective, nvblox-style integration. On a mature synthetic Pinus radiata model, the method labels 56.8% of visible pine branch surfaces as physically inaccessible.

eess.IV · cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2408.02845

Heterogeneous Graph Attention Network With Multimodal Feature Selection for Cancer Multiomics Integration

Sina Tabakhi, Charlotte Vandermeulen, Ian Sudbery, Haiping Lu

原文摘要

The increase in high-dimensional multiomics data demands advanced integration models to capture the complexity of human diseases. Graph-based deep learning integration models, despite their promise, struggle with small patient cohorts and high-dimensional features, often applying independent feature selection without modeling relationships among omics. Furthermore, conventional graph-based omics models focus on homogeneous graphs, lacking multiple types of nodes and edges to capture diverse structures. We introduce a Heterogeneous Graph ATtention network for omics integration (HeteroGATomics) to improve cancer diagnosis. HeteroGATomics performs multimodal feature selection through a multi-agent system, creating dedicated networks of feature and patient similarity for each omics modality. These networks are then combined into one heterogeneous graph for learning holistic omics-specific representations and integrating predictions across modalities. Experiments on three cancer multiomics datasets show that HeteroGATomics outperforms baseline methods across most evaluation metrics in cancer diagnosis. External validation on an independent breast cancer cohort further demonstrates its generalizability, where HeteroGATomics outperforms graph-based multiomics integration methods across evaluation metrics. Moreover, HeteroGATomics enhances interpretability by identifying important biomarkers contributing to the diagnosis outcomes.

cs.LG · cs.MA · q-bio.BM · q-bio.GN

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04114

SciPhi-Chat: Training a Spatial Audio Agent for Question Answering, Dialogue, and Tool Use

Gijs Wijngaard, Hannes Gamper, Sebastian Braun

原文摘要

Spatial audio question answering requires associating sound identities with locations and reasoning about their relations. This paper introduces SciPhi-Chat, a spatial audio agent that answers questions, conducts multi-turn dialogue, and calls tools on first-order ambisonic recordings. The accompanying MASU benchmark contains 12,000 samples across 42 task categories, the widest task coverage of any spatial model, with real recordings and simulated audio scenes. SciPhi-Chat achieves a score of 44.9%, 7.4 percentage points above the previous state-of-the-art. On the external ASA-Bench, SciPhi-Chat achieves 74.5% azimuth and 60.0% elevation accuracy, the highest among locally evaluated models. Removing reasoning supervision reduces ASA-Bench accuracy by 17.4 percentage points. All materials will be released via https://github.com/microsoft/sciphi-chat

cs.SD

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04113

The Evolution Strategies Paradox: Better Models from Worse Perturbations

Roberto Dailey, Kajetan Schweighofer, Jamieson Warner, Conor F. Hayes, Elliot Meyerson, Babak Hodjat, Risto Miikkulainen, Xin Qiu

原文摘要

How is it possible for perturbation-based methods like Evolution Strategies (ES) to optimize billion-parameter large language models (LLMs)? Classical search methods are highly dependent on finding good candidates within generated populations. However, when using ES to fine-tune LLMs, a phenomenon can occur where all offspring models consistently perform worse than the parent model, yet during training, the parent model consistently improves. This paradox occurs across models, tasks, and even modalities. This paper characterizes the paradox, then examines how the local reward landscape and properties of the ES estimator contribute to it: in high-dimensional spaces, exploration leads to worse performance, yet still highlights good directions; aggregating those directions then allows for making local improvements. The result can be seen as a new optimization paradigm where a principled interaction of noise and curvature produces the paradox as a byproduct of effective optimization. These findings provide guidance for taking advantage of perturbation-based methods in LLM fine-tuning and in high-dimensional optimization in general.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04112

Co-Occurrence Priors Override Visual Spatial Evidence: Understanding False Alarms in Vision-Language Models for Construction Struck-By Hazard Detection

Xiaoshan Zhou

原文摘要

Vision-language models (VLMs) increasingly support construction hazard monitoring, yet their frequent false alarms remain poorly understood. This study investigates whether false struck-by hazard warnings arise from failures in object recognition, spatial relation encoding, or hazard decisions dominated by learned co-occurrence priors. We evaluated Qwen2.5-VL-7B and InternVL3.5-8B using 355 counterfactual worker-excavator scene pairs with controlled spatial proximity and worker-removal variants, combining behavioral analysis, internal representation probing, causal interventions, and decision correction. Both models recognized the entities and encoded their spatial relations (out-of-fold AUROC: 0.82 and 0.91), assigning higher hazard scores to near than far scenes in 79% and 73% of pairs. Nevertheless, 74% and 100% of far scenes were falsely classified as hazardous. Removing the worker reduced Qwen's hazard rate to 11%, while worker presence exerted 2.4- and 3.2-times greater influence on hazard scores than spatial proximity in the respective models. Attention-head interventions, neuron-level interventions, and activation steering failed to eliminate this co-occurrence bias. However, conditioning hazard decisions on the model's own spatial-relation representation improved Qwen's specificity from 0.36 to 0.58 at matched sensitivity, outperforming threshold recalibration. These findings reveal that VLM struck-by false alarms stem primarily from decision-making failures rather than deficiencies in object recognition or spatial encoding, suggesting that future efforts should prioritize conditioning hazard decisions on encoded spatial evidence rather than further improving visual recognition alone.

cs.CV · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04111

Generalizable AI-Generated Image Detection with Embedding-Space Explanations

Nikos Giakoumoglou, Paschalis Giakoumoglou, Tania Stathaki, Symeon Papadopoulos

原文摘要

AI-generated images have reached unprecedented realism, while instruction-driven AI editors re-synthesize images for local changes, leaving no splice boundary to localize. Most detectors are black boxes with unverifiable verdicts and generalize poorly because adapting a backbone on the detection distribution memorizes generator-specific fingerprints that quickly become obsolete. Specialist detectors built on pure-vision representations, particularly DINOv3, generalize well across generators, but explainable methods forgo this advantage: jointly training one multimodal model for classification and language generation subjects its visual representation to the language objective. We therefore require that explanation training cannot alter the verdict. We introduce FOX, which first builds and locks a specialist DINOv3 detector, then uses its representation to guide a lightweight VLM toward forensic explanation. A joint-embedding predictor maps locked visual evidence into a frozen text encoder's semantic space, producing guidance that the VLM verbalizes on demand. FOX achieves 97.4 mean AUC across 14 unseen generators, 69.8 BERTScore on XPlainVerse, and 56.1 on LOKI.

cs.CV · cs.AI · cs.MM

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2603.07618

SMAT: Staged Multi-Agent Training for Co-Adaptive Exoskeleton Control

Yifei Yuan, Ghaith Androwis, Xianlian Zhou

原文摘要

Effective exoskeleton assistance requires co-adaptation: as the device alters joint dynamics, the user reorganizes neuromuscular coordination, creating a non-stationary learning problem. Most learning-based approaches do not explicitly account for the sequential nature of human motor adaptation, leading to training instability and poorly timed assistance. We propose Staged Multi-Agent Training (SMAT), a four-stage curriculum inspired by how users acclimate to a wearable device. In SMAT, a musculoskeletal human actor and a bilateral hip exoskeleton actor are trained progressively: the human first learns unassisted gait, then adapts to the added device mass; the exoskeleton subsequently learns a positive assistance pattern against a stabilized human policy, and finally both agents co-adapt with full torque capacity and bidirectional feedback. We implement SMAT in the MyoAssist simulation environment using a 26-muscle lower-limb model and an attached hip exoskeleton. Our musculoskeletal simulations demonstrate that the learned exoskeleton control policy produces an average 10.1% reduction in simulated hip muscle activation relative to the no-assist condition. We validated the learned controller in an offline setting using open-source gait data, then deployed it to a physical hip exoskeleton for treadmill experiments with five subjects. The resulting policy delivers consistent assistance and predominantly positive mechanical power without the need for any explicitly imposed timing shift (mean positive power: 13.6 W at 6 Nm RMS torque to 23.8 W at 9.3 Nm RMS torque, with minimal negative power) consistently across all subjects without subject-specific retraining.

cs.RO · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2610.28929

Scriptorium: System Design and Implementation of a Multi-Human, Multi-AI Collaborative Digital Library Search Platform

Abbas Pirmoradi, Milad Momeni, Anna Chu, Orland Hoeber

原文摘要

This technical report documents the design and implementation of Scriptorium, a digital library search interface designed to support multi-human, multi-AI collaboration: full cross-user transparency of search activity combined with a Manager--Worker architecture of seven role-specialized generative AI agents. A companion short demonstration paper describes Scriptorium's interaction design; because a demonstration format is constrained to a few pages, it necessarily omits most implementation detail. This report documents Scriptorium's complete system architecture, data model, collaborative-interaction design rationale, and AI orchestration mechanism, including verbatim agent prompts, the deterministic guardrails that constrain them, the language model and retrieval configuration, and the steps needed to run the system, so that the design can be inspected, critiqued, reproduced, and reused by other researchers. We also discuss the system's current limitations and the user study we are planning to evaluate how the shared record of search activity and the scoped agents support collaborative search, including collaborators' awareness of one another's work and how they use the agents.

cs.HC · cs.IR

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.18147

Dynamic Pooling and Regional Participation in Deceased-Donor Organ Allocation

Genta Okada

原文摘要

Moving from geographically fragmented to pooled waiting lists in deceased-donor organ transplantation can improve efficiency, but it raises concerns about regional fairness and participation incentives. This paper studies Pareto gains from such transitions in a multi-class queueing model with impatient agents, perishable items, and a tractable homogeneous compatibility friction. In contrast to standard approaches that emphasize static match quality or overlook regional incentives in dynamic contexts, this analysis identifies dynamic risk pooling as the primary source of Pareto improvements for regions. Pooling reduces organ waste resulting from stochastic supply-demand fluctuations, temporal mismatches, and random compatibility outcomes. Although this dynamic surplus theoretically enables Pareto improvements, a purely utilitarian policy that treats all patients equally across regions often fails to achieve these improvements when regions are asymmetric. In these situations, utilitarian redistribution removes an excessive number of organs from high-supply regions and violates their participation constraints. To address this conflict, it is demonstrated that the set of achievable utilities can be represented by mixtures of static priority index policies. Building on this structure, the Constrained Pareto Frontier Search (CPFS) is proposed as a polynomial-time procedure that computes an approximately optimal allocation policy with arbitrarily small participation-constraint violations. The results indicate that, by adjusting priority indices to ensure redistribution remains within the gains from pooled supply, demand, and compatibility risk, policymakers can achieve global efficiency while maintaining local participation incentives.

econ.TH · cs.GT

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04107

Say Less, Mean Better: Training-Free Verbosity Control for Language Models

Mahmoud Alyosify, Mirna Imbabi, Eman Elkhamisi, Rahatara Ferdousi

原文摘要

Large language models (LLMs) often generate responses longer than necessary, increasing inference latency and computational cost. While prompting can encourage conciseness, controlling output length without compromising response completeness remains challenging. Aggressive early stopping can reduce generation costs but risks terminating responses prematurely. We introduce MinimalLM, a training-free framework for reliable verbosity control that combines uncertainty-aware, length-dependent stop-token adjustment with structural completeness guardrails. MinimalLM separates the decision of when to stop from whether stopping is structurally appropriate, coordinating response-level instructions, decoding-time intervention, and completion checks without modifying model parameters or requiring auxiliary models. We evaluate MinimalLM across three open-source LLMs and 330 questions, producing 5,940 responses across baseline and ablation configurations. MinimalLM reduces generated tokens by 47.6%, accelerates individual responses by 19-68%, and eliminates structurally truncated outputs in the primary evaluation. Despite substantial token savings, it achieves an average quality score of 8.04/10, compared with 8.42/10 for unmodified models and 6.46/10 for length-matched truncation. Evaluations across diverse subject domains (such as healthcare, programming, law, and science) demonstrate robust efficiency gains, while ablations establish the complementary roles of decoding control and completeness constraints. Our findings highlight that efficient generation requires more than producing fewer tokens: it requires controlling when generation ends while preserving response integrity. MinimalLM offers a lightweight, training-free approach toward this goal.

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04106

Structure, Dimensionality, and Distribution Shape - What Makes Pretraining Data Transferable, and Why the Input Pipeline Decides Which

Sangyoon Bae, Jiook Cha

原文摘要

What makes a pretraining corpus transferable? We describe a corpus by three properties: its structure, its dimensionality, and the distribution shape of each unit's values. We propose the Transfer-Invariance Principle: the same property can be irrelevant to transfer under one input transform and decisive under another, so the model's input transform, not the data modality, decides which properties can matter. To test it we build TransferForge, a generator that places a synthetic population at any prescribed structure, dimensionality, and distribution shape. KernelSynth, the synthetic prior of Chronos, cannot, because it draws independent series. When the model standardizes each input window (PatchTST on UCR), the distribution shape is inert and the structural setting governs. Synthetic pretraining then matches a real corpus of the same size and exceeds it once scaled. When the model encodes calcium fluorescence signals without normalization (POYO Plus), the shape becomes a governing constraint, and a shape-matched synthetic corpus replaces real pretraining with zero real neural activity. Flipping only the input normalization, with the model and data fixed, changes the effect of a shape manipulation by a factor of 5.3, so the pipeline, not the modality, sets the switch. The principle yields a recipe that chooses what to synthesize before any training. Pretraining data then becomes something to manufacture for the model, most valuably where real data is scarce, expensive, or private.

cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2511.20836

Structured Prompts Improve Evaluation of Language Models

Asad Aali, Muhammad Ahmed Mohsin, Vasiliki Bikia, Arnav Singhvi, Richard Gaus, Suhana Bedi, Hejie Cui, Miguel Fuentes 等 18 位作者

原文摘要

As language models (LMs) are increasingly adopted across domains, high-quality benchmarking frameworks are essential for guiding deployment decisions. However, frameworks such as Holistic Evaluation of Language Models (HELM) typically evaluate models under a single static prompt configuration, even though model behavior depends strongly on prompt choice. As a result, reported scores can reflect prompt choice as much as model capability. Declarative prompting frameworks, such as DSPy, offer a scalable way to operationalize this evaluation protocol by evaluating models under structured prompting strategies rather than static prompts. Here, we conduct a reproducible DSPy+HELM benchmarking study and quantify how prompt choice impacts benchmark outcomes. Using five prompting methods, we evaluate four frontier and two open-source LMs across seven benchmarks against HELM baseline scores and find that prompt choice can materially impact leaderboards. Structured prompting improves performance (by 6% on average), alters comparisons (leaderboard rankings shift on $5/7$ benchmarks); chain-of-thought provides the strongest cost-performance tradeoff while more advanced optimizers offer smaller additional gains. Beyond the known intuition that prompts affect outputs, our benchmarking study integrates structured prompting into an established evaluation framework and quantifies how prompt choice alone impacts leaderboards. We open-source (i) DSPy+HELM Evaluation (https://github.com/stanford-crfm/helm/pull/3893) and (ii) Prompt Optimization Pipeline (https://github.com/StanfordMIMI/dspy-helm).

cs.CL · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04104

ModelBlaster: Closing the Performance Gap through Learning-Augmented Agents Across Models, Kernels and Hardware Backends

Shengjun Kris Dong, Dima Nikiforov, Sahil Modi, Sana Damani, Edward C Lin, Siva Kumar Sastry Hari, Christos Kozyrakis

原文摘要

Optimizing accelerator code across hardware architectures is difficult. ML compilers search predefined spaces of schedules and micro-kernels, while LLM-guided optimization can explore alternative algorithms, operator decompositions, and hardware-specific techniques. However, existing LLM agents typically target a single programming interface, toolchain, or accelerator family, focus on kernel-level optimization without model-level transforms, and offer limited reuse of experience across hardware targets. We present ModelBlaster, a Memory-Augmented In-Context Reinforcement Learning (MAIC-RL) framework for optimizing data-parallel kernels and whole models across heterogeneous SIMD and SIMT accelerators. Its mutable policy is a persistent knowledge base updated at inference time by Actor-Critic Textual Gradient Descent (AC-TGD). The knowledge base maps performance states to actions, outcomes, and constraints, enabling experience reuse across kernels, models, and hardware targets. A model-level optimizer uses kernel-level knowledge to inform whole-model decisions. The same optimization loop and knowledge schema support CUDA on NVIDIA GPUs, OpenCL on Qualcomm GPUs, and RVV intrinsics on RISC-V vector cores, with only lowering and measurement layers specific to each target. Compared with PyTorch, ModelBlaster achieves geometric-mean speedups of 1.85x, 5.58x, and 5.18x on KernelBench Levels 1, 2, and 3, and 1.69x on SOL-ExecBench Levels 1 and 2 on a B200 GPU. On Qualcomm Adreno GPUs, it achieves 5.8x over the vendor SDK and 1.7x over Apache TVM; on RISC-V vector cores, it achieves 2.7x over ExecuTorch with XNNPACK across 70 KernelBench problems.

cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.18685

WeaveRL: Weaving Reconstruction into Scene-Aware Fabrics for Perceptive Reinforcement Learning

Remo Steiner, Vikram Ramasamy, David Tingdahl, Sam Mady, Karl Van Wyk, Nathan Ratliff, David Recasens, Soha Pouya 等 10 位作者

原文摘要

Reinforcement learning allows robots to acquire complex skills, but producing policies for geometrically complex manipulation remains difficult. A promising approach is to learn on top of collision-avoidant controllers, such as geometric fabrics. However, these approaches have relied on static, hand-specified representations of the scene. Integrating active, online 3D perception into massively parallel RL training has so far been inaccessible. We introduce a GPU-accelerated method that reconstructs the scene as a collection of surfels across thousands of parallel simulation instances during active rollouts. This lets policies operate over sensor-derived, rather than hand-specified, geometry. On a suite of collision-dense manipulation tasks, our surfel fabrics enable policies to tackle geometrically complex scenes where primitive-based baselines fail, while maintaining sim-to-real transfer. Furthermore, policies learned with a scene-aware fabric are more robust to the introduction of novel geometry at test time, improving collision-free task completion under unseen obstacles from 35% to 61%. We release our reconstruction system, training code and test dataset to spur research in this direction.

cs.RO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04102

Is This 3D Asset Usable? Beyond the Beauty Render for Asset Curation

Rushikesh Zawar, Vladimir Kim

原文摘要

Large 3D repositories make it easy to collect assets, but difficult to determine whether they can be reused reliably across scene assembly, editing, rendering, game production, and training-data construction. An asset can look plausible in RGB while hiding disconnected geometry, problematic UV organization, or materials that fail to resolve correctly after import. We introduce Diagnostic Rendered Evidence (DRE): aligned beauty, wireframe, and UV-seam-plus-wireframe views scored by pass-specific frozen ViT classifiers and combined through late fusion. On 1,234 human labeled TexVerse assets, nested grouped evaluation achieves 86.36% balanced accuracy and 88.32% bad-asset recall. On a separate, balanced 764 external reference test, DRE achieved 88.9% balanced reference-label accuracy. Our method requires only an imported asset and renderer-derived evidence, without repository identity, creator, popularity, text, or source metadata at inference. Our results support cross-corpus portability and show that purpose built diagnostic views expose defects that just the beauty views can miss. Webpage: https://rbz-99.github.io/is-this-3d-asset-usable/

cs.CV

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.18877

Lotus: Learning to Understand Time-series by Learning to Forecast

Van Dai Do, Huu Hiep Nguyen, Minh Hoang Nguyen, Manh Nguyen, Hung Le

原文摘要

Time-series language models typically align numerical signals with language using labels, captions, or question-answer pairs, without explicitly requiring their representations to preserve how the underlying signals evolve. We introduce forecast grounding, a pretraining principle that requires language-aligned time-series representations to remain predictive of the future evolution before they are adapted to downstream tasks. We instantiate this principle in Lotus, which divides each time-series channel into non-overlapping patches and maps them into the LLM embedding space. It then predicts the next patch from the LLM hidden states of the time-series tokens, encouraging the learned representation to retain temporal information. To capture both temporal changes and periodic patterns, we also propose a temporal-spectral patch representation that combines each raw time-series patch with its spectral representation before mapping it into the LLM embedding space. Across three language model backbones and six downstream evaluation datasets, Lotus achieves the best average performance with minimal trainable parameters, reaching 59.91% accuracy and outperforming the strongest matched baseline by 3.36%. Our representation analyses further show that Lotus uses the numerical input and preserves temporal structure in its learned patch representations. These results establish forecast grounding as a promising pretraining principle, enabling compact time-series language models to outperform models 25-50 times their size on TSRBench. Our code is publicly available at https://github.com/DA2I2-SLM/Lotus

cs.LG · cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04101

Structured Clinical Memory as Context for Probabilistic Decision Models: A Synthetic Evaluation with Jev

Jash Sayani, Gaurang Karwande, Achinth Bharadwaj

原文摘要

Longitudinal clinical context is often distributed across noisy, repetitive forms and narrative records. We study whether a compact structured memory can improve a probabilistic decision model's answers when the underlying patient vignette is already available. We created 30 fully synthetic cases spanning twelve clinical domains, paired each with synthetically generated structured memory, and evaluated 120 typed questions with Jev. The raw-vignette arm achieved 90.0% accuracy (95% normal-approximation confidence interval: 84.6%-95.4%); adding memory achieved 93.3% (88.9%-97.8%). Five questions changed from incorrect to correct and one changed in the opposite direction (exact McNemar $p=0.219$). Choice log loss decreased from 0.258 to 0.175 and binary Brier score from 0.054 to 0.022. Memory added 310 input tokens per call on average, while observed mean latency was unchanged at approximately 0.5 seconds. These results are directionally positive rather than clinically validating: the cases are synthetic, the labels are single-author judgments, and the memory is an oracle upper bound rather than extracted production output. The pilot supports evaluating structured clinical memory as a reusable state interface for decision models.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04100

The Agent Board: Kanban for Machines; Durable Coordination for Agents That Outlive the Conversation

Tyler Hunt, Donald Nguyen, Gurbinder Gill

原文摘要

Harnesses for large language models have converged on a single-threaded shape: one agent, one history, one machine, one sitting. That shape fails work that spans many agents, outlives any one model call, or waits on systems that are not agents. We describe the Agent Board, the coordination substrate in the Corvic harness. The board is a durable, shared table of semi-semantic tasks: a few exact fields the system routes on, and a free-form body that only language models read. Agents post tasks, watch tasks, and block on them, ending their turn and holding no process until a matching post wakes them. A system agent, the Board Keeper, acts as the interrupt handler for the board: it carries events from non-agent parts of the harness onto the board, matches them to waiting tasks by reading their prose, and allocates new agents for tasks that name a kind of agent rather than an instance. Loadable coordination policy kernels give statuses and roles meaning for a given pattern of work, without the board enforcing them. The design mirrors the kanban board a human team puts on a wall. We argue it is what persistent agent products such as Meta Muse, OpenAI dots and SpaceXAI Grok Bot need once an agent must wait, hand off work, or work with agents from another vendor. We view the board as an operating-system scheduler for agents and state where the analogy breaks. We separate what the mechanism guarantees from what it leaves to convention, and report a trace from a live board and three small simulated pilots. In the pilots, the wait cost of the board was flat in wait length while naive polling grew about 165-fold, but a plain external watcher was cheaper than the board. The Board Keeper matched prose waiters with no false wakes but missed 3.6 percent of real wakes.

cs.AI · cs.DC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04099

Evaluation Contracts and Evaluation Leverage: Sliced-Gap Readings in Structured Prediction Can Move Without Model Change

Hao Lu, Weixing Zhu, Ruiyang Li

原文摘要

NLP evaluation reports routinely cite sliced performance differences, most prominently the intra-inter gap in document-level relation extraction, as capability evidence. We show that such gap readings are joint properties of models and evaluation contracts: scoring function, gold version, threshold, output format, and prompt language. With predictions frozen, a method ordering reverses under a gold-standard exchange (+0.0363, CI excluding zero; tie-level after), movement localizing where gold re-partitions: renames are gauge transformations (Delta=0; re-annotation +0.862). We formalize this as an evaluation leverage account centered on the rank-flip condition. Constructively, the audited evaluation reading (AER) is a contract-marginalized, tie-declared, seed-prescribed, gate-filtered procedure whose tie declarations cover every contract-induced flip on the declared WMT19 grids (14/14; held-out 73% vs. union-of-ties 89%), matching the best single metric contract at epsilon=0.10 equivalence. Evidence spans DocRE, HacRED, a UD probe, and released gold revisions; an alternative-reference-set replication extends it beyond our testbeds. Failed predictions retained as boundaries; a 5% violation gate flags unverifiable settings and bounds survivor-selection bias. We release EvalLever, a preregistered workflow with an amendment trail; on LLM-as-judge grids, the audited ensemble stabilizes every unstable pair (13-15 to zero; generic ensemble shrinkage: the preregistered correlation criterion failed, the companion stabilization criterion passed).

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04098

When Strong Reasoners Join the Pool: Small Union Premiums and Costly Runaway Generations in LLM Routing

Teng-Ruei Chen

原文摘要

Routing benchmarks often judge a router against a hindsight oracle that counts a query as solved whenever any model's recorded answer is correct, but under stochastic decoding that oracle mixes two different objects: the clairvoyant ceiling of committing to one model per query and the premium of calling several models and keeping any correct answer. A companion study measured this premium at 3.5 percentage points (pp) for 11 small open models on MATH-500; how large it is once current reasoning models enter the pool was unknown. We measure it with a fixed-protocol, pre-committed generation on a single NVIDIA B300: all 500 MATH-500 problems, 30 seeded samples per problem, Llama-3.3-70B-Instruct and NVIDIA's Nemotron Nano 9B v2 and Nemotron Super 49B v1.5 (a NAS derivative of the same 70B) under vendor sampling, a 65,536-token cap, a pinned Math-Verify scorer, and an append-only attempt ledger, for 45,000 scored responses. The two reasoning models both reach 96.1% accuracy against 76.2% for the 70B baseline. The empirical product-coupling premium is 0.215-0.588 pp across all two- and three-model pools; the empirical plug-in selection headroom over always calling Nano is 1.27 pp; and resampling Nano three times gives practically the same empirical any-correct rate as calling all three models once. Only 31 responses reached the cap, 25 of them with exactly periodic tails, yet the capped responses account for up to 60.2% of a model's output-length-squared proxy, and long generations made a fourth planned model, Llama-3.1-8B, too slow to finish. The premium is far below the companion's 3.5 pp, but models, pool size and protocol changed together; an oracle gap should therefore be reported with the pool, the protocol, the samples per cell and the decoding budget that produced it.

cs.LG · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04097

ScaleTraj: Scale-Decomposed Human Trajectory Generation

Yulin Zeng, Yiheng Zhao, Jun Yan, Yuhong Liu

原文摘要

Individual-level trajectories are essential for applications that rely on user-level mobility behavior. Existing mobility-generation approaches face a fundamental tension between individual-level and population-level statistical fidelity: top-down models reproduce aggregate mobility statistics but struggle to guarantee realistic individual trajectories, while bottom-up generators synthesize individual trajectories yet often fail to preserve key population-level mobility regularities after aggregation. In particular, many bottom-up generated populations fail to reproduce the displacement distribution, whose heavy tail is well summarized by a truncated power law. We formalize this phenomenon as the \textbf{Population-Level Displacement Gap}. To investigate its prevalence, we conduct a systematic evaluation of representative model-driven and data-driven trajectory generators. Our results provide evidence that the Population-Level Displacement Gap is prevalent across diverse generation paradigms. To diagnose this gap, we introduce the \textbf{Scale Fidelity Profile (SFP)}, a scale-decomposed metric that quantifies discrepancies in trip allocations across spatial scales. SFP indicates that population-level displacement mismatches co-occur with systematic errors in how trips are allocated across spatial scales. Motivated by this observation, we develop \textbf{ScaleTraj}, an autoregressive Transformer that decomposes location prediction into a Scale Selection Head that decides the movement scale and a Scale-Aware Cell Selection Head that predicts the destination conditioned on the selected scale. Experiments on two datasets show that ScaleTraj reduces the Population-Level Displacement Gap versus model-driven and data-driven baselines, while preserving fidelity. Our code is available at https://github.com/SCU-TrustworthyAI/ScaleTraj.

cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.16286

A Cheeger Inequality for Hypergraphs and Its Applications

Raj Kamal, Amitabha Bagchi

原文摘要

Hypergraphs provide a natural framework for modeling higher-order relationships, but the development of spectral techniques with provable guarantees for general non-uniform hypergraphs remains challenging. Building on Banerjee's normalized adjacency matrix and Spiro's averaging-based diffusion framework, we develop a spectral framework for non-uniform hypergraphs and establish Cheeger's inequality for their conductance. A fundamental result in the spectral theory of hypergraphs asserts that, for every non-covering hypergraph, the second-smallest eigenvalue of its normalized Laplacian is at most one. This spectral characterization yields an improved Cheeger's inequality for non-covering hypergraphs, and we show that the resulting inequality is tight on both sides using cycle and cube hypergraphs. Our framework further yields higher-order Cheeger inequalities and provides theoretical guarantees for Fiedler's spectral partitioning algorithm, all in the setting of hypergraphs. Finally and most notably, we construct a new family of optimal hypergraph expanders that is tight for the Alon-Boppana bound.

math.CO · cs.DS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.04096

Online Hamilton-Jacobi Reachability Onboard the Robot Control Loop: A Real-Time GPU Solver for Safety-Critical Autonomy

Kevin Weekly, Alexandre M. Bayen

原文摘要

Hamilton-Jacobi (HJ) reachability provides certified backward reachable tubes (BRTs) for safety-critical robotic systems, but the level-set computation scales exponentially in the state dimension and has historically been performed offline. As autonomy stacks (ADAS/AV, humanoid robotics, and vision-language-action models) increasingly rely on black-box, probabilistic policies, a deterministic safety layer with formal guarantees is an essential complement. We present a CUDA implementation of the canonical WENO5 + TVD-RK3 + Lax-Friedrichs level-set Hamilton-Jacobi-Isaacs solver for 2D, 3D, and 4D dynamics that is fast enough to run inside a robot's control loop. In 32-bit floating point (FP32) on a single workstation GPU (NVIDIA RTX 4000 Ada), it computes the 3D Air3D BRT at 112 Hz, a 4D pursuit-evasion BRT at 61 Hz, and a 2D wind-affected drifter BRT at 74-147 Hz, 5.9-17.5x faster than the JAX reference hj_reachability at matched precision and 1.7-16x faster than a torch.compile reimplementation of the same scheme. In 64-bit floating point (FP64) it agrees with hj_reachability on eight benchmark cases, reproduces Mitchell's Air3D BRT to within 0.04% of the reference cell count, and runs 2.0-4.6x faster. On an embedded NVIDIA Jetson AGX Thor it computes Air3D at 66 Hz (N=41) and the 2D drifter at 54 Hz, 3.8-8.9x faster than hj_reachability on the same device. As a multi-agent application, an in-kernel label-propagation extension recovers the K-agent capture-basin Voronoi tessellation from a single combined HJ solve, independent of fleet size; it runs at 4.0 ms per tick in a closed-loop wildfire-response simulation of six quadcopters under time-varying, non-uniform wind.

cs.RO · eess.SY

查看论文
arXiv ↗PDF ↗