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

+2471新增论文
+2025其中 AI 相关
+163已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-11 04:56 · 全站最近入库：2026-11-10 15:22各信源进度 →

统计口径与数据新鲜度

统计 2026-11-10 05:03 — 2026-11-11 05:03（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：2026-11-11 04:56最近发现新增版本：2026-11-10 15:22库内论文最新更新时间：2026-11-10 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。

20,643 已收录论文17,531 AI 相关5,643 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 16,219 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-11 04:55（北京时间）
新论文 · v1arXiv:2611.12392

More Than Just Style: Benchmarking Personality Influence in Large Language Models

Yuhao Liu, Mengyu Wang, Xiao Chen, Lu Liu

原文摘要

In addition to the effects of explicit input prompts and instructions, large language model outputs are influenced by underlying behavioral preferences similar to human personality tendencies. These tendencies can affect how models interpret situations and make decisions. However, their influence is difficult to disentangle from the effects of model capabilities and instructions. Therefore, we introduce SPACE, a benchmark framework combining a controlled evaluation protocol with complementary datasets for personality verification, capability assessment, and hierarchical situational tasks. Controlled comparisons across guidance conditions help distinguish personality influence from measured capability changes and immediate instruction effects. Experimental results demonstrate that although prompting can shift generation style, models' internal personality tendencies are persistent and not readily overridden by opposing instructions. Furthermore, these underlying tendencies have an observable influence on LLM task performance, highlighting the need for greater attention to these characteristics and potential applications. Training code and SPACE benchmark specifications are available at https://github.com/Niklauseik/SPACE-training.

cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12391

Semantic-Aware Communications in Industrial IoT:What Can Be Tested Today, and What Is Missing?

Rute C. Sofia

原文摘要

Classical networks treat every bit as equally important, which wastes time and energy in factories and requires extra traffic engineering for critical traffic. Semantic communications send only what a task needs, but how can they be tested under realistic conditions? This paper examines what can be tested today in 6G industrial IoT, that is, networks of factory machines, robots, and sensors, and what is still missing. It introduces semantic importance, a score attached to each message that states how much the message matters, and uses it in a context-driven design to derive eight testing needs. Existing simulators, emulators, testbeds, codecs, and learning frameworks are assessed against these needs, which reveals five gaps. We propose a toolchain of four components built around the ns-3 simulator for simulated and emulated scenarios.

cs.NI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.39164

Quantum-Inspired Score-Based Variational Inference

Yuchen Cong, Zerui Tao, Chao Li, Zhe Sun, Qibin Zhao

原文摘要

Score-based variational inference (SBVI) provides an alternative to classical Kullback--Leibler (KL)-based variational inference by learning a distribution using only score queries of the target. EigenVI~\cite{cai2024eigenvi} formulates SBVI as an eigenvalue problem, but faces two limitations: the number of variational parameters grows exponentially with the data dimension, and the single-eigenvector formulation cannot represent the full minimum-eigenvalue eigenspace. We propose QuanVI, a quantum-inspired SBVI algorithm that addresses both limitations. Rather than selecting a single eigenvector, QuanVI captures a learned subspace through a density-operator formulation. To enable scalable computation, \model{} employs a matrix-product-operator (MPO) structure parameterized by a quantum tensor network (QTN) with unitary-constrained local tensors and auxiliary ancilla wires. Numerical experiments demonstrate the scalability of QuanVI beyond EigenVI and competitive performance on synthetic distributions and Bayesian posterior-approximation benchmarks, including challenging non-Gaussian targets. The code is available at https://github.com/ChaoLiAtRIKEN/quanvi.

cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2602.01312

Understanding What TRAK Gets Right and Wrong: A Theory of Data Attribution

Han Tong, Haolin Zou, Shubhangi Ghosh, Arian Maleki

原文摘要

Data attribution, which traces a model's prediction back to specific training data, is an important tool for interpreting sophisticated AI models. The widely used TRAK algorithm addresses this challenge by first approximating the underlying model with a kernel machine and then leveraging techniques developed for approximating the leave-one-out (ALO) risk. Despite its strong empirical performance, the theoretical conditions under which TRAK approximations are accurate, as well as the regimes in which they break down, remain largely unexplored. In this paper, we provide a theoretical analysis of the TRAK algorithm, characterizing its performance and quantifying the errors introduced by the approximations on which the method relies. We show that although the approximations can incur significant errors, TRAK preserves the separation between highly influential and weakly influential data points. We corroborate our theoretical results through extensive simulations and empirical studies.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12386

Say Shibboleth: Fingerprinting the LLMs and Harnesses Behind Agentic Attacks

Dario Pasquini, Reworr, Giuseppe Ateniese, Manish Bhatt, Sergej Epp

原文摘要

We study the fingerprinting of agentic cyberattacks. Our objective is to reliably infer information about the attacker's configuration, such as LLM version and harness, from telemetry. To achieve this, we introduce a technique we call shibboleth traps: modular honeypot services designed to elicit discriminative behaviors from agentic attackers. The traps make informative interactions a condition for obtaining resources the agent seeks, eliciting model declarations and characteristic writing without requiring a successful prompt injection. To achieve harness fingerprinting, we introduce general mechanisms to indirectly probe peculiar execution behaviors of harnesses. These probes reveal how a harness masks secrets in tool output, enforces command time limits, and schedules tool calls, using only the effects visible in subsequent requests. We show that combined declarative and behavioral inference distinguishes among 15 candidate models with an average 90% top-1 accuracy, with behavioral classifiers evaluated on held-out harnesses or task families. In a separate experiment across four harnesses, fixed probe rules identify the correct harness with 92% accuracy. When instructed to conceal their identity, agents still leave useful behavioral evidence even as their declarations become uninformative.

cs.CR · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.33958

How Much Do Structural Tools Help Small Coding Agents? A Pre-Registered Scale Study of Code Localization

Saurav Bellam

原文摘要

Code-graph and embedding tools are routinely added to LLM coding agents on the premise that they supply repository structure the model lacks. We report a pre-registered study of how much such tools help off-the-shelf agents localize the code an issue requires changing and how the benefit varies with model size. Nine open-weight models (Qwen3 0.6B to 32B, Gemma 4 E2B to 12B) run one fixed localization harness on 500 held-out SWE-bench tasks under four tool sets (grep only; plus embedding search; plus code-graph tools; both) and a placebo arm whose structural tools return random entities, with five sampled seeds per condition: 170,500 episodes under hypotheses, tests and claim criteria frozen before the runs. The tools raise canonical function Acc@10 by +0.06 to +0.10 for the Qwen3 models up to 14B and the two small Gemma 4 models, by +0.03 at 32B (a near miss after correction), and not at all for Gemma 4 12B, which rarely submits; the gain is larger for small models than for the largest model in both families. The placebo shows that the gain is caused by what the tools return rather than by having them, and that random tool output harms two small models. The pre-registered claim that a 4B model with tools matches a 32B model without them fails. Secondary analyses show that the embedding index, not the code graph, carries the gain; that the gain replicates on tasks created after every model's release and survives a shell harness, thinking mode, FP8 serving and a file-level metric that output formatting cannot inflate; and that it is mostly exposure: the tools show the gold code more often. Injecting the gold code directly explains the size trend. A 32B model already finds the code without the tools, so they add little exposure, while small models act on a shown answer less often (conversion 0.09, 0.24 and 0.40 at 1.7B, 4B and 32B). We release registration, code, episodes and analyses.

cs.SE · cs.CL · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12383

PCTac: Mechanochromic Photonic-Crystal Tactile Sensor for Dexterous Manipulation

Ruilin Zhang, Xuanshuo Liu, Junzhe Dai, Kefei Wu, Xuhao Qin, Ziyuan Jiao, Wanlin Li, Yurui Qu 等 9 位作者

原文摘要

Modern visuotactile sensors primarily sense contact by imaging the deformation of a compliant interface. Using techniques such as photometric stereo, they can recover detailed contact geometry, often represented as surface normals or depth. However, force-related information typically still needs to be inferred indirectly through calibration, geometric reconstruction, or learned mappings. Mechanochromic photonic-crystal (PC) materials offer an alternative physical encoding mechanism: contact loading changes their periodic microstructure and spectral response, directly transforming local stress into a spatially resolved structural-color field. Despite this potential, PC materials have rarely been explored as visuotactile sensors for closed-loop robot learning. We present PCTac, an open-source mechanochromic PC fingertip that uses a dense hue field as its tactile observation. Loading characterization demonstrates a repeatable relationship between the applied normal load and the observed hue response. We further integrate this representation directly into a Diffusion Policy and evaluate it on three real-robot manipulation tasks. Across 90 physical rollouts, PCTac Hue achieves 68 successes (75.6\%), compared with 60 (66.7\%) for PCTac RGB, 57 (63.3\%) for Wedge RGB, and 29 (32.2\%) for vision-only. PCTac Hue also achieves the highest observed success rate on each evaluated task, including Perturbed Test-Tube Insertion, where it achieves 16/30 successes versus 13/30 for tactile RGB baselines and 1/30 for vision-only. These results demonstrate the potential of mechanochromic structural color as a compact, load-sensitive tactile representation for robot learning and closed-loop manipulation. Project page: https://bey0ndt1me.github.io/PCTac-page/.

cs.RO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12382

AutoVLA-D: Learn a Trajectory-Geometric Condition from Reasoning for Autonomous Driving VLA

Yuguang Yang, Zhewen Tan, Canyu Chen, Linlin Yang, Cheng Chi, Chunyang Liu, Kehua Sheng, Jinyu Yang 等 12 位作者

原文摘要

Autoregressive VLA planning usually predicts continuous trajectories as the final output, but driving instructions often describe actions rather than ego states, which weakens direct conditioning on the final continuous trajectory. We address this gap with TrajCond, which introduces a discrete trajectory-geometric condition for VLA planning. Specifically, we represent TrajCond as a 1D-Brohan differential action trajectory and learn it with a two-stage training framework before final continuous trajectory prediction. With the simultaneous autoregressive generation of discrete and numeric trajectories, we name our method AutoVLA-D (Double). Experiments on NAVSIM navtest, nuScenes, and NeuroNCAP show that discrete trajectories are weaker as direct outputs but more effective as TrajCond, consistently improving final continuous trajectory prediction across closed-loop and open-loop driving benchmarks. Code is available at https://github.com/ShivaTeam/AutoVLA-D.

cs.RO · cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12379

Reconstructing hourly traffic on every street of a capital from a sparse counting network: a graph neural network applied to Paris

Xavier Jurado, C{é}dric Wemmert, Nicolas Reiminger

原文摘要

Micro-scale air quality models require hourly traffic data on every road segment, whereas permanent counters cover only a small part of the network. A graph neural network (GNN) was developed to reconstruct hourly traffic flows on the 32,163 road segments of Paris intra-muros from the 1,573 segments equipped with a counter (4.9%). The model adapts the Graph Neural Network for Urban Interpolation (GNNUI) to a real, sparse counting network. Flows are decoded on the edges of the junction graph (one edge per segment, both directions summed), a traffic climatology serves as an offset in a zero-inflated negative-binomial likelihood, and training proceeds by masking counters. Eight-fold spatial cross-validation, with counters sharing a published series merged before splitting, distinguishes interpolation at a withheld counter in an instrumented district (R${}^2$ = 0.89, error of a quarter of the mean flow) from extrapolation to a district without counters (R${}^2$ = 0.74, a third of the mean flow). These scores hold on main axes where counters are located, while relative errors are about 50% on the few counted local streets. Ablations show that the climatology offset tends to help extrapolation. Road attributes, time cycles and the live state of the counting network matter most, while weather inputs have no significant effect. The masking rate is the only inherited hyperparameter with a clear effect. When evaluated on January--August 2026 (outside the training period), the frozen model stayed around the error band of previous years.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12378

Safety-Constrained Intelligent Control of a 2-DOF Robotic Manipulator Using Control Barrier Functions

Gabriel da Silva Lima, Made Widhi Surya Atman, Wallace Moreira Bessa

原文摘要

Control barrier function (CBF) filters depend on dynamic models that can be inaccurate when their parameters and external forces are uncertain. This paper investigates a radial basis function (RBF) neural network compensator whose output is shared by a Cartesian tracking controller and a joint-space CBF filter. An auxiliary state driven by the difference between requested and applied torques corrects the learning signal, reducing the direct influence of filter intervention on adaptation. A two-link manipulator is evaluated through constrained regulation, ablations, matched uncertainty and disturbance trials, and operating-range and timing studies. The results demonstrate that the proposed controller improves tracking while maintaining joint, velocity, and torque limits under parameter uncertainty, and that both the shared compensation and the auxiliary correction contribute to this improvement.

cs.RO · eess.SY

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12377

Detecting Shadow AI: Unsanctioned LLM Agents on Enterprise Endpoints: A Taxonomy, an Open Detector and a Controlled Evaluation

Jeremy Canale, Fabien Soulis

原文摘要

Shadow IT, the use of technology that the IT organisation has not approved, is a long-standing governance problem. We argue that autonomous AI agents can make it grow much faster: agents make building tools nearly free, act with their user's privileges, and leave behind scripts, automations and MCP servers that are themselves unsanctioned. A simple compounding model shows that, under its assumptions, shadow IT grows exponentially once agentic artifacts reproduce faster than the organisation retires them; retiring them starts with detecting them. We then ask whether unsanctioned agents can be detected on Windows endpoints. We propose a taxonomy of nine classes of agents and six layers of endpoint footprint, and argue that rules based only on the structure of processes and connections cannot separate an agent from its behavioural twin, the same program talking to a non-AI service. We present Agent Detector, an open-source Windows detector that anchors behaviour to the identity of the remote service and correlates it with configuration, installation and execution history, using a catalog of 39 tools and 184 LLM endpoint entries; it informs users through a notice they must acknowledge instead of blocking. On a controlled bench of 12 synthetic agents, 3 AI components and 9 benign workloads, repeated 5 times on one workstation, two behavioural rules flagged every behavioural twin, while Agent Detector flagged none and detected every agent in the majority of runs. Its limits are temporal: a connection to a provider behind a shared content-delivery network is attributable only while the DNS cache holds the provider's name, one to five minutes, and an intermittent agent was seen by about one scan in three. A scan costs about 154 ms. The bench is too small to rank detectors; it maps what each kind of evidence can see: unsanctioned agents can be partly detected, and not by behaviour alone.

cs.AI · cs.CR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12375

AudioCodex: Streaming Autoencoders for Live Audio

Marco Pasini, Javier Nistal, Cyran Aouameur, Stefan Lattner, George Fazekas

原文摘要

We introduce AudioCodex, a fully causal autoencoder for live generative audio applications that compresses stereo audio into compact continuous latent sequences at ~10 Hz. It combines a Mean Flow magnitude autoencoder with latent hypersphere projection, bounded-noise augmentation, representation alignment, and adversarial phase recovery to learn representations that support both efficient downstream generation and high-quality reconstruction. Through extensive ablations, we show that AudioCodex outperforms the state-of-the-art in terms of generation-reconstruction trade-off, whilst remaining strictly causal in both encoding and decoding. This work paves the way towards truly real-time interactive and streaming generative audio applications.

cs.SD · cs.LG · eess.AS

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2605.07907

Consistency Regularised Gradient Flows for Inverse Problems

Alessio Spagnoletti, Tim Y. J. Wang, Marcelo Pereyra, O. Deniz Akyildiz

原文摘要

Vision-Language Latent Diffusion Models (LDMs) (Rombach et al., 2022) provide powerful generative priors for inverse problems. However, existing LDM-based inverse solvers typically require a large number of neural function evaluations (NFEs) and backpropagation through large pretrained components, leading to substantial computational costs and, in some cases, degraded reconstruction quality. We propose a unified Euclidean-Wasserstein-2 gradient-flow framework that jointly performs posterior sampling and prompt optimization in the latent space through a single flow that aligns the prior and posterior with the observed data. Combined with few-step latent text-to-image models, this formulation enables low-NFE inference without backpropagation through autoencoders. Experiments across several canonical imaging inverse problems show that our method achieves state-of-the-art performance with significantly reduced computational cost.

stat.ML · cs.CV · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2411.16890

U-WNO:U-Net-enhanced Wavelet Neural Operator for fetal head segmentation

Pranava Seth, Deepak Mishra, Veena Iyer

原文摘要

This article describes the development of a novel U-Net-enhanced Wavelet Neural Operator (U-WNO),which combines wavelet decomposition, operator learning, and an encoder-decoder mechanism. This approach harnesses the superiority of the wavelets in time frequency localization of the functions, and the combine down-sampling and up-sampling operations to generate the segmentation map to enable accurate tracking of patterns in spatial domain and effective learning of the functional mappings to perform regional segmentation. By bridging the gap between theoretical advancements and practical applications, the U-WNO holds potential for significant impact in multiple science and industrial fields, facilitating more accurate decision-making and improved operational efficiencies. The operator is demonstrated for different pregnancy trimesters, utilizing two-dimensional ultrasound images.

eess.IV · cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12374

TANSAR: Editable and Attributable Correlation Matrix Memories under Correlated Keys, with Rule-Constrained Multi-Hop Knowledge Editing

Abdulhady Abas Abdulla, Jiyan Salim Mahmud, Fatemeh Daneshfar, Hussein Mohammed Ali, Mourad Oussalah, Muhammad Bilal

原文摘要

Correlation Matrix Memories (CMMs) store associations as sums of rank-one outer products, providing closed-form addressability, exact deletion, and linear attribution. While classical analyses assume random, orthogonal keys, semantic encoders produce correlated representations. We analyse CMMs under correlated keys using frame-theoretic coherence: (i) capacity bounds yielding an encoder-independent limit $n \le d(1+\varepsilon^2)$ for mean squared crosstalk $\varepsilon^2$; (ii) an exact characterisation of the collateral effect of deleting an association, guaranteeing that deletion never invalidates certified associations; and (iii) a geometric transparency bound demonstrating why attribution requires candidate sets of centred keys. On this basis, we introduce Topology-Aware Neuro-Symbolic Associative Retrieval (TANSAR). TANSAR traverses a directed acyclic association graph, prunes invalid paths via sound ground Horn constraints, and injects a representation derived strictly from the verified path into the residual stream of a frozen language model, guarded by a Clopper--Pearson selective-risk gate. Controlled experiments confirm all theoretical bounds. In multi-hop knowledge editing, Horn pruning reduces answer contradictions from 25.1% to 0.5%, whereas Hopfield retrieval incurs 24% contradictions when stale associations remain stored (and 0% if stale associations are removed). An end-to-end edit transaction requires 13.6 ms on a standard CPU and 2.37 ms on GPU. The symbolic component matches gold labels across all 600 ProofWriter theories and achieves 94.0% agreement on FOLIO, demonstrating sound and efficient multi-hop associative retrieval with exact unlearning.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2505.24622

Rule Forest: Interpretable Ensembles of LLM-Generated Questions for Rare-Outcome Prediction

Ben Griffin, Aaron Ontoyin Yin, Diego Vidaurre, Ugur Koyluoglu, Joseph Ternasky, Fuat Alican, Yigit Ihlamur

原文摘要

Many high-stakes screening decisions in finance amount to predicting a rare outcome from unstructured text, such as whether a founder will successfully build a large company, or which crowdfunded loan will fail to attract funding. When positives are this rare, simple models may be preferable to more complex ones. For example, a model that learns a weight for each feature may have too little signal to estimate those weights reliably, and also inspecting the model means reading a list of fitted coefficients rather than a simple rule that can be checked by eye. We introduce Rule Forest, which uses a large language model to generate simple YES/NO questions, rather than using it to directly predict outcomes, and combines the strongest questions into an equal-weight "green-flags" scorecard. This scorecard is as accurate as an equivalent model that learns a weight per question, within noise, and its advantage appears largest when positives are rarest. The result is an interpretable list of rules that a domain expert can edit and extend. When predicting founder success, the fraction of shortlisted founders who succeed increases from about 1 in 50 to 1 in 10, and to about 1 in 6 with a handful of expert-written questions. A second dataset which focuses on identifying failed loans, Kiva microloans, shows the same qualitative pattern. Across both tasks, the lesson is the same: a language model is most useful not as the predictor but as a generator of features a person can read, edit, and improve, and when positives are scarce that interpretability costs no measurable accuracy.

cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12370

Adapting Multi-Stage Distributed Real-Time Systems Design to Vehicular Edge Computing

Niraj Kumar, Anant Gautam

原文摘要

Vehicular Edge Computing (VEC) aims to support computation-intensive applications in intelligent transportation systems while meeting stringent latency requirements. When processed remotely, applications require access to different resource types in a specific sequence, forming a multi-stage processing pipeline. This structure closely resembles multi-stage distributed applications executing across heterogeneous resources, making established multi-stage distributed real-time scheduling techniques well suited for modeling VEC systems. The resulting model accommodates diverse resource configurations and access patterns. Building on this model, we propose a joint mapping and scheduling approach that accounts for vehicle mobility and real-time constraints. The proposed heuristic extends optimal priority assignment for multi-stage systems to the VEC setting, to construct feasible schedules while minimizing local-device energy consumption and ensuring that offloaded jobs meet their deadlines. Extensive simulations show that the proposed approach consistently outperforms both a baseline based on the same delay-composition framework and a deadline decomposition-based approach.

cs.DC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12367

An Agile Homomorphic Encryption and ML-KEM Scheme for Privacy-Preserving Health Data Processing

Daniel Resende, Nadia El Mrabet, Ulrich K{ü}hne

原文摘要

The storage and processing of health data raise serious privacy concerns; therefore, the data must be encrypted. Fully Homomorphic Encryption (FHE) allows computations on encrypted data, which makes it a very promising technology in this context. However, although the efficiency of FHE algorithms has improved in recent years, they remain very complex. The security of several prominent FHE schemes is based on variants of the Learning With Errors (LWE) problem, which is also the basis of Module-Lattice-Based Key-Encapsulation Mechanism (ML-KEM), the new post-quantum standard for asymmetric encryption. In this paper, we propose an agile cryptographic scheme -combining ML-KEM and BGV -that allows efficient implementations in an embedded use case. We describe the parameter selection for the proposed scheme and show its feasibility with a proof-of-concept implementation.

cs.CR

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2602.03777

Statically Verifying Well-Definedness of Separately Compiled Attribute Grammars

Federico Bruzzone, Walter Cazzola, Luca Favalli

原文摘要

The development of programming languages involves complex theoretical and practical challenges, particularly when addressing modularity and reusability through language extensions. While language workbenches aim to enable modular development under the constraints of the language extension problem, one critical constraint--separate compilation--is often relaxed due to its complexity. However, this relaxation undermines artifact reusability and integration with common dependency systems. A key difficulty under separate compilation arises from managing attribute grammars, as extensions may introduce new attributes that invalidate previously generated abstract syntax tree structures. Existing approaches, such as the use of dynamic maps in the Neverlang workbench, favor flexibility at the cost of compile-time correctness, leading to potential runtime errors due to undefined attributes. This work addresses this issue by introducing nlgcheck, a theoretically sound static analysis tool based on data-flow analysis for the Neverlang language workbench, which detects potential runtime errors--such as undefined attribute accesses--when separately compiled language components are composed, before the resulting language is ever run, thus preserving separate compilation while maintaining strong static correctness guarantees. Experimental evaluation using mutation testing on Neverlang-based projects demonstrates that nlgcheck effectively enhances robustness without sacrificing modularity or flexibility and with a level of performance that does not impede its adoption in daily development activities.

cs.PL · cs.SE

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.03043

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

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

原文摘要

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

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12364

TRIAD: Multimodal Retrieval across Abstraction, Constraints, and State

Xiangyang Wu, Chen Ju, Xu Chen, Likang Peng, Jinsong Lan, Haihong Tang, Yang Hu

原文摘要

Multimodal retrieval increasingly supports agentic systems, where user requests can involve cross-level relations, combine multiple constraints, and evolve across dialogue turns. In these settings, a highly similar result may still violate the user's requirements, while retrieved content may directly inform downstream actions without human screening. We introduce TRIAD, a multimodal retrieval benchmark for evaluating how well retrieval systems satisfy these complex, evolving requests. TRIAD comprises three tracks: abstraction, testing bidirectional retrieval between visual instances and concepts at different hierarchical levels; constraints, testing the joint satisfaction of positive and negative conditions; and state, testing the integration of requirements across dialogue turns and the updating of earlier conditions when users revise their requests. The benchmark contains six subsets, each with 1,000 queries and a fixed pool of 1,000 candidates per query, evaluated using a common multi-positive mean average precision (mAP) protocol. Simpler reference tasks and hard negatives grouped by the requirements they violate help characterize specific retrieval difficulties. We evaluate 20 configurations of 17 systems and find that no system leads across all six subsets. Performance declines with hierarchical distance, and negative-only queries generally underperform positive-only queries with the same number of constraints. Across paired tasks expressing the same final intent, multi-turn retrieval yields lower mAP in 39 of 40 configuration-domain comparisons. The benefits of explicit reasoning also vary across models and tasks. These findings highlight the gap between retrieving similar content and satisfying a complete request.

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12362

Release Sooner, Expose Less: Offloading MoE Dispatch in NCCL-EP under Dual-Batch Overlap

Hsin-Chun Yin, Khaled Hamidouche, Ian Ziemba, Arjun Kashyap, Pak Markthub, Subhadeep Bhattacharya, Ke Wen, Assaf Ravid 等 10 位作者

原文摘要

Dual-batch overlap (DBO) hides mixture-of-experts (MoE) communication behind another microbatch's computation, but only after dispatch send releases the GPU. The time dispatch send holds the GPU therefore stays exposed, while network progress after release can be hidden. We evaluate three research configurations built on NCCL-EP that target this exposed communication time: GPI, in which the GPU constructs GPU-Friendly Descriptors (GFDs) and submits them to NIC-accessible memory for transport; GPI-MC, which describes MoE token fan-out in one GFD; and GDAKI Free Flow, which moves send progression to a host thread. On a GB300 NVL72 system, all three approaches reduce dispatch-send time relative to default GDAKI over most operating points in the evaluated parameter sweep. An experiment that inserts a GPU delay between dispatch send and receive shows that offload increases the in-flight interval. Depending on the configuration, 32 to 512 microseconds of computation or other independent work between send and receive hides the deferred progress, leaving less exposed time than default GDAKI. In disaggregated vLLM serving with DBO, these approaches lower median TPOT by 15-25% relative to NCCL-EP default GDAKI and by 16-42% relative to DeepEP v1.

cs.DC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12361

Law-Invariant Operators for Stochastic Mechanics

Vissarion Papadopoulos

原文摘要

A central objective in stochastic mechanics is to separate deterministic physical behavior from the probability laws governing uncertain inputs. While classical spectral approaches achieve this separation in linear settings, nonlinear responses introduce interactions beyond conventional second-order variability response functions (VRF). Inspired by the spirit of VRFs, this work establishes that, for a fixed physical system, uncertainty propagation is described by a law-invariant operator mapping input realizations to responses, governed by the underlying physics and the deterministic parameters of the system. Different probability laws affect the input sampling and resulting response statistics, but not the mapping operator. We establish this law invariance under complete, invertible coordinate transformations and develop complementary, Fourier, fixed-reference polynomial-chaos (PCE) and finite element method (FEM) transfer formulations. Fourier captures nonlinear cross-frequency interactions and recovers the classical VRF in the appropriate linear limit, PCE describes coupling among stochastic modes, while FEM expresses the deterministic inverse-stiffness action. Neural approximations of these deterministic operators are investigated through analytical and finite-element benchmarks, examining cross-law generalization, the influence of training distributions and stochastic excitation, and the prediction of response fields and statistics. The results demonstrate the potential for reusing learned physical operators across different stochastic laws without retraining, while highlighting the importance of adequate training coverage and the challenges of accurately estimating low-probability events.

stat.CO · cs.CE · math.NA

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2605.13282

Differentiable Learning of Lifted Action Schemas for Classical Planning

Jonas Reiter, Jakob Elias Gebler, Hector Geffner

原文摘要

Classical planners can effectively solve very large deterministic MDPs represented in STRIPS or PDDL where states are sets of atoms over objects and relations, and lifted action schemas add or delete these atoms. This compact representation yields strong search heuristics and provides an ideal setting for structural generalization, since lifted relations and action schemas give rise to infinitely many domain instances. A central challenge is to learn these relations and action schemas from data, and recent approaches have addressed this problem using different types of observations. In this work, we develop a novel neural network architecture for learning action schemas from traces where states are fully observed but action arguments are unobserved. The problem is a simplification but an important step towards learning planning domains from sequences of images and action labels, and we aim to solve this simplification in a nearly perfect manner. The challenge lies in learning the action schemas while simultaneously identifying the action arguments from observed state changes. Our approach yields a robust differentiable component that can then be integrated into larger neuro-symbolic models. We evaluate the architecture on various planning domains, where the learned lifted action schemas must recover the ground-truth structure. Additionally, we report experiments on robustness to observation noise, on component ablations, and on a variation related to slot-based dynamics models.

cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12356

DForkTree: Accelerating Speculative Decoding with Fork-Aware Block Diffusion Draft Trees

Zhicheng Zheng, Cheng Zhang, Chenkai Tang

原文摘要

Speculative decoding speeds up language models by letting a lightweight drafter propose several future tokens that the target model verifies in parallel. Block diffusion drafters such as DFlash draft a whole block in one forward pass, and DDTree builds a draft tree from their per-position distributions. Each of these distributions assumes that the drafter's own most likely continuation is correct. A tree, however, gains over a chain only on branches that leave this continuation, where the drafter's top-1 accuracy drops from 89% to 18% and its candidate lists miss the right token far more often. When re-run from the fork token, the same drafter is usually correct, so the errors come from the prefix that the distributions condition on rather than from what the drafter knows. We introduce DForkTree, a training-free method that re-anchors each branch on its own tokens. DForkTree adds n-gram statistics from the target model's own outputs to the drafter's scores as pointwise mutual information, weights them more after the fork, and builds the optimal tree under these scores exactly on the GPU. On Qwen3.5-4B and 9B, with all tree methods sharing one drafter, tree pipeline and verifier, DForkTree is the fastest tree method on every benchmark. It is up to 41.5% faster than DFlash and 15.1% faster than DDTree in the same setting, with average speedups of up to 4.75x over autoregressive decoding.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.32313

MemTransfer: Benchmarking Memory Beyond Matched Experience in Embodied Decision-Making

Haiming Tang, Xianjie Dai, Gujie Shao, Zuyi Guo, Jingguang Li, Kailang Ma, Yihong Tang, Heye Huang

原文摘要

Memory lets an embodied agent reuse past experience, yet retaining useful information does not ensure that the agent can apply it when conditions change. We present MemTransfer, a benchmark comparing six memory representations, a working-memory baseline and five representations of past experience, under a shared frozen vision-language-model policy. It comprises 100 navigation cases across ten task types in a simulated warehouse, with expert demonstrations supplying the history. Three comparisons vary the starting pose, route availability, and amount and task relevance of history. With one demonstration per task, Full-context and Episodic memory reach 95.3% and 100.0% success at the original demonstration start, but lose 48-49 percentage points at a new test start. Summary changes little between these two test starts, yet with four demonstrations per task it retains a smaller fraction of its unchanged-route success after blocking (39.3%) than Working memory (44.8%) or the two trajectory memories (56-58%). At the new test start, increasing from one to four relevant demonstrations raises Episodic success by 14.3 percentage points, while the other evaluated representations gain no more than 1.3 percentage points. Replacing half of the relevant histories with other-task experience lowers success for both trajectory memories. These results show that robustness to one kind of mismatch does not imply robustness to another, motivating evaluation of both stored information and its use at decision time.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2606.15912

On-Policy Distillation with Curriculum Turn-level Guidance for Multi-turn Agents

Gengsheng Li, Mao Zheng, Mingyang Song, Ruiqi Liu, Tianyu Yang, Jie Sun, Qiyong Zhong, Haiyun Guo 等 11 位作者

原文摘要

Multi-turn agents that plan, invoke tools, and interact with environments offer a promising paradigm for solving complex tasks, yet their capabilities typically rely on very large models whose inference cost is prohibitive in practice. On-Policy Distillation (OPD) is a natural recipe for transferring such capabilities to smaller students, but we find that it suffers a characteristic failure mode in this setting: small student errors compound across turns and push the trajectory out of the teacher's familiar state distribution, so the teacher's supervision becomes least reliable precisely where the student needs it most. We propose Guided On-Policy Distillation (Guided-OPD), a simple yet effective algorithm that mixes teacher- and student-generated turns within each rollout and schedules the teacher's intervention probability along a curriculum that decays to zero. Strong guidance keeps early trajectories close to the teacher distribution and is then gradually withdrawn to recover the purely on-policy regime used at inference. On ALFWorld, ScienceWorld, and WebShop, distilling Qwen3 students from a Qwen3-30B-A3B teacher, Guided-OPD yields average relative gains of 21.1\% in Score and 25.5\% in Success Rate over vanilla OPD, with larger gains on smaller students.

cs.LG · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12351

UniOMA: Unified Optimal-Transport Multi-Modal Structural Alignment for Robot Perception

Xinrui Zu, Mark Hoogendoorn, Kevin Sebastian Luck, Shujian Yu

原文摘要

Achieving generalizable and well-aligned multimodal representations remains a core challenge in robot learning. While recent approaches have attempted to align modalities by modeling conditional or higher-order statistical dependencies, they do not explicitly consider the structural coherence across modalities. In this work, we propose a novel multimodal alignment method that augments existing contrastive losses with a geometry-aware Gromov--Wasserstein (GW) distance-based regularization. To this end, we encode intra-modal geometry for each modality and reduce cross-modal structural discrepancies through a dynamically learned GW barycenter, thereby encouraging structural consistency in addition to instance-level contrastive alignment. We apply this optimal-transport-based alignment strategy to robot perception tasks involving underexplored modalities such as force and tactile signals, where modality data often exhibit varying sample densities. Extensive experiments on five robotic benchmarks show that our method improves structural alignment and downstream performance over state-of-the-art multimodal contrastive baselines across a wide range of downstream tasks, including robot and environment state prediction. Moreover, our GW-based regularization can be seamlessly integrated into most InfoNCE-like objectives.

cs.RO · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2505.14260

Speculative Decoding Reimagined for Multimodal Large Language Models

Luxi Lin, Zhihang Lin, Zhanpeng Zeng, Rongrong Ji

原文摘要

Multimodal Large Language Models (MLLMs) deliver strong performance across vision-language tasks but suffer from high autoregressive decoding latency. Although speculative decoding can accelerate Large Language Models (LLMs) without changing their output distribution, directly applying existing approaches to MLLMs yields limited speedups because text and visual tokens exhibit distinct characteristics. We present Multimodal Speculative Decoding (MSD), a framework specifically designed for efficient MLLM inference. MSD is motivated by two key observations: text and visual tokens should be handled separately during drafting, and an effective draft model requires both strong language modeling and visual perception capabilities. Accordingly, MSD decouples text and visual token processing in the draft model and applies token-specific strategies to improve drafting efficiency. We further introduce a two-stage training strategy. The draft model is first trained on text-only instruction-tuning data to strengthen its language modeling ability and is then progressively exposed to multimodal data to develop visual perception while preserving its language competence. Extensive experiments across multimodal benchmarks demonstrate that MSD achieves up to $2.29\times$ inference speedup for LLaVA-1.5-7B and $2.46\times$ for LLaVA-1.5-13B without sacrificing generation quality. These results establish MSD as an effective and general approach for accelerating multimodal autoregressive generation. Our code is available at https://github.com/Lyn-Lucy/MSD

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12348

How Hard Is Tokenisation? Algorithms and Complexity Beyond NP-Hardness

Andrei Robert Ion, Rares Mateiu, Adrian Miclaus, Alexandru Popa

原文摘要

Given a collection of strings and a vocabulary budget, how well can we minimize its total token count, and how does the answer change with the tokenisation model? For binary direct tokenisation, previous work proved APX-hardness but left constant-factor approximability unresolved \citep{kastreva2026}. We rule out every polynomial-time $M^{1-\varepsilon}$ approximation, for fixed $\varepsilon>0$, unless $\mathrm{P}=\mathrm{NP}$, even on distinct equal-length strings with $K=m-1$. Here $M$ is the total input length and $m$ is the number of input strings. We complement this lower bound with a deterministic polynomial-time $O(M/\log M)$ approximation algorithm and stronger bounds for selected vocabulary budgets. For the different objective of maximizing compression savings, GreedTok was shown to attain a constant fraction of the optimum on unrestricted alphabets \citep{baliga2026}; we establish a deterministic EPTAS for direct savings on every fixed alphabet and analyze greedy selection by true marginal gains. For exact optimization, we give a binary direct algorithm running in $2^{O(n\log\log n/\log n)}\poly(I)$ time, where $n$ is the total length of the distinct input strings and $I$ is the input encoding length. Finally, we resolve two unary merge-model hardness questions left by the bounded-alphabet study \citep{kastreva2026}: standard unary bottom-up tokenisation is APX-hard, and unary optimal pair encoding is strongly NP-complete, strengthening the previously known weak hardness.

cs.DS

查看论文
arXiv ↗PDF ↗