跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+0新增论文
+0其中 AI 相关
+0已有论文修订
+92新增 AI 导读

arXiv 最近完成扫描:2026-11-23 01:07 · 全站最近入库:2026-11-21 12:19各信源进度 →

统计口径与数据新鲜度

统计 2026-11-22 04:28 — 2026-11-23 04:28(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:2026-11-23 01:07最近发现新增版本:2026-11-21 12:19库内论文最新更新时间:2026-11-20 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。

29,158 已收录论文24,181 AI 相关8,992 有修订112 HF 社区发现

浏览论文库

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

找到 18 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-23 01:05(北京时间)
版本更新 · v3arXiv:2610.12830

Cross-Behavior Policy Optimization: Unified Learning from Offline Demonstrations and On- and Off-Policy Rollouts via a Shared Regression Target

Du Cheng, Junyu Lu, Jingfan Zhang, Haoxuan Jia, Shichao Weng

原文摘要

Reasoning post-training draws on offline demonstrations and policy-generated rollouts, yet conventional supervised fine-tuning (SFT) followed by reinforcement learning (RL) separates their objectives and limits continued reuse of collected responses. We propose \textbf{Cross-Behavior Policy Optimization} (CBPO), which makes these sources compatible with one regression loss through a fixed-reference target. Because the generating policy does not enter this target, positive and negative offline completions, reference responses, and current and historical rollouts can share a common reward-labeled record format and replay buffer. As new responses enter the buffer, optimization retains the same regression objective without coordinating separate SFT and RL objectives or requiring stored behavior-policy probabilities, importance weighting, or a learned critic. Restricting the available sources gives offline, on-policy, and off-policy variants, while CBPO-Unified combines them. Across two model scales and four mathematical reasoning benchmarks, CBPO delivers leading \texttt{avg@32} and \texttt{pass@32} performance across benchmarks and data regimes. On Qwen3-1.7B, CBPO-Unified improves benchmark-averaged \texttt{avg@32} and \texttt{pass@32} over SFT DAPO by 2.33 and 4.44 percentage points, respectively. In the replay ablation, retaining historical rollouts improves these metrics by 1.87 and 3.72 percentage points over the no-history configuration.

cs.LG

新论文 · v1arXiv:2611.21029

Sparse Attention for Visual Generation: Toward Algorithm-System Co-Design

Majun Zhang, Yuan Guo, Decheng Wu, Jiayang Xu, Changhao Pan, Fan Zhuo, Xiaoda Yang, Chenyuhao Wen 等 9 位作者

原文摘要

Rapid advances in image and video generation have enabled high-fidelity synthesis at higher resolutions and longer durations, making the quadratic cost of dense self-attention a major bottleneck as visual token sequences grow. Sparse attention seeks to preserve generation quality at lower computational cost by retaining informative query-key interactions and omitting less consequential ones, with end-to-end gains depending on the cost of constructing sparse support and the efficiency with which the system maps the resulting workload to hardware. We survey sparse attention for visual generation through a 4+1+3 analytical framework that links algorithmic design to system implementation. The framework examines how static and dynamic methods use source information and the evidence derived from it to decide which interactions to retain and at what granularity, while accounting for the roles of reuse in reducing repeated computation and correction in compensating for omitted contributions. We relate these algorithmic choices to their implementation on hardware through the framework's three system dimensions of data organization, execution, and dataflow. We summarize common evaluation metrics and discuss open challenges and research directions, providing researchers and practitioners with a reference for the joint design of sparse attention algorithms and systems.

cs.CV

新论文 · v1arXiv:2611.19864

Local Order Matters: A Lightweight CPU-Trainable CNN for Website Fingerprinting on Defended Tor Traffic

Puyang Xu, Zhizhao Wen, Yifan Jiang, Chao Wang, Yusi Gong

原文摘要

Website fingerprinting (WF) attacks let a passive adversary infer which site a Tor user visits from the encrypted cell sequence. On undefended traffic, hand-crafted-feature attacks are already near-ceiling, so deep learning offers only a marginal advantage there. We show that the decisive regime is defended traffic: against the FRONT padding defense, which injects dummy cells that corrupt the volume and count features feature attacks rely on, a compact 95,103-parameter 1-D convolutional network trained on the raw cell-direction sequence, on CPU only, stays robust while three implemented direction-only feature baselines degrade. On the closed-world WFlib FRONT set (95 sites), our model reaches 54.2% top-1 accuracy versus 38.5% for the strongest implemented feature baseline (k-FP-style RF), a gain of 15.7 percentage points (3 seeds; seed-0 paired-bootstrap 95% CI [+14.5, +17.0] pp on per-trace correctness, excluding zero). On undefended traffic the CNN reaches 94.1% versus 92.3% for k-FP-style RF. A controlled order-ablation permutes only nonzero traffic cells within each trace, preserving direction counts, retained length, and trailing zero-padding positions while destroying traffic order. This collapses the CNN below every feature baseline (from 54.2% to 7.6%, a 46.6 pp drop), showing that the model depends strongly on order-derived information under FRONT.

cs.CR · cs.AI · cs.LG

新论文 · v1arXiv:2611.16211

One Fuzzer, Many Reasoners: Adaptive Neuro-Symbolic Fuzzing with Reliability-Calibrated and Context-Gated Guidance

Jinghao Wen, Xianglong Yu, Shaohuang Wang, Xun Jiao

原文摘要

Neural-symbolic AI (NSAI) combines learned perception with programs, constraints, and compositional reasoning, but remains difficult to test because small semantics-preserving changes can alter grounded symbols and trigger discontinuous reasoning failures. Existing neural fuzzers mainly rely on neural coverage or representation-space feedback and do not provide a common testing interface for heterogeneous NSAI systems. We introduce NSAI-Fuzz, a unified semantics-aware fuzzing framework that separates task-specific valid mutation from architecture-neutral guided search and represents each candidate using five availability-aware, rank-normalized risk signals: boundary pressure, grounding uncertainty, representation drift, constraint violation, and archive novelty. We further propose Reliability-Calibrated Weighting (RCW), which learns dataset-model-level signal allocations from calibrated reliability, and Contextual Signal Gating (CSG), which adapts signal weights to the current candidate pool. We evaluate four NSAI systems across four reasoning tasks, nine compatible dataset-model cells, 20 policies, and 1,800 on-policy runs. Fixed five-signal guidance outperforms Random and Uniform in every cell, while RCW and CSG further improve over Fixed in all nine cells. Under matched mutation and query budgets, RCW and CSG achieve 16.17% and 16.06% aggregate failure discovery, compared with 6.06% for the strongest neural-fuzzing baseline. Wall-clock evaluation further shows that TensorFuzz requires 1.11x the runtime of RCW while achieving only 0.45x its failure throughput. These results show that effective and efficient fuzzing of heterogeneous NSAI systems benefits from a common neural-symbolic risk interface with adaptive signal allocation.

cs.AI · cs.LG

新论文 · v1arXiv:2611.15276

Verify by Doing: Co-Evolving Agent Execution and Evaluation without External Feedback

Yitong Lin, Jianda Chen, Yihao Hu, Pan Wang, Hang Wang, Zhihao Wen, Xiaoxue Ren, Wei Wu

原文摘要

Agent self-improvement requires feedback about task success, yet external feedback may be unavailable at test time. Agents must then infer success from observable evidence, and incomplete verification can misguide subsequent updates. We introduce Verify by Doing (VbD), a method that co-evolves Actor and Evaluator harnesses within a single task while keeping model weights fixed. The Evaluator actively inspects artifacts and environment state with tools; its findings ground an internal signal for output selection and guide updates to both execution and verification procedures. Proposed Evaluators undergo comparative review on common candidates, while Actor updates require support from their own fresh outputs. Across AutomationBench, $τ^3$ Banking, and selected GDPval tasks, VbD achieves the highest overall scores among the compared methods operating without external feedback, with gains of 3.3, 4.1, and 2.3 percentage points over the strongest baselines on the main backbone. Ablations support the contributions of active verification and Evaluator evolution, and selection from fixed candidate pools shows that improved verification can recover successful outputs without generating new ones.

cs.AI

新论文 · v1arXiv:2611.13254

Dependable Long-Horizon LLM Agents: A Trajectory-Centered Empirical Study of Where Errors Enter and Which Controls Reach Them

Shaohuang Wang, Jinghao Wen, Xulei Zhang, Xun Jiao

原文摘要

When an LLM agent fails part-way through a long task, the usual analysis records a symptom: a failure-mode label, or the agent and step judged responsible. It does not say where the error entered, which state carried it to the symptom, whether anything in the trajectory contradicted it, or which control could have caught it. We define an episode record that ties a candidate fault to its origin, propagation, detection, response and consequence, located by operational phase and state domain, and that names the control families whose observation and reach cover it. We turned the record into a written annotation protocol and applied it to 95 failed runs of six multi-agent systems from the public Who&When and MAD corpora, reading each trajectory in full. Faults enter early (median at 0.18 of the run), mostly as wrong decisions or fabrications over task state, and in 37 of 88 Who&When runs before the step the corpus' human labels mark as decisive. In 72% of failed runs the trajectory already contained evidence against the error before the failure; in 88% of those no agent acted on it. Something flagged a deviation in 31% of runs, but no response removed the deviated state. Independent judges reading the unused evidence, task contracts fixed at admission, and re-checking beliefs against the environment reach most episodes; restoring the agent's own state would have sufficed in 35 of 95 runs. A text scan of the raw corpus also shows that 45% of the hand-crafted Who&When failures ended with a provider content-filter crash after which the harness emitted a final answer no agent produced, a failure class that neither symptom taxonomies nor responsible-agent labels represent. Protocol, records and scripts are included.

cs.AI

新论文 · v1arXiv:2611.11125

LongCap: Hour-Scale Timestamped Audio Captioning via Chunk-wise Processing

Jinchuan Tian, Sen Chen, Wenbin Huang, Chao Weng, Zhengwei Huang, Jiatong Shi, Yichen Lu, Haoran Wang 等 10 位作者

原文摘要

Describing hours of audio requires accurate event timing and sustained descriptive detail throughout the recording. We present LongCap, a framework for timestamped audio captioning (TAC), with the human-annotated LongCapBench and three diarization-inspired metrics: Timeline Error Rate (TER), Granularity Error Rate (GER), and Caption Similarity (tcpCapSim). LongCap interleaves audio chunks with structured event captions, using relative timestamps and retained context to support length extrapolation within a 1M-token inference context window. Event-level reannotation restores caption detail, while systematic studies examine model scaling, dependency supervision, and self-distillation. LongCap-Diar achieves strong benchmark results and extrapolates to 16 hours of concatenated audio despite training on recordings no longer than 40 minutes. Our TAC models are competitive with Qwen3.5-Omni-Plus on short clips, outperform evaluated baselines in caption similarity and timeline accuracy in segment-level comparisons on LongCapBench-TAC-Long, and caption full recordings approaching two hours. On LongCapBench-TAC, 9B and 35B-A3B offer favorable size-performance tradeoffs, while 122B-A10B substantially improves event-dependency recovery. These findings support long-horizon general audio understanding and provide a foundation for always-on voice agents.

cs.CL · cs.SD

新论文 · v1arXiv:2611.11116

Query-Only Prompt-Agnostic Attacks on Visual In-Context Learning

Xiao Hu, Muhammad Ibrahim, Hao Wen, Pin-Yu Chen, Jianliang Wu, Antonio Bianchi

原文摘要

Visual In-Context Learning (VICL) enables promptable segmentation by conditioning RGB predictions on visual demonstrations, making it attractive for auto-labeling, but its robustness under query-only access and unknown deployment prompts remains underexplored. We study query-only, prompt-agnostic targeted attacks where the attacker observes only returned RGB outputs, without access to model internals or deployment prompts. Treating prompt-conditioned RGB rendering as a distinct attack surface, we introduce two appearance-space attacks: Color Hijack (CH) drives foreground colors toward the background to suppress visibility, while Task Hijack (TH) collapses instance-distinguishing colors toward a shared color assignment. To transfer across unseen prompts, we propose a Self-Prompt Surrogate (SPS) constructed from the clean query and its output. In a paired, matched-budget SegGPT comparison, SPS outperforms the Random-Prompt Surrogate (RPS), reducing CH instance foreground retention from 0.918 to 0.155 and TH palette-aware tolerance distance from 0.345 to 0.225. Across four VICL models on COCO and HierText, CH suppresses foregrounds across semantic-, instance-, and text-style outputs, while TH collapses instance colors under held-out deployment prompts. After palette-based decoding, TH also reduces the decoded instance count and mask retention while increasing instance merges. Finally, at 30% corruption, CH-corrupted annotations from a publicly accessible VICL-based auto-labeling pipeline reduce downstream YOLO-seg mAP50 from 0.761 to 0.281.

cs.CV

版本更新 · v2arXiv:2605.06978

Group of Skills: Group-Structured Skill Retrieval for Agent Skill Libraries

Kun Zeng, Yu Huo, Siyu Zhang, Haoyue Liu, Yuecheng Zhuo, Zi Ye, Yuquan Lu, Junhao Wen 等 9 位作者

原文摘要

Skill-augmented agents increasingly rely on large reusable skill libraries, but retrieving relevant skills is not the same as presenting usable context. Existing methods typically return atomic skills or dependency-aware bundles whose internal roles remain implicit, leaving the agent to infer the execution entry point, support skills, visible requirements, and failure-avoidance guidance. We introduce Group of Skills (GoSkills), an inference-time group-structured retrieval method that changes the agent-facing retrieval object from a flat skill list to a compact, role-labeled execution context. GoSkills builds anchor-centered skill groups from a typed skill graph, expands support groups through a group graph, bottlenecks the selected group plan into a bounded set of atomic skill payloads, and renders a fixed execution contract with Start, Support, Check, and Avoid fields, without changing the downstream agent, skill payloads, or execution environment. Across SkillsBench, ALFWorld, and Terminal-Bench 2 with six backbones, GoSkills attains the highest mean reward or accuracy among the evaluated methods in every benchmark-backbone setting, and matched controls show higher reward and fewer agent turns with group selection than with fixed-contract or post-hoc role-annotation alternatives.

cs.CL · cs.AI

版本更新 · v2arXiv:2605.06161

Are LLM Safety Judges Policy-Invariant? A Three-Principle Stress-Test

Shihao Weng, Yang Feng, Xiaofei Xie

原文摘要

LLM-as-a-Judge pipelines have become the de facto evaluator for agent safety, yet existing benchmarks treat their verdicts as ground-truth proxies without checking whether the verdicts depend on the agent's behavior or merely on how the evaluation policy happens to be worded. We argue that any trustworthy safety judge must satisfy a basic property we call policy invariance, and we operationalize it as three testable principles: rubric-semantics invariance under certified-equivalent rewrites, rubric-threshold invariance under intentional strict-to-lenient shifts, and ambiguity-aware calibration so that verdict instability concentrates on genuinely ambiguous cases. Instantiating these principles as a stress-test protocol with four agent-class judges on trajectories drawn from ASSEBench and R-Judge, we surface a previously unmeasured failure mode: today's judges respond to meaningful normative shifts and to meaningless structural rewrites with comparable strength, and cannot tell the two apart. Content-preserving policy rewrites flip up to 9.1% of verdicts above baseline jitter, and 18-43% of all observed flips occur on unambiguous cases under such rewrites, so safety scores produced by these judges partly conflate what the agent did with how the evaluator was prompted. These findings persist with a frontier-scale judge, a third benchmark, and a second base policy. Beyond the diagnosis, we contribute the Policy Invariance Score and the Judge Card reporting protocol, which expose an order-of-magnitude spread in judge reliability that is invisible to accuracy-only leaderboards. We release the protocol and code so that future agent-safety benchmarks can audit their own evaluators rather than trust them by default: https://github.com/wengshihao/policy-invariance-judge

cs.AI · cs.SE

新论文 · v1arXiv:2611.05937

Deep Meets Classical: A CPU-Efficient Decision-Fusion Model for Automatic Modulation Classification

Zhiyi Yin, Zhizhao Wen, Ruoxin Zhang, Chao Wang, Puyang Xu, Yifan Jiang, Yusi Gong

原文摘要

Automatic modulation classification (AMC) underpins spectrum sensing and non-cooperative signal identification in cognitive radio, yet the two standard CPU-feasible approaches to it, a compact convolutional neural network (CNN) on raw in-phase and quadrature (I/Q) samples and a classical higher-order-cumulant feature classifier, are rarely compared against each other conditioned on signal-to-noise ratio (SNR). We show that they are, in fact, complementary: the CNN is markedly more robust at low SNR, where hand-crafted statistics collapse toward chance, while the cumulant classifier is more accurate at high SNR, where higher-order statistics cleanly separate dense constellations. We exploit this complementarity with FuseAMC, a lightweight decision-fusion model that trains a logistic-regression meta-combiner over the concatenated class-posteriors of an approximately 198k-parameter residual CNN and a gradient-boosted-tree cumulant classifier, on a validation split disjoint from test. On RadioML 2016.10a (5 seeds, 44k-frame held-out test set), FuseAMC reaches 55.8% overall accuracy, improving on the best cumulant-feature baseline by 5.1 points and on the CNN alone by 3.2 points (paired-bootstrap 95% confidence-interval lower bounds of 5.0 and 3.1 points, both positive), while matching the best classical baseline in its own strong high-SNR regime and the CNN in its strong low-SNR regime. With about 0.2 million parameters, a model footprint of a few megabytes, and roughly 3 MFLOPs per frame, the whole pipeline trains and runs on one CPU core, targeting general-purpose and embedded processors without hardware acceleration.

cs.LG · cs.AI

新论文 · v1arXiv:2611.04273

Deep Learning Compilers: Architecture, Optimization, and Empirical Characterization

Jinghao Wen, Ruixuan Wang, Xun Jiao

原文摘要

As deep learning (DL) models become increasingly complex and hardware architectures grow more heterogeneous, DL compilers play a critical role in optimizing computational efficiency and enabling scalable deployment. This paper presents a systematic analysis of DL compilers, covering their architectural evolution, compilation workflows, optimization techniques, and emerging research directions. We organize DL compiler architectures into frontend, middle-end, and backend stages, emphasizing the role of the middle-end in connecting high-level model representations with hardware-specific optimizations. We further examine representative DL compilers and recent optimization techniques across different compilation stages. Beyond architectural and technical analysis, we conduct an empirical evaluation of compiler behaviors to investigate the effectiveness of graph optimization, compilation overhead, and dynamic-shape handling. Experiments using PyTorch/TorchInductor on an NVIDIA RTX A6000 demonstrate that compiler benefits depend strongly on workload characteristics, with speedups of up to 8.21\times for fusion-friendly computation graphs, while other workloads exhibit limited improvements. Our results further reveal that compilation overhead can require thousands of executions to amortize, and dynamic-shape compilation reduces cumulative shape-transition costs by approximately 70\% at the expense of modest steady-state performance degradation. By combining systematic architectural analysis with quantitative experimental evidence, this work provides insights into the design, optimization, and practical deployment of modern DL compilers.

cs.LG · cs.AR

版本更新 · v2arXiv:2605.24426

SEAL: Synergistic Co-Evolution of Agents and Learning Environments

Yihao Hu, Zhihao Wen, Pan Wang, Xiujin Liu, Yitong Lin, Xin Zhang, Wei Wu

原文摘要

Large language model (LLM) agents increasingly learn through interaction, yet their training process often evolves unevenly: the policy changes with experience while the learning environment continues to provide largely fixed forms of support. As training progresses, this creates a growing mismatch between what the agent currently struggles with and what its learning process emphasizes. We call this phenomenon Agent-Learning Environment Misalignment (ALEM). We introduce SEAL, a framework for synergistic co-evolution of agents and learning environments. SEAL turns verifier-grounded execution failures into a shared diagnostic signal that coordinates two complementary forms of adaptation: it reshapes the support available in subsequent interactions and adjusts how the policy learns from collected trajectories. In this way, the learning environment follows the agent's evolving failure profile while both sides remain grounded in the same executable feedback. Across three backbones on BFCL V3, SEAL consistently improves over matched Vanilla RL, reaching 40.25% versus 30.75% with Qwen2.5-7B. A matched factorial ablation further yields a +5.00-point interaction estimate for the coupled configuration. The same trained checkpoints also show positive transfer to the evaluated BFCL V4 and $τ^2$-bench domains. These results suggest that coordinating how an agent learns with the environment it learns from is a promising direction for self-evolving tool-use agents.

cs.CL

新论文 · v1arXiv:2610.36134

When Connectivity Misleads: Guiding Agentic Search with Contrastive Evidence Paths

Junchen Li, Liang Xu, Rongzheng Wang, Qizhi Chen, Junhao Wen, Shuang Liang

原文摘要

Search agents can use cross-document graphs to discover sources through multi-hop connections and choose subsequent actions based on candidate rankings and relationships within a limited observation window. However, documents may rank highly because their numerous connections make them easier to reach, without necessarily providing more valuable evidence. Graph diffusion methods such as Personalized PageRank (PPR) aggregate propagation contributions from multiple sources, allowing highly connected distractor nodes to crowd the observation window. Flat node scores also obscure cross-document evidence trails, limiting continued exploration. We propose Contrastive Evidence Paths (CEP), which scores candidates by their strongest individual paths on a query-conditioned document graph and calibrates these scores against the same graph's background visitation distribution to mitigate structural bias. CEP ranks candidates by the calibrated scores and retains the corresponding source paths. Building on CEP, we develop Structured Corpus Interaction (SCI), which organizes candidates and their connecting paths into a structured Map, enabling agents to verify source documents and pursue missing information through expansion from anchor documents. Experiments on BrowseComp-Plus and TREC-COVID show that SCI achieves state-of-the-art answer accuracy and nDCG@10 with the lowest generation cost. Compared with the strongest baselines by task performance, maximum reductions in tool calls and total generation costs exceed 50% and 40%, respectively. Further analysis confirms that SCI with CEP suppresses ranking inflation induced by structural connections and improves gold recall and exploration efficiency.

cs.IR · cs.AI

新论文 · v1arXiv:2610.34926

Supervising the Visible State: Separating Representation from View Selection in Streaming Video

Yihao Hu, Pan Wang, Yitong Lin, Zhihao Wen, Deyuan Liu, Hang Wang, Mingda Wan, Yiwen Zhao 等 11 位作者

原文摘要

An action seen once and the same action seen three times receive the same action-recognition label, so that supervision does not require a model to count. Answers shared across histories that differ in event order, accumulation, or relations leave those distinctions unconstrained; we call this state under-specification. Visible-State Supervision (ViSS) addresses it by training on source-grounded probes of event identity, order, accumulation, relation change, and answer sufficiency; under the same recent-frame interface it improves OVO-Bench over a strict-matched generic temporal supervision baseline while remaining comparable on StreamingBench, with the largest gain in hallucination detection. Test-Time Causal View Search (TVS) then adapts evidence access while keeping the learned model fixed, expanding from recent frames to sparse event and compact history views on demand. The two interventions separate representation from view selection: supervision particularly benefits hallucination detection and causal reasoning, access particularly benefits event recall and forward reasoning. Together they reach the best reported accuracy among the published systems compared here on StreamingBench and OVO-Bench at a fraction of the processed frames of exhaustive selection, and strict replay shows that most of the remaining gap to a three-view oracle is a correct answer that was later replaced. Streaming accuracy depends on what supervision asks the model to keep, what evidence inference exposes, and whether the model keeps a correct answer once it has one.

cs.CV

新论文 · v1arXiv:2610.34283HF 社区推荐

ChronoReplay: One Answer Interface Is Not Enough for Streaming Spatial Intelligence

Yihao Hu, Yitong Lin, Pan Wang, Zhihao Wen, Hang Wang, Yiwen Zhao, Deyuan Liu, Yutong Li 等 11 位作者

原文摘要

Training a spatial VLM to answer which side an object lies on does not require it to preserve the metric displacement that determines that answer, while supervision for a distance estimate may leave categorical relations largely unconstrained. When post-training emphasizes one readout, improvements can therefore concentrate on that interface while transfer to others remains weak or even degrades; we call this observed specialization interface lock-in. ChronoReplay addresses it by replaying heterogeneous spatial supervision as chronologically ordered episodes, retaining source-native targets, and balancing supervision across families. Under the same observation access, a single ChronoReplay checkpoint improves OVO-S-Bench, OST-Bench, and UCS-Bench together, including a 13.41-point gain on OST-Bench, and achieves the highest three-benchmark mean among the supervision settings compared here. Same-data controls further separate the ingredients of this transfer: answer serialization redistributes gains across OST and UCS, family-balanced learning improves over ordinary joint SFT and natural turn weighting, and chronological replay exceeds a shifted-order control by about 2.41 OST points across two paired seeds. Adaptive frame selection then changes evidence access while keeping the trained model fixed, providing an additional inference-time gain at lower visual cost. Streaming spatial performance depends on what distinctions supervision preserves, how heterogeneous histories are organized during learning, and what causal evidence inference makes available at test time.

cs.CV

新论文 · v1arXiv:2610.34073HF 社区推荐

Memory Tax: Dissecting the Net Value of Agent Memory

Xuan He, Sining Zhoubian, Ruozhen Yang, Yuhao Wen, Zhineng Wang, Yuyang Hu, Ke Yang, Yuyu Zhang 等 11 位作者

原文摘要

Memory is increasingly used in long-horizon agents, yet the net value of building memory over direct retrieval from raw history remains largely unclear. Through a large-scale evaluation across multiple benchmarks and model backbones, we find many dedicated memory systems underperform naive RAG in many cases, indicating a negative net value. Taking RAG as a reference, we dissect the net value of memory into the gains and losses over direct retrieval, termed memory benefit and memory tax. These gains and losses arise because memory process changes how the raw information is organized and accessed. We view such changes as the double-edged deviations from the raw history. They are originally designed and introduced to benefit the task, for example by reorganizing facts or distilling preferences, but some may also remove or obscure evidence, giving rise to memory tax. Since memory benefit reflects the expected gains of its design, we then focus more on the overlooked memory tax, study how it arises from the system and evaluation side, and generalize the concept to more application scenarios of memory. We hope the notion of memory tax can provide a new lens and shift more attention from standalone performance to the net value of memory, informing more reliable memory designs and evaluations. We will open-source the code and results to the community. Project Page: https://hexuan21.github.io/MemoryTax/

cs.CL

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.185ACL Anthology · ACL 2026

SDiaReward: Modeling and Benchmarking Spoken Dialogue Rewards with Modality and Colloquialness

Jingyu Lu, Yuhan Wang, Fan Zhuo, Xize Cheng, Changhao Pan, Xueyi Pu, Yifu Chen Zhejiang University, Chenyuhao Wen 等 10 位作者

原文摘要

The rapid evolution of end-to-end spoken dialogue systems demands transcending mere textual semantics to incorporate paralinguistic nuances and the spontaneous nature of human conversation. However, current methods struggle with two critical gaps: the modality gap, involving prosody and emotion, and the colloquialness gap, distinguishing written scripts from natural speech. To address these challenges, we introduce SDiaReward, an end-to-end multi-turn reward model trained on SDiaReward-Dataset, a novel collection of episode-level preference pairs explicitly targeting these gaps. It operates directly on full multi-turn speech episodes and is optimized with pairwise preference supervision, enabling joint assessment of modality and colloquialness in a single evaluator. We further establish ESDR-Bench, a stratified benchmark for robust episode-level evaluation. Experiments demonstrate that SDiaReward achieves state-of-the-art pairwise preference accuracy, significantly outperforming general-purpose audio LLMs. Further analysis suggests that SDiaReward captures relative conversational expressiveness beyond superficial synthesis cues, improving generalization across domains and recording conditions.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表