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
+1已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-07 19:34 · 全站最近入库：2026-11-07 17:43各信源进度 →

统计口径与数据新鲜度

统计 2026-11-06 20:10 — 2026-11-07 20:10（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：2026-11-07 19:34最近发现新增版本：2026-11-07 17:43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4,056 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 19 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-07 19:30（北京时间）
新论文 · v1arXiv:2611.07979

Perception Over Geometry: Rethinking Gaussian Allocation in Mesh-based Splatting

Yuting Wang, Yang Zhao, Shanding Diao, Penglin Wang, Yuan Chen, Wei Jia, Ronggang Wang

原文摘要

3D Gaussian Splatting (3DGS) enables photorealistic novel-view synthesis but lacks explicit topology for geometric editing. Existing mesh-based methods enable geometric manipulation by associating Gaussians with triangular mesh faces, typically assigning a fixed number of Gaussians to each face. However, the representation capacity required for appearance reconstruction depends on local appearance complexity rather than merely on mesh tessellation density. To address these limitations, we propose P-MGS, a framework that separates appearance modeling from geometric control to achieve high quality rendering, compact representation, and flexible mesh editing. P-MGS introduces a perceptual complexity guided allocation strategy that distributes Gaussians according to surface appearance demand estimated from multiview observations. To preserve reconstruction flexibility, Gaussians are optimized without mesh constraints and subsequently associated with the original mesh through post hoc binding, retaining adaptive refinement while preserving mesh topology. During mesh editing, local affine deformation transfer updates Gaussian centers and covariances under mesh deformation, reducing misalignment and floating artifacts. Experiments demonstrate superior average novel view synthesis quality with fewer Gaussians and coherent mesh editing.

cs.GR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.07689

MCPHallu: Benchmarking Reasoning, Execution, and Memory Hallucinations in MCP Agents

Yizhen Jiang, Haowei Guo, Tianyi Bai, Zengjie Hu, Lu Ma, Zhengyang Zhao, Haoze Sun, Peng Pei 等 9 位作者

原文摘要

Large language model agents are increasingly deployed through the Model Context Protocol (MCP), which allows them to discover and invoke tools across multiple servers at runtime. Existing MCP benchmarks primarily measure task completion; where they diagnose failures, they do so after execution rather than building specific reasoning, execution, and memory failure risks into task construction. We introduce MCPHallu, a benchmark for diagnosing hallucination-prone agent failures in MCP environments along three axes: reasoning, execution, and memory. MCPHallu contains 358 tasks across five capability domains and runs on 36 production MCP servers in isolated containers. Each task is designed with a structural trigger that makes one of four subtypes the dominant failure risk: Branch Collapse, Unreachable Goal, Tool Misuse, or Context Forgetting. This design enables subtype-level analysis rather than aggregate task success alone. We evaluate fourteen frontier LLM agents from eight vendors on 5{,}012 trajectories. Current MCP agents remain vulnerable to hallucination-inducing tool environments, achieving an average score of 0.642 on a $[0,1]$ scale (higher is better). Failures concentrate on Unreachable Goal tasks, where the average score drops to 0.458 and 23.2\% of runs pass, indicating that agents often continue issuing tool calls or claim successful completion rather than recognizing infeasible goals. Subtype rankings diverge sharply: a model's rank shifts by 6.9 of 13 positions on average across subtypes, and Branch Collapse is uncorrelated with Unreachable Goal across models ($r = -0.12$). MCPHallu is released at https://github.com/J1zz/mcp_hallu.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.07355

Constant-Factor Approximation Algorithms for Weighted Budgeted Multisensor Sweep Coverage

Jingyang Zhao, Jason William Setiawan, Uman Khalid, Chaoning Zhang, Een-Kee Hong, Hyundong Shin

原文摘要

Sweep coverage requires each point of interest (PoI) to be revisited by some mobile sensor at least once during every sweep period, and it underlies persistent patrolling, environmental monitoring, and infrastructure inspection. Given a set of $N$ PoIs, each associated with a nonnegative weight representing its importance, the weighted budgeted sweep coverage problem (WB-SCP) seeks routing trajectories for a budget of $M\leq N$ identical mobile sensors with a common maximum speed to maximize the total weight of the sweep-covered PoIs. For the unweighted special case in which every PoI has unit weight, the first constant-factor approximation algorithm achieves a ratio of $0.0116-O(\varepsilon)$ for any constant $\varepsilon>0$, and whether the WB-SCP itself admits such an algorithm has remained open. In this paper, we resolve this open question by proposing two constant-factor approximation algorithms for the WB-SCP. The first converts any bifactor $(α,β)$-approximation algorithm into a $β/(2α)$-approximation algorithm, yielding a ratio of $1/16=0.0625$. The second builds on a greedy maximum-coverage framework and improves the ratio to $1-e^{-1/2}-\varepsilon$ for any fixed $\varepsilon>0$. Both algorithms also improve the best-known ratio for the unweighted case, and the second extends to heterogeneous sensors with different speeds while preserving the same guarantee.

cs.DS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.07348

Constant-Factor Approximations for the Biased Traveling Salesman Problem and Its Path Variant

Jingyang Zhao, Jason William Setiawan, Uman Khalid, Chaoning Zhang, Een-Kee Hong, Hyundong Shin

原文摘要

The metric traveling salesman problem (TSP) is a fundamental problem in combinatorial optimization that asks for a minimum-cost Hamiltonian cycle in a complete undirected graph with $N$ vertices, where edge costs satisfy the triangle inequality. The biased TSP (BTSP) is a natural generalization of the metric TSP, in which the vertex set is partitioned into $k$ subsets: edge costs satisfy the triangle inequality within each subset, while edges crossing the subsets may violate the triangle inequality by a certain bias factor. This model captures scenarios in which distances are metric only locally but may be distorted across different regions or types of vertices. In this paper, we study approximation algorithms for the BTSP and its path variant, the biased traveling salesman path problem (BTSPP), which asks for a minimum-cost Hamiltonian path. Previously, constant-factor approximation algorithms were known only for the case $k=2$. Whether the BTSP and the BTSPP admit constant-factor approximation algorithms for any fixed integer $k>2$ has remained an open question. We resolve this question affirmatively. We show that both the BTSP and the BTSPP admit $O(k)$-approximation algorithms with running time $N^{O(2^k k!)}$. Moreover, for the BTSPP with $k=2$, we propose an improved $(α+2+\varepsilon)$-approximation algorithm, where $α<3/2-10^{-34}$ is the best-known approximation ratio for the metric TSP, improving upon the previous ratio of $4$

cs.DS

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2605.09214

Fast Rates for Offline Contextual Bandits with Forward-KL Regularization under Single-Policy Concentrability

Qingyue Zhao, Kaixuan Ji, Heyang Zhao, Quanquan Gu

原文摘要

\emph{Kullback-Leibler} (KL) regularization is ubiquitous in reinforcement learning algorithms in the form of \emph{reverse} or \emph{forward} KL. Recent studies have demonstrated $ε^{-1}$-type fast rates for decision making under reverse KL regularization, in contrast to the standard $ε^{-2}$-type sample complexity. However, for forward-KL-regularized objectives, existing statistical analyses are either not applicable or result in $\tilde{O}(ε^{-2})$ slow rates. We take the first step towards addressing this problem via a streamlined analysis of forward-KL-regularized offline CBs. We give the first $\tilde{O}(ε^{-1})$ upper bounds in tabular and general function approximation settings, both under notions of \emph{single-policy concentrability}. In particular, our convex-analytical pipeline unifies these settings by exploiting the pessimism principle in a novel way and completely bypasses the proof routines in previous works based on the mean value theorem, which might be of independent interest. Moreover, we provide rate-optimal lower bounds, manifesting the tightness of our upper bounds in terms of statistical rates. Our lower bounds also demonstrate that the forward-KL-regularized sample complexity recovers the unregularized slow rate in the low-regularization regime, similarly to the reverse-KL regularization.

cs.LG · cs.AI · cs.IT · math.ST · stat.ML

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.06410

Polynomial $(2α+\varepsilon)$- and Quadratic $4$-Approximation Algorithms for the Biased Traveling Salesman Problem in Bipartitioned Networks

Jingyang Zhao, Trung Q. Duong, Een-Kee Hong, Hyundong Shin

原文摘要

The traveling salesman problem (TSP) is one of the fundamental problems in combinatorial optimization that asks for a minimum-cost Hamiltonian cycle in a complete undirected graph with $N$ vertices. For the metric TSP, where edge costs satisfy the triangle inequality, a long line of work has produced constant-factor approximation algorithms, culminating in the best-known approximation ratio $α<3/2-10^{-36}$. The biased TSP (BTSP) generalizes the metric TSP to bipartitioned networks: the vertex set is partitioned into two subsets, edge costs satisfy the triangle inequality within each subset, while edges crossing the partition may violate it by a certain bias factor. This model captures heterogeneous routing scenarios in which distances are metric only locally but may be distorted across different regions or vertex types. Previously, the only known result for the BTSP was a $7/2$-approximation algorithm with running time $O(N^3)$. In this paper, we present two new approximation algorithms that significantly improve the state of the art. The first achieves an approximation ratio of $2α+\varepsilon$ (strictly less than $3$) in polynomial time $N^{O(1)}$ for any constant $\varepsilon>0$, where the exponent grows at least linearly with $1/\varepsilon$. The second achieves a $4$-approximation in only $O(N^2)$ running time, reducing the previous cubic-time bound by a full polynomial factor. These advances directly support emerging networked systems---including multi-tier 6G architectures and hybrid quantum-classical networks---where cross-region transitions incur biased and non-metric overheads.

cs.DS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02721

IssueLighter: An Issue Type-Aware LLM Agent for Issue Report Validity Assessment

Jiawei Ran, Yuxia Zhang, Ziyang Zhao, Hui Liu

原文摘要

Invalid issue reports are prevalent in open source software (OSS) projects, imposing substantial burdens on maintainers who already face a high risk of burnout. Prior research has primarily focused on identifying invalid reports for specific issue categories (e.g., bug and wontfix), leaving the broader landscape of OSS issue types largely unexplored and limiting the generalizability of existing approaches. Moreover, existing methods provide little support for explaining or validating their predictions, reducing their practical usefulness for maintainers. With the rise of LLM-based agents, it remains unclear whether a comprehensive understanding of the causes of invalidity across all issue types can improve validity assessment performance while producing well-grounded rationales. To this end, we first manually annotated the validity status of 520 issue reports from three OSS projects, i.e., pandas, scikit-learn, and mypy, resulting in 257 invalid reports and 263 valid ones. Then, we conducted a thematic analysis on these invalid reports and identified 13 causes of invalidity that are closely specific to issue types. Based on these insights, we propose IssueLighter, an agent-based framework with a type-aware validity assessment strategy that dynamically executes commands. Experimental results show that IssueLighter consistently and substantially outperforms all baseline methods, with an average F1 improvement of over 20 percentage points. Ablation studies confirm that both the type-aware strategy and the taxonomy of causes of invalidity contribute to the performance gains. Qualitative analysis further demonstrates that IssueLighter generates rationales grounded in accurately identified causes of invalidity and supported by evidence. Moreover, leveraging these rationales, we challenged maintainers' original decisions in four disputed cases, and all received approval.

cs.SE · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.35252

Labference: Benchmarking the Reasoning-Action Connection in Robotic Laboratories

Shitong Liu, Xu Yang, Jingwei Zhang, Sheng Lian, Xinyu Zhuang, Kaiwen Guo, Quan Liu, Chunhai Zhao 等 21 位作者

原文摘要

Realizing Embodied AI for Science requires robots to translate experimental goals into reliable physical procedures, reducing hazardous and repetitive human work while generating evidence for scientific discovery. Although language-based planning and learned manipulation provide complementary capabilities, their integration requires robots to adapt subsequent operations to actual execution outcomes. For example, an unsuccessful placement changes what must be done next, even when the experimental goal remains unchanged. Existing benchmarks assess planning, manipulation, and laboratory workflows through task completion and reasoning or execution metrics; however, aggregate performance leaves the connection between outcome-dependent choices and physical execution unresolved. To address this evaluation gap, we introduce Labference, a benchmark for evaluating how observed physical outcomes guide subsequent choices and how those choices translate into robotic laboratory actions. Its evaluation protocol is designed to test outcome-dependent behavioral adaptation through controlled evidence variations, while independently measuring operation selection, physical execution, and task completion. Matched Oracle-assisted continuations from shared checkpoints then assess which forms of support improve subsequent choices and task completion. Initial weighing evaluations reveal target-selection errors in the tested VLA policies despite successful object transfers, whereas GPT-driven control achieves correct selection but encounters execution failures and substantial computation costs. Together, these findings highlight the need to connect task-appropriate reasoning with reliable, timely action. By making this connection measurable and diagnosable, Labference supports targeted improvements toward autonomous experimentation for Embodied AI for Science.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2510.03199

Best-of-Majority: Minimax-Optimal Strategy for Pass@$k$ Inference Scaling

Qiwei Di, Kaixuan Ji, Xuheng Li, Heyang Zhao, Quanquan Gu

原文摘要

LLM inference often generates a batch of candidates for a prompt and selects one via strategies like majority voting or Best-of- N (BoN). For difficult tasks, this single-shot selection often underperforms. Consequently, evaluations commonly report Pass@$k$: the agent may submit up to $k$ responses, and only the best of them is used when computing regret. Motivated by this, we study inference scaling in the more general Pass@$k$ inference setting, and prove that neither majority voting nor BoN exhibits the desirable scaling with $k$ and the sampling budget $N$. Combining the advantages of majority voting and BoN, we propose a new inference strategy called Best-of-Majority (BoM), with a pivotal step that restricts the candidates to the responses with high frequency in the $N$ samples before selecting the top-$k$ rewards. We prove that when the sampling budget is $N=\tildeΩ(C^*)$, the regret of BoM is $O(ε_{\mathrm{opt}}+\sqrt{ε_{\mathrm{RM}}^2C^*/k})$, where $C^*$ is the coverage coefficient, $ε_{\mathrm{RM}}$ is the estimation error of the reward model, and $ε_{\mathrm{opt}}$ is the estimation error of reward at the optimal response. We further establish a matching lower bound, certifying that our algorithm is minimax optimal. Beyond optimality, BoM has a key advantage: unlike majority voting and BoN, its performance does not degrade when increasing $N$. Experimental results of inference on math problems show BoM outperforming both majority voting and BoN.

cs.LG · stat.ML

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.36681

From Biological Curves to Soft Modular Robots: Learning-Based Inverse Design with Uncertainty-Guided Refinement

Dixiang Wang, Minghao Zha, Arifur Rahman, Yinhong Qin, Yitao Jiang, Shiyang Lu, Zebing Mao, Muhao Chen 等 9 位作者

原文摘要

Animals exhibit diverse shape-changing behaviors, from the undulatory swimming of eels to the arching movements of inchworms. Reproducing these biological shapes with modular soft robots offers a promising approach to developing versatile bioinspired robotic systems. However, translating a desired shape into a feasible robotic configuration remains challenging. Existing cable-driven modular soft robots rely on computationally expensive physics-based modeling and combinatorial search for inverse design, making it difficult to efficiently reproduce diverse target shapes. To address this challenge, we present a learning-based real-to-sim-to-real framework that translates biological curves into robotic configurations through surrogate-assisted inverse design and uncertainty-guided refinement. The proposed method combines learned inverse prediction with targeted candidate exploration, using a fast forward surrogate to evaluate and select suitable designs. The learned forward surrogate reduces deformation prediction time from $1.53 \times 10^5$ ms to 0.0008 ms, achieving approximately a $1.9 \times 10^8$-fold speedup. At the same time, the proposed inverse design method reduces design time from $1.99 \times 10^6$ ms to 4.18 ms. On 5,000 test samples generated from previously unseen threading patterns, the method achieves a 93% shape reconstruction success rate under surrogate-based evaluation. Physical demonstrations with four bioinspired robots further illustrate the translation of biological curves into realizable robotic shapes.

cs.RO · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.37149

GCSR-OCC: Geometry Calibration and Semantic Routing for Open-Vocabulary Occupancy Prediction

Bohua Wang, Congyang Zhao, Tianxiong Lv, Chenxu Zhao, He Zheng, Hangqi Fan, Leigang Luo, Zhihao Liao 等 10 位作者

原文摘要

Open-vocabulary 3D occupancy prediction represents scene geometry and semantics in a unified voxel grid, enabling recognition beyond predefined categories. One approach reconstructs scenes with 3D Gaussians and converts them into voxel occupancy. However, Gaussian scale inflation and excessive overlap can distort voxel occupancy predictions, while semantic mixing can arise within Gaussians spanning multiple categories and during 3D--2D semantic feature alignment. We propose GCSR-OCC, which combines geometry calibration and multi-mode query-conditioned semantic routing for indoor occupancy prediction from a single RGB image. The Evidence-Conserving Gaussian-to-Occupancy operator (EC-G2O) uses kernel mass normalization, $p$-norm aggregation, and a learnable gain to suppress scale inflation and excessive local evidence accumulation while calibrating occupancy responses. Multi-mode Query-First Gaussian Semantic Routing (MM-QGSR) matches multiple semantic modes per Gaussian to text queries and aggregates query-specific evidence into voxels through a learnable text-conditioned router. Fine-grained subclass queries and smoothed semantic aggregation further refine voxel-level predictions. Without 3D semantic annotations, GCSR-OCC achieves state-of-the-art performance in open-vocabulary occupancy prediction on Occ-ScanNet, with 63.31% geometric IoU and 27.44% semantic mIoU. It also reduces geometric aggregation latency from 19.76 ms to 12.88 ms. Our code will be made publicly available.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.36557

QESChunker: A Single Objective Unifies Overlapping and Non-Overlapping Chunking for RAG

Yifan Zhao, Qianyang Zhao, Xinyu Guan, kai wei, Yuming Deng

原文摘要

Chunking is a fundamental preprocessing step in retrieval-augmented generation (RAG), yet most existing chunkers rely primarily on heuristics. We introduce the Question--Evidence-Supervised Chunker (QESChunker), to our knowledge the first method for constructing static chunk indices from question--evidence supervision. QESChunker constructs document-specific chunk indices by optimizing over human-annotated or LLM-generated question--evidence pairs, without access to held-out test queries. We formulate non-overlapping and overlapping index construction through a unified question--evidence utility-maximization objective, turning chunking into a supervised combinatorial optimization problem. Under the non-overlap constraint, the objective decomposes over chunks and can be optimized exactly in polynomial time by dynamic programming. When overlap is allowed, the objective becomes monotone submodular; exact optimization is NP-hard, but our greedy achieves a $(1-1/\sqrt{e})$-approximation. Across two self-constructed and three public datasets, both variants outperform existing chunking methods in complete-evidence recall in nearly all evaluated settings, under the same overlap regime and a fixed hybrid retriever. These results recast chunking as supervised, budget-aware index construction for evidence retrieval.

cs.IR · cs.CL · math.OC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.36506

GarmentSegAgent: Autonomous Garment Segmentation for Web Fashion Images via Textual Self-Refinement

Jia Chen, Chao Huang, Xinrong Hu, Mingdi Huang, Ruyang Zhao, Jun Bai

原文摘要

Web fashion images support applications such as visual product search and consumer-to-shop matching, yet their visual heterogeneity makes reliable garment segmentation challenging. Despite advances in promptable segmentation, autonomously identifying and correcting garment-specific errors without human-provided spatial prompts remains difficult. We propose GarmentSegAgent, an autonomous framework that integrates semantic planning, spatial execution, and visual evaluation through iterative textual self-refinement. At its core, we introduce CAST (Constrained Attribution-guided Selective TextGrad), which connects visual error attribution with targeted textual optimization. CAST identifies instruction components responsible for segmentation errors, selectively updates them through textual gradients, and validates the resulting instructions against garment-specific constraints. This enables image-specific error correction at inference time while keeping the underlying foundation models frozen. Experiments on DeepFashion2, ModaNet, and Fashionpedia demonstrate consistent improvements in region overlap and boundary accuracy, alongside more reliable refinement than conventional TextGrad. By enabling visual feedback to guide controlled instruction updates, GarmentSegAgent offers a promising direction for autonomous garment understanding, with potential applications in scalable fashion content analysis, visual retrieval, and consumer-to-shop matching on the Web.

cs.CV · cs.MA

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2605.02141

On the Optimal Sample Complexity of Offline Multi-Armed Bandits with KL Regularization

Kaixuan Ji, Qiwei Di, Heyang Zhao, Qingyue Zhao, Quanquan Gu

原文摘要

Kullback-Leibler (KL) regularization is widely used in offline decision-making and offers several benefits, motivating recent work on the sample complexity of offline learning with respect to KL-regularized performance metrics. Nevertheless, the exact sample complexity of KL-regularized offline learning remains largely from fully characterized. In this paper, we study this question in the setting of multi-armed bandits (MABs). We provide a sharp analysis of KL-PCB (Zhao et al., 2026), showing that it achieves a sample complexity of $\tilde{O}(ηSAC^{π^*}/ε)$ under large regularization $η= \tilde{O}(ε^{-1})$, and a sample complexity of $\tildeΩ(SAC^{π^*}/ε^2)$ under small regularization $η= \tildeΩ(ε^{-1})$, where $η$ is the regularization parameter, $S$ is the number of contexts, $A$ is the number of arms, $C^{π^*}$ policy coverage coefficient at the optimal policy $π^*$, $ε$ is the desired sub-optimality, and $\tilde{O}$ and $\tildeΩ$ hide all poly-logarithmic factors. We further provide a pair of sharper sample complexity lower bounds, which matches the upper bounds over the entire range of regularization strengths. Overall, our results provide a nearly complete characterization of offline multi-armed bandits with KL regularization.

cs.LG · cs.AI · math.ST · stat.ML

查看论文
arXiv ↗PDF ↗
首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1019ACL Anthology · ACL 2026

MAESTRO: Meta-learning Adaptive Estimation of Scalarization Trade-offs for Reward Optimization

Yang Zhao, Hepeng Wang Harbin Institute of Technology, Xiao Ding, Yangou Ouyang, Bibo Cai, Kai Xiong Harbin Institute of Technology, Jinglong Gao Research Center for Social Computing and Information Retrieval, Zhouhao Sun 等 11 位作者

原文摘要

Group-Relative Policy Optimization (GRPO) has emerged as an efficient paradigm for aligning Large Language Models (LLMs), yet its efficacy is primarily confined to domains with verifiable ground truths. Extending GRPO to open-domain settings remains a critical challenge, as unconstrained generation entails multi-faceted and often conflicting objectives—such as creativity versus factuality—where rigid, static reward scalarization is inherently suboptimal. To address this, we propose MAESTRO (Meta-learning Adaptive Estimation of Scalarization Trade-offs for Reward Optimization), which introduces a meta-cognitive orchestration layer that treats reward scalarization as a dynamic latent policy, leveraging the model’s terminal hidden states as a semantic bottleneck to perceive task-specific priorities. We formulate this as a contextual bandit problem within a bi-level optimization framework, where a lightweight Conductor network co-evolves with the policy by utilizing group-relative advantages as a meta-reward signal. Across seven benchmarks, MAESTRO consistently outperforms single-reward and static multi-objective baselines, while preserving the efficiency advantages of GRPO, and in some settings even reducing redundant generation.

cs.CL

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

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1084ACL Anthology · ACL 2026

I²B-LPO: Latent Policy Optimization via Iterative Information Bottleneck

Huilin Deng, Hongchen Luo Northeastern University, Yue Zhu, Long Li, Zhuoyue Chen Alibaba, Xinghao Zhao, Ming LI Hithink RoyalFlush Information Network Co.,Ltd., Chuyang Zhao Baidu 等 12 位作者

原文摘要

Despite recent advances in Reinforcement learning with verifiable rewards (RLVR) for large language model (LLM) reasoning, most methods suffer from exploration collapse, as the semantic homogeneity of random rollouts traps models in narrow, over-optimized behaviors. Existing methods leverage policy entropy to encourage exploration, but face inherent limitations: global entropy regularization is susceptible to reward hacking, inducing meaningless verbosity, whereas local token-selective updates struggle with the strong inductive bias of pre-trained models. To this end, we propose Latent Policy Optimization via Iterative Information Bottleneck ( I²B-LPO), which shifts from statistical perturbation of token distributions to topological branching of reasoning trajectories. I²BLPO triggers latent branching at high-entropy states to diversify reasoning trajectories and applies the Information Bottleneck as a trajectory filter and self-reward to ensure concise and informative exploration. Empirical results on four mathematical benchmarks demonstrate that I²B-LPO achieves state-of-the-art performance, with margins of up to 5.3% in accuracy and 7.4% in diversity metrics. Code is available at https://github.com/denghuilin-cyber/IIB-LPO.

cs.CL

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

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1097ACL Anthology · ACL 2026

Consolidation or Adaptation? PRISM: Disentangling SFT and RL Data via Gradient Concentration

Yang Zhao, Yangou Ouyang, Xiao Ding, Hepeng Wang Harbin Institute of Technology, Bibo Cai, Kai Xiong Harbin Institute of Technology, Jinglong Gao Research Center for Social Computing and Information Retrieval, Zhouhao Sun 等 11 位作者

原文摘要

While Hybrid Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL) has become the standard paradigm for training LLM agents, effective mechanisms for data allocation between these stages remain largely underexplored. Current data arbitration strategies often rely on surface-level heuristics that fail to diagnose intrinsic learning needs. Since SFT targets pattern consolidation through imitation while RL drives structural adaptation via exploration, misaligning data with these functional roles causes severe optimization interference. We propose PRISM, a dynamics-aware framework grounded in Schema Theory that arbitrates data based on its degree of cognitive conflict with the model’s existing knowledge. By analyzing the spatial geometric structure of gradients, PRISM identifies data triggering high spatial concentration as high-conflict signals that require RL for structural restructuring. In contrast, data yielding diffuse updates is routed to SFT for efficient consolidation. Extensive experiments on WebShop and ALFWorld demonstrate that PRISM achieves a Pareto improvement, outperforming state-of-the-art hybrid methods while reducing computational costs by up to 3.22 \times. Our findings suggest that disentangling data based on internal optimization regimes is crucial for scalable and robust agent alignment.

cs.CL

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

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.115ACL Anthology · ACL 2026

Toward Secure Tuning: Mitigating Security Risks from Instruction Fine-Tuning

Yanrui Du, Fenglei Fan City University of Hong Kong, Sendong Zhao, Jiawei Cao, Ming Ma, Danyang Zhao Harbin Institute of Technology, Shuren Qi City University of Hong Kong, Ting Liu Harbin Institute of Technology 等 9 位作者

原文摘要

Instruction Fine-Tuning (IFT) has emerged as a critical technique for customizing Large Language Models (LLMs) to meet diverse downstream applications. However, recent studies have revealed that IFT can compromise the built-in security mechanisms of LLMs, thereby posing significant security risks. Although defense methods targeting various training stages have been proposed, they either face challenges in practical deployment or exhibit instability and limited performance gains. In our study, we propose a novel SWAT method that introduces a key idea: shifting more of the learning burden onto security-robust parameters. To this end, our study investigates how module-level parameters affect LLMs’ internal security feature space, aiming to uncover robustness patterns in parameters. Guided by this analysis, we identify a robust module set (Mods_Rob) that exhibits minimal effects on LLMs’ security feature space. Leveraging this insight, SWAT proceeds in two phases: (1) a warm-up phase that preferentially trains Mods_Rob to learn low-level features with minimal security risk, followed by (2) standard tuning to achieve optimal task performance. Across diverse knowledge-intensive datasets, scenarios, and LLMs, SWAT substantially reduces security risks without sacrificing task performance gains.

cs.CL

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

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.352ACL Anthology · ACL 2026

DART: Disambiguation-Aware Reasoning for Video-guided Machine Translation

Boyu Guan Institute of automation, Chinese academy of science, Chinese Academy of Sciences, Chuang Han, Yang Zhao, Chengqing Zong Institute of automation, Chinese Academy of Sciences

原文摘要

Video-guided Machine Translation (VMT) seeks to enhance translation quality by incorporating contextual information derived from paired short video clips. However, many VMT samples are text-sufficient; even when visual information is needed, only minimal cues are required. Aiming to tackle these issues, we propose a novel framework DART (Disambiguation-Aware Reasoning for Video-guided Machine Translation). Reinforcement learning is used to incorporate multimodal large language models’ multimodal reasoning into VMT. The model dynamically switches between text-only processing and multimodal integration, contingent on the necessity of visual disambiguation. Furthermore, we present TVRF (Translation-oriented Video Relevance Filtering), a systematic pipeline for constructing training data based on multimodal relevance to translation. This pipeline filters samples where video information is translation-relevant, mitigating training collapse caused by video-irrelevant data in conventional VMT. Experimental results show that our approach improves multimodal information utilization in VMT, yielding gains in both translation quality and computational efficiency.

cs.CL · cs.CV

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