FrontierPhysics: Evaluating Agents for End-to-End Physics Research
Bingran You, Wenjun Ke, Ria Rosenauer, Bo Liu, Wenbo Chen, Min Chen, Shuqi Xu, Yue Ma 等 72 位作者
AI 速览：作者提出包含56个由物理学家基于自身研究设计的端到端物理研究基准，用确定性验证器与专家评分量表同时考察智能体研究的正确性与可发表性。
AI agents are playing an increasingly important role in physics research. However, existing physics benchmarks only evaluate fixed questions, isolated sub-tasks, or paper reproduction with given solutions, and rarely check research quality alongside correctness. We introduce FrontierPhysics, an end-to-end benchmark with 56 tasks made by 42 physicists based on their own research, covering nine domains across theoretical, computational, and experimental physics. Each task requires agents to review the literature, develop a research plan, conduct research and experiments, and produce structured outputs and scientific manuscripts. Combining deterministic verifiers with expert-authored rubrics, FrontierPhysics evaluates whether agents' research is correct and publishable. Across 22 models in 26 configurations, no configuration passes more than 46 of its 168 rollouts. Manual review of agent trajectories shows that frontier agents' main weaknesses are insufficiently rigorous reasoning and limited expert intuition, rather than gaps in physics knowledge or execution errors.
Hanyang Liu, Hong Zhang, Ding Yuan, Yang Han, Yifan Yang
Tracking-by-propagation (TBP) methods perform end-to-end multi-object tracking by propagating track queries across frames, jointly handling detection and identity association without heuristic post-processing while naturally modeling temporal information. However, as detection and association are tightly coupled, detection queries in TBP are expected to discover currently untracked objects and provide complementary current-frame evidence for challenging tracks whose queries and identities are retained. Existing methods largely inherit detection-query formation from single-frame detection and construct such queries before the decoder with little awareness of trajectory states. Consequently, currently untracked objects may be missed, while challenging tracked objects with reduced visibility under occlusion may lack complementary evidence, causing association failures or duplicate responses. To address these issues, we propose TAQ-MOTR, a tracking-aware detection query formation framework that employs a pre-decoder Hybrid Query Generator (HQG) to fuse current-frame image features, track memory, and learnable memory. By conditioning detection queries on trajectory states, TAQ-MOTR adapts them to \rev{the discovery of currently untracked objects and tracked-object complementation We further introduce a training-only auxiliary detection branch whose one-to-one label assignment is decoupled from that of the main tracking branch, providing additional detection supervision to strengthen the shared HQG and decoder for the discovery of currently untracked objects. TAQ-MOTR improves performance on DanceTrack, SportsMOT, and BFT, achieving 72.0 HOTA, 63.1 AssA, and 76.6 IDF1 on the DanceTrack test set. Code is available at https://github.com/Magicmiya/TAQ-MOTR
Graphical user interface (GUI) agents can reuse past interaction experience. We study how illustrated procedures support action prediction and students deployed without retrieval. We introduce SkillLens, which represents procedures, applicability cues, and interface views in a shared Visual Skill Card (VSC) schema. Retrieved cards provide a frozen executor with reference evidence alongside the current screen. CardDistill uses this schema to construct privileged teacher context from labeled training steps for on-policy distillation; the student learns from native inputs and requires no retrieval at deployment. With matched training data and update counts, students trained with full-card teachers improve step success rate by an average of 5.04 percentage points over those trained with compact-text teachers across three training seeds on 119 tasks from ten websites absent from training. An independent runtime control uses training-source memory and pre-action retrieval queries: full cards achieve 12/64 correct actions (18.75%) versus 4/64 (6.25%) for text-only cards. These results establish the value of the complete VSC teacher-context package for retrieval-free students and provide evidence for pre-action retrieval of illustrated procedures.
Low-Rank Adaptation for Critic Learning in Off-Policy Reinforcement Learning
Yuan Zhuang, Yuexin Bian, Sihong He, Jie Feng, Qing Su, Songyang Han, Jonathan Petit, Shihao Ji 等 10 位作者
AI 速览：作者提出将 LoRA 作为从零训练评论家的结构正则化手段，冻结随机初始化的基矩阵、只训练低秩适配器，并在连续控制任务上评估。
Scaling critic capacity is a promising direction for improving off-policy reinforcement learning (RL). However, recent work shows that larger critics are prone to overfitting and instability in replay-based bootstrapped training. In this paper, we propose using Low-Rank Adaptation (LoRA) as a structural regularizer for critic learning from scratch. Our approach freezes randomly initialized base matrices and optimizes only the corresponding low-rank adapters, thereby constraining the learned weight changes to low-rank residuals. We evaluate our method on state-based continuous-control tasks, using SAC with SimbaV2 and BRC on DMC-Hard, and FastTD3 with its SimbaV2 variant on IsaacLab. Empirically, LoRA reduces critic loss during training and improves overall policy performance, achieving the best or competitive results on most tasks. Extensive experiments demonstrate that our low-rank updates provide a simple and effective form of structural regularization for critic learning in off-policy RL.
Interactive environments offer a practical way to evaluate large language model (LLM) agents, and competitive games extend this evaluation to decisions shaped by competing goals and other agents' choices. Yet existing evaluations confined to individual games or execution frameworks leave unclear which competitive advantages persist across settings. We introduce DECE (Diverse Eurogame Competitive Environments), a unified arena spanning ten Eurogames with shared interfaces and evaluation protocols. These games combine lasting strategic commitments, competition over shared resources, and multiple routes to victory. Our primary study evaluates 12 model-harness configurations, spanning nine models and four execution harnesses, in 480 three-player matches across ten games. The main and supplementary studies comprise 1,032 completed matches in total. These comparisons reveal broad cross-game ranking agreement alongside game-specific strengths. Matched studies show that the same model can perform differently across execution harnesses, adding another source of variation to these profiles. These performance advantages do not consistently translate into better token or cost efficiency, making resource use an additional consideration when comparing configurations. Together, the findings support evaluating complete agent configurations by both their competitive performance across games and the resources needed to achieve it.
Yifeng Ding Meta FAIR and Department of Computer Science, University of Illinois at Urbana-Champaign, Hung Le, Songyang Han NVIDIA, Kangrui Ruan AWS AI Lab, Zhenghui Jin NVIDIA, Varun Kumar Amazon, Zijian Wang Meta Superintelligence Labs, Anoop Deoras
Training Large Language Models (LLMs) for multi-turn Tool-Integrated Reasoning (TIR) – where models iteratively reason, generate code, and verify through execution – remains challenging for existing reinforcement learning (RL) approaches. Current RL methods, exemplified by Group Relative Policy Optimization (GRPO), suffer from coarse-grained, trajectory-level rewards that provide insufficient learning signals for complex multi-turn interactions, leading to training stagnation. To address this issue, we propose Group Turn Policy Optimization (GTPO), a novel RL algorithm specifically designed for training LLMs on multi-turn TIR tasks. GTPO introduces three key innovations: (1) turn-level reward assignment that provides fine-grained feedback for individual turns, (2) return-based advantage estimation where normalized discounted returns are calculated as advantages, and (3) self-supervised reward shaping that exploits self-supervision signals from generated code to densify sparse binary outcome-based rewards. Our comprehensive evaluation demonstrates that GTPO outperforms GRPO by 3.0% across diverse math reasoning benchmarks, establishing its effectiveness. GTPO also improves GRPO by 3.9% on commonsense reasoning and program synthesis tasks, demonstrating its generalizability to non-math domains. Importantly, GTPO incurs negligible overhead, ensuring its practicality for real-world scenarios.
尚无 AI 解读，展示原文摘要
cs.CL · cs.AI
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表