arXiv 论文速递

Coding agents have emerged as a promising paradigm for robot manipulation: a language model writes the robot controller as a program, and agents built in this way now operate robot…

Embedding spaces define notions of semantic similarity and distance. We study whether those embeddings reflect physical measurements of mass, distance, time and volume, which admit…

Complex robotic manipulation tasks frequently require a long-term memory of past events and actions. As conditioning on full histories renders policies prone to spurious correlatio…

从稀疏单目视图建模铰接物体因观测不完整而困难，该前馈方法实现稀疏观测下的三维铰接建模，对机器人操作与场景理解有重要价值。

Modeling articulated objects from sparse monocular views is challenging because each observation reveals only partial geometry and motion evidence. Most feed-forward methods infer …

Professional design requires any-color control: the ability to specify an object's target color with any 24-bit hex value for image generation and editing. Prior work has explored …

Hereditary polyposis syndromes can be precursor lesions to colorectal cancer and are associated with a broad spectrum of extracolonic tumors. Early identification and accurate clas…

Pretraining a neural PDE surrogate can reduce the amount of new CFD data needed when geometry or modeled physics changes. However, it remains unclear how different components of di…

前沿编码智能体被长期自主信任，但其最终回复常是唯一记录，论文量化其夸大倾向，对自主智能体的可信度评估与安全部署至关重要。

Frontier coding agents are increasingly trusted to work autonomously for long periods, yet an agent's final response is often the only account of that work a user sees. We quantify…

大模型强化学习对训练与推理引擎的微小差异极为敏感，该工作提出分数中心化方法稳定离策略RL，提升训练稳定性与可复现性。

Reinforcement learning (RL) of large language models is notoriously sensitive to small differences between training and inference engines, often referred to as the training-inferen…

编码框架决定智能体能力如何转化为长程软件工程表现，该实证研究系统揭示框架设计的关键因素，为构建更可靠的编码智能体提供指导。

Coding harnesses shape how autonomous coding agents translate model capabilities into long-horizon software-engineering performance, yet existing work typically evaluates harnesses…

Generative models are increasingly used to solve scientific inverse problems, but existing evaluations still focus primarily on whether a method can produce a single plausible reco…

多轮智能体RL仅获单一标量奖励，该工作提出自退役在策略蒸馏，缓解信用分配难题，提升智能体强化学习训练效率。

Multi-turn agents trained with reinforcement learning (RL) receive a single scalar reward per trajectory, which motivates self on-policy distillation (OPD) to supply dense token-le…