17 papers
Embodied Task Agents Can Perform Fine Manipulation through Tactile In-Context Learning
Yubang Wang, Zezheng Huai, Yifei Zhang +8
Can embodied task agents use tactile experience to perform fine manipulation without task-specific model training? We investigate this capability through Tactile-ETA, a system that…
Learn from Move: the Next Step for GUI Agents
Jiahao Lu, Yize Shen, Chuanyuan Tan +6
Existing benchmarks for GUI agents primarily focus on increasing task difficulty by extending task horizons. Besides, enabling agents to explore unfamiliar environments and success…
MARCO: Multi-Round Agentic Reinforcement for Conditional Molecular Optimization
Shicheng Fang, Yuxin Wang, Zhuo Yang +6
Molecular optimization is inherently iterative: a candidate is proposed, evaluated against several objectives, and revised while preserving a relationship to the source molecule. M…
VehicleArena: A Realistic Urban Environment for Multi-Agent Driving
Jie Yang, Jiajun Chen, Jiazheng Zhou +4
Real-world embodied agents often pursue independent objectives within a shared physical environment, where their actions can alter the conditions faced by others. Existing benchmar…
ORPG: Reconciling Multiple Reward Objectives through Objective-wise Policy Gradients
Shicheng Fang, Yiwen Zhao, Wenbo Tian +4
Multi-reward policy optimization requires a joint update that reflects both the learning signals and the intended relationships among objectives. We introduce Objective-wise Reconc…
SWE-bench Science: Can Coding Agents Resolve Engineering Tasks in Science?
Zhipeng Xu, Jiahao Lu, Yining Zheng +2
Software increasingly functions as part of the scientific instrument itself, making failures in scientific code capable of compromising not only program behavior but also the evide…
ETA: A New Agentic Paradigm for Embodied Tasks
Yitong Chen, Zezheng Huai, Sixian Li +7
When will robots have their ChatGPT moment? Such a breakthrough requires a general-purpose robot that can handle unfamiliar tasks in unfamiliar environments, remain controllable ov…
Let It Be Simple: One-Step Action Generation for Vision-Language-Action Models
Yitong Chen, Shiduo Zhang, Jingjing Gong +1
One-step generation from standard diffusion or Flow-Matching training is rarely effective for text-to-image or video generation without distillation or specialized one-step objecti…
Rethinking Scientific Discovery in the Agentic Era
Yining Zheng, Yuxin Wang, Jiahao Lu +27
Artificial intelligence has advanced scientific discovery, but most AI4Science systems remain fragmented tools that rely on humans to coordinate problem formulation, literature gro…
CoRE-VLA: Towards Scalable and Robust Vision-Language-Action Modeling via Conditional Routing of Experts
Haozhe Zhang, Sixian Li, Yifei Zhang +5
Vision-language-action (VLA) models have advanced generalist robotic manipulation, yet real-world deployment reveals a fundamental challenge: robots are equipped with diverse and h…
In-Context World Modeling for Robotic Control
Siyin Wang, Junhao Shi, Senyu Fei +4
Modern Vision-Language-Action (VLA) models often fail to generalize to novel setups, such as altered camera viewpoints or robot morphologies, because they are typically conditioned…
EvoCUA-1.5: Online Reinforcement Learning for Multi-turn Computer-Use Agents
Mianqiu Huang, Taofeng Xue, Chong Peng +13
Computer-use agents must solve long-horizon tasks through repeated interaction with partially observable, multimodal desktop environments. Although imitation learning and offline t…
Advancing Omnimodal Embodied Agents from Isolated Skills to Everyday Physical Autonomy
Junhao Shi, Zezheng Huai, Siyin Wang +7
Building persistent embodied agents in unstructured environments demands unified orchestration of heterogeneous tools spanning both cyber (APIs, IoT) and physical (manipulation, na…
ResearchEnvBench: Benchmarking Agents on Environment Synthesis for Research Code Execution
Yubang Wang, Chenxi Zhang, Bowen Chen +7
Autonomous agents are increasingly expected to support scientific research, and recent benchmarks report progress in code repair and autonomous experimentation. However, these eval…
What Makes Position Zero Special? A Mechanistic Study of Position Zero Attention Sinks in LLMs
Runyu Peng, Ruixiao Li, Mingshu Chen +5
Transformers frequently allocate disproportionate attention to specific tokens, a phenomenon known as attention sinks. Causal large language models reliably form one at position ze…
AgentLongBench: A Controllable Long Benchmark For Long-Contexts Agents via Environment Rollouts
Shicheng Fang, Yuxin Wang, Xiaoran Liu +6
The evolution of Large Language Models (LLMs) into autonomous agents necessitates the management of extensive, dynamic contexts. Current benchmarks, however, remain largely static,…
EvoCUA: Evolving Computer Use Agents via Learning from Scalable Synthetic Experience
Taofeng Xue, Chong Peng, Mianqiu Huang +13
The development of native computer-use agents (CUA) represents a significant leap in multimodal AI. However, their potential is currently bottlenecked by the constraints of static…