dongsheng zhu

dongsheng

·

AI & ML interests

nlp, cv

Recent Activity

authored a paper about 7 hours ago

AgentCompass: A Unified Evaluation Infrastructure for Agent Capabilities

upvoted a paper 4 days ago

AgentCompass: A Unified Evaluation Infrastructure for Agent Capabilities

submitted a paper about 1 month ago

When Tools Fail: Benchmarking Dynamic Replanning and Anomaly Recovery in LLM Agents

View all activity

Organizations

dongsheng 's papers 7

arxiv:2607.13705

arxiv:2606.05806

arxiv:2603.25040

arxiv:2511.08487

arxiv:2508.15763

arxiv:2508.15754

arxiv:2501.12432