2 papers
cs.CL2026
ReMCTS: Reflection-Enhanced Monte Carlo Tree Search for Code Generation
Huifei Wang, Xinying Huang, Yiheng Sun +1
Open-weight large language models (LLMs) can generate function-level programs from natural-language prompts, but plausible candidates still fail on hidden semantics and repeat mist…
cs.AI2026
When "Must" Becomes "Maybe": Erosion of Workflow Constraints in LLM Agent Handoffs
Yiheng Sun, Huifei Wang, Yancheng Zhu +3
Large language model (LLM) agents build on one another's work through summaries, plans, and other intermediate artifacts. These records let a downstream agent continue without revi…