conftrace_

Yongding Tao

2 papers · 2026–2026 · 1 conference · across top CS/AI conferences

Conferences

ACL (2)

Top co-authors

Zhi Jin (2) Xue Jiang (2) Yihong Dong (2) Ge Li (2) Binhua Li (1) Deng Hongyi (1) Yingwei Ma (1) Rongyu Cao (1) Kechi Zhang (1) Lili Mou (1)

Keywords

reinforcement learning with verifiable reward (2) multiple importance sampling (1) execution semantics alignment (1) variable level execution trajectory (1) hybrid-policy optimization (1) capability boundary collapse (1) code generation (1) exploration-based advantage function (1)

Papers

CODERL+: Improving Code Generation via Reinforcement with Execution Semantics Alignment ACL 2026 RL-PLUS: Countering Capability Boundary Collapse of LLMs in Reinforcement Learning with Hybrid-policy Optimization ACL 2026