8 papers
Independent Patch Verification for Coding Agents with a Bidirectional Reconstruct-and-Verify Framework
Chenglin Li, Yisen Xu, Zehao Wang +3
Autonomous coding agents powered by large language models can now generate code patches directly from bug reports, but a fundamental gap remains: once a patch is produced, no mecha…
Turning Interaction History into Execution State: A Runtime Layer for Long-Horizon Coding Agents
Zehao Wang, Yisen Xu, Chenglin Li +5
Long-horizon coding agents accumulate hundreds of actions and observations in their trajectories, yet nothing in this record indicates which observations still describe the reposit…
Preventing Premature Commitment in Coding Agents with an Evidence-Conditioned Execution Layer
Yisen Xu, Chenglin Li, Zehao Wang +2
LLM-based coding agents often edit source code or submit patches before examining enough repository evidence to justify the change, a failure pattern we call premature commitment.…
Probe to Generate: Program Variant-Guided Test Augmentation for Repository-Level Repair Benchmarks
Chenglin Li, Yisen Xu, Zehao Wang +3
Test-based benchmarks such as SWE-bench have become a standard basis for evaluating automated issue resolution agents, deeming a patch correct if it passes a provided regression te…
Safe Reinforcement Learning with Preference-based Constraint Inference
Chenglin Li, Grant Ruan, Hua Geng
Safe reinforcement learning (RL) is a standard paradigm for safety-critical decision making. However, real-world safety constraints can be complex, subjective, and even hard to exp…
From Historical Patches to Repair Plans: Outcome-Conditioned Reasoning for Repository-Level Program Repair
Chenglin Li, Yisen Xu, Zehao Wang +3
Repository-level automated program repair (APR) requires long-horizon reasoning over interdependent decisions. However, most LLM-based approaches reconstruct repair reasoning indep…
What Makes Code Generation Ethically Sourced?
Zhuolin Xu, Chenglin Li, Qiushi Li +1
Several code generation models have been proposed to help reduce time and effort in solving software-related tasks. To ensure responsible AI, there are growing interests over vario…
Polygames: Improved Zero Learning
Tristan Cazenave, Yen-Chi Chen, Guan-Wei Chen +21
Since DeepMind's AlphaZero, Zero learning quickly became the state-of-the-art method for many board games. It can be improved using a fully convolutional structure (no fully connec…