Lin Jing

Agents trained on procedurally generated, harder environments significantly outperform those trained on original benchmarks, even with identical compute and optimization.

Software engineering agents learn significantly more from failed attempts by conditioning recovery trajectories on explicit verifier feedback, boosting resolution rates by up to 10% on SWE-bench.

Industrial code generation gets a reasoning boost: InCoder-32B-Thinking leverages error-driven feedback and a code world model to achieve top-tier performance on complex hardware-aware tasks.

Code LLMs can achieve SOTA performance in agentic tasks by explicitly modeling the dynamic evolution of software logic across different training stages.

A new 32B code LLM trained specifically for industrial tasks crushes existing models on specialized domains like chip design and GPU kernel optimization, while remaining competitive on general coding benchmarks.