Egor Skopin
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
Software Eng.×1NLP×1
Frequent co-authors
Research Timeline
2026
Improving Small Language Models for Code Generation with Reinforcement Learning from Verification Feedback
The paper demonstrates that using Reinforcement Learning from Verifiable Rewards (RLVR) significantly improves small language models' functional correctness in code generation, particularly when combining unit-test and static-analysis feedback.
Highlighted terms show continued research focus across papers