Skip to content
conftrace_
2026 EMNLP EMNLP 2026

LogitsCoder: Towards Efficient Chain-of-Thought Path Search via Logits Preference Decoding for Code Generation

AbstractCode generation requires executable reasoning, making test-time chain-of-thought (CoT) search an attractive strategy for improving solution quality. However, existing CoT scaling methods often face a tension between underthinking, where shallow reasoning misses important algorithmic cases, and overthinking, where verbose or redundant reasoning consumes substantial tokens without improving correctness. We propose LogitsCoder, a lightweight framework that revisits CoT search from the perspective of logits-level control. LogitsCoder improves efficient reasoning search through two complementary mechanisms: Logits Preference Decoding adaptively calibrates local token expansion using public-test feedback, while Logits Rank Based Path Selection explores and selects compact high-confidence reasoning trajectories based on logit-level stability. Together with Thoughts Aggregation, LogitsCoder refines reasoning paths without relying on costly MCTS-style tree expansion. Extensive experiments demonstrate that LogitsCoder produces more efficient and higher-quality reasoning chains, leading to superior code generation performance compared to baseline methods.

More from EMNLP 2026

All EMNLP 2026 papers →