AI Paper Atlas

Research map / Large language models / Instruction and fine-tuning

ReCode: Updating Code API Knowledge with Reinforcement Learning

Haoze Wu, Yunzhi Yao, Wenhao Yu, Ningyu Zhang

AAAI 2026 · Large language models › Instruction and fine-tuning · approach: Fine-tuning for reasoning data · 12 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add ReCode (DAPO) RL trainingQwen2.5-Coder-7B-Instruct · CodeUpdateArenaPass@1: 67.3 -> 78.7Improved
add ReCode (GRPO) RL trainingQwen2.5-Coder-7B-Instruct · CodeUpdateArenaPass@1: 67.3 -> 74.6Improved
add ReCode (DAPO) RL trainingDS-v1.5-Coder-7B-Instruct · CodeUpdateArenaPass@5: 72.5 -> 78.2Improved
replace SFT with ReCode (DAPO)DS-v1.5-Coder-7B-Instruct · CodeUpdateArenaPass@1: 53.4 -> 63.6Improved
replace SFT with ReCode (DAPO) to preserve general abilityQwen2.5-Coder-7B-Instruct · HumanEval+Pass@1: 70.2 -> 81.7Improved
add syntax checking (AST) to ES reward (ablation)ReCode GRPO · CodeUpdateArena reward ablationPass@1: +5.4 -> +7.4Improved
add syntax checking (AST) to EM reward (ablation)ReCode GRPO · CodeUpdateArena reward ablationPass@1: -1.2 -> +1.1Improved
use edit-similarity (ES*) instead of exact match (EM*) reward (ablation)ReCode GRPO · CodeUpdateArena reward ablationPass@5: +2.0 -> +8.0Improved
add correctness reward to format-only reward (ablation)ReCode GRPO · CodeUpdateArena reward ablationPass@1: -2.3 -> +7.4Improved
add ReCode to model on multiple API updatesQwen2.5-Coder-7B-Instruct · 20 multi-API-update questionsPass@1: 35 -> 60Improved
fix CodeUpdateArena test-case errorsDS-Coder-1.5 · CodeUpdateArenaPass@1: 46.8 -> 59.1Improved
add ReCode (DAPO) on DS-Coder for general abilityDS-v1.5-Coder-7B-Instruct · HumanEval+Pass@1: 71.3 -> 68.9Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy