AI Paper Atlas

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

Learn from Your Mistakes: Tree-like Self-Play for Secure Code LLMs

Wenqi Chen, Ziyan Zhang, Wang Bin, Lin Liu, Hengheng Zhang, Zhengsu Chen

ICML 2026 · Large language models › Instruction and fine-tuning · approach: Fine-tuning generalization and delta compression · 0 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 tree-structured risk-node generation to self-play (ablation)TSP / CodeLlama-7B · SecurityEval (Python)SPR@1: 69.6 -> 75.8Improved
add tree-structured risk-node generation to self-play (ablation)TSP / CodeLlama-7B · DiverseVul C/C++total vulnerabilities: 103 -> 94Improved
add tree-structured risk-node generation to self-play (ablation)TSP / Qwen2.5-Coder-7B · DiverseVul C/C++total vulnerabilities: 142 -> 57Improved
add tree-structured risk-node generation to self-play (ablation)TSP / Qwen2.5-Coder-3B · DiverseVul C/C++total vulnerabilities: 130 -> 41Improved
train with TSP instead of base modelCodeLlama-7B · SecurityEval (Python)SPR@1: 55.0 -> 75.8Improved
train with TSP instead of SFT on same dataQwen2.5-Coder-7B · Unseen CWEsvulnerabilities: 58 -> 26Improved
apply SFT on C/C++ dataQwen2.5-Coder-7B · Unseen CWEsvulnerabilities: 30 -> 58Hurt
train with TSP instead of SFT on C/C++ unseen CWEsCodeLlama-7B · Unseen CWEsvulnerabilities: 50 -> 27Improved
add TSP training, effect on general codingCodeLlama-7B · HumanEvalpass@1: 34.5 -> 34.0No clear effect
add TSP training, effect on general codingQwen2.5-Coder-7B · HumanEval (Table 2 run)pass@1: 51.6 -> 58.9Improved
add self-play without trees to SafeCoder-level baseline (ablation)CodeLlama-7B · HumanEvalpass@10: 54.7 -> 44.9Hurt
TSP versus Self-Play on local-flow CWE-215 (debug-mode info)TSP · DiverseVul C/C++ CWE-215vulnerabilities: 38 -> 4Improved

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