LiveCodeBench

The 'LiveCodeBench' dataset/benchmark contains a collection of coding tasks and is used to evaluate the performance of large language models in generating correct and robust code solutions.

Papers using LiveCodeBench (17)

Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models2026Sakana Fugu Technical Report2026Don't Let Gains FADE: Breaking Down Policy Gradient Weights in RL2026DecompRL: Solving Harder Problems by Learning Modular Code Generation2026Learning to Orchestrate Agents in Natural Language with the Conductor2025ATLAS: Agentic Test-time Learning-to-Allocate Scaling2026EvoSyn: Generalizable Evolutionary Data Synthesis for Verifiable Learning2025CodeHacker: Automated Test Case Generation for Detecting Vulnerabilities in Competitive Programming Solutions2026ReVeal: Self-Evolving Code Agents via Reliable Self-Verification2025ACE: Self-Evolving LLM Coding Framework via Adversarial Unit Test Generation and Preference Optimization2026CollabCoder: Plan-Code Co-Evolution via Collaborative Decision-Making for Efficient Code Generation2026Think Anywhere in Code Generation2026Team of Thoughts: Efficient Test-time Scaling of Agentic Systems through Orchestrated Tool Calling2026TRINITY: An Evolved LLM Coordinator2025A Self-improving Coding Agent2025SEW: Self-Evolving Agentic Workflows for Automated Code Generation2025Pangu Embedded: An Efficient Dual-system LLM Reasoner with Metacognition2025