Performance Gain of Self-Invoking Code Generation Models in Multilingual MBPP-Pro Benchmarks

We introduce self-invoking code generation, a new task designed to evaluate the progressive reasoning and problem-solving capabilities of LLMs. In this task, models are presented with a base problem and a related, more complex problem. They must solve the base problem and then utilize its solution to address the more complex one. This work features three key contributions. First, we propose a general recipe for generating more challenging versions of existing benchmarks, resulting in three new benchmarks: HumanEval Pro, MBPP Pro, and BigCodeBench-Lite Pro, specifically designed to assess LLMs

Research goal: What is the relative performance gain (measured in pass@k accuracy) of models trained on self-invoking code generation versus traditional code generation tasks when evaluated on unseen programming languages in multilingual MBPP-Pro benchmarks?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.9/10.