Search papers, labs, and topics across Lattice.
This paper introduces a compete-then-collaborate framework for training a coding student (Qwen2.5-Coder) using four frontier AI teachers (Claude, Codex-GPT, Grok, Gemini) ranked through execution-based judging. The study reveals that while all teachers achieve near-perfect performance on standard problems, their effectiveness varies significantly on more challenging tasks, with Gemini outperforming the others. Notably, the collaborative curriculum approach using reinforcement learning with verifiable rewards leads to substantial performance gains for the student, demonstrating that AI-teacher collaboration enhances learning through active problem-solving rather than mere imitation.
AI-teacher collaboration outperforms imitation learning, boosting student performance by nearly 50% on challenging coding tasks.
Large language models increasingly serve as teachers generating training data for smaller students. Prior multi-teacher knowledge distillation methods merge outputs without determining which frontier model teaches best, often relying on an LLM judge biased toward its own outputs. We introduce a compete-then-collaborate framework where four frontier AI teachers (Claude, Codex-GPT, Grok, Gemini) are ranked head-to-head by an execution-based judge (unit tests and stdin-stdout checks) with fairness controls, and then collaborate to build a verifiable curriculum for a student (Qwen2.5-Coder). We report three findings. (1) Under execution verification, all teachers solve standard problems near-perfectly after self-correction (99-100%) due to a saturation effect, but harder competition problems separate them (Gemini 77%>Claude 69% = Codex 69%>Grok 50%); however, the robust student-side results do not depend on teacher ranking. (2) Imitation (SFT) on verified solutions does not improve, and can degrade, an already-competent student at 7B and 32B (e.g., from 76.7% to 72.7% on MBPP-test, and 5.9% to 2.9% on competition problems). (3) Using the same collaborative curriculum as a reinforcement learning with verifiable rewards (RLVR) environment improves the student (from 5.9% to 8.8% peak on competition problems, a +49% relative gain), reversing SFT's direction. The value of AI-teacher collaboration lies not in pooling answers to imitate, but in jointly constructing a verifiable environment where the student learns by doing. We release a reproducible on-prem pipeline (NVIDIA GB10) with framework patches for running GRPO on a bleeding-edge stack.
