Initialization Improves LLM-Driven Discovery

The paper investigates how the set of prior iterates influences success in large language model (LLM)-driven discovery tasks. It introduces 12 new harnesses called 'Modular' and evaluates them on five diverse discovery problems, revealing that success is fragile and highly dependent on harness design. The study identifies mode collapse—a sharp loss of iterate diversity—as a common failure, and shows that early discoveries predict final outcomes, leading to a new initialization strategy that consistently improves performance across harnesses and applications.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

• llms
• benchmarks