about
Training Reasoning Models to Generate Diverse Scientific Ideas (arxiv.org)
3 points by theanonymousone 30 days ago | hide | past | pdf | discuss on HN

In plain words: A model writes a set of research ideas at once, scoring each one higher when it is both good and different from earlier ones. This training let a small model beat two commercial systems on quality-plus-diversity, without hurting whether the ideas could be run.

Abstract

One of the first steps of a research pipeline is generating ideas. Large Language Models (LLMs) are increasingly used to automate this step, but most of this work optimizes the quality of individual ideas, while overlooking diversity. In our work, we study document-grounded ideation, focusing on increasing the diversity of LLM generated ideas and ensuring they are high quality. Given prior papers, a model must propose a set of ideas that are each high-quality and clearly distinct from one another, while also being grounded in the input documents. To optimize for this, we propose training with a set-level reinforcement learning objective. In our setup, in a single generation, the model writes K structured idea cards grounded in the source papers, and the set is scored jointly, where each idea's quality is weighted by its distance from the ideas written before it. This teaches the model to steer each new idea away from its predecessors. We show that our trained 8B model achieves higher creative utility (a measure combining quality and diversity) than Claude Opus 4.8 and GPT-6 Astra on our contamination-free benchmark. In a pilot human study with four experts, our sets are judged as promising as Claude Opus 4.8's and contain marginally more unique ideas. When a coding agent runs proof-of-concept experiments on the ideas, our model's success@k is comparable to Claude Opus 4.8's, so we see that the added diversity does not cost any executability.

Manya Wadhwa, Pranav Narayanan Venkit, Hiroaki Hayashi, Yu Li, Tuhin Chakrabarty, Chien-Sheng Wu
arXiv:2610.22324 · cs.CL · submitted Oct 15, 2026
abstract · pdf · html · Code and Data at:https://github.com/ManyaWadhwa/diverge_ideation | Under Review

add comment on HN