Unlocking LLM Creativity in Science through Analogical Reasoning

Andrew Shen, Shaul Druckmann, James Zou

arXiv:2605.11258 · 2026-05-13 공개 · arXiv · PDF

llm mode-collapse diversity-metrics biomedicine cell-communication brain-region-interactions oligonucleotide-prediction analogical-reasoning

Abstract

Autonomous science promises to augment scientific discovery, particularly in complex fields like biomedicine. However, this requires AI systems that can consistently generate novel and diverse solutions to open-ended problems. We evaluate LLMs on the task of open-ended solution generation and quantify their tendency to mode collapse into low-diversity generations. To mitigate this mode collapse, we introduce analogical reasoning (AR) as a new approach to solution generation. AR generates analogies to cross-domain problems based on shared relational structure, then uses those analogies to search for novel solutions. Compared to baselines, AR discovers significantly more diverse generations (improving solution diversity metrics by 90-173%), generates novel solutions over 50% of the time (compared to as little as 1.6% for baselines), and produces high-quality analogies. To validate the real-world feasibility of AR, we implement AR-generated solutions across four biomedical problems, yielding consistent quantitative gains. AR-generated approaches achieve a nearly 13-fold improvement on distributional metrics for perturbation effect prediction, outperform all baselines on AUPRC when predicting cell-cell communication, infer brain region interactions with a high Spearman correlation ($\rho$=0.729) to published methods, and establish state-of-the-art performance on 2 datasets for oligonucleotide property prediction. The novel and diverse solutions produced by AR can be used to augment the search space of existing solution generation methods.

한국어 요약

📋 한 줄 요약

**[과학 AI / 창의성]** 모드 붕괴(mode collapse)에 빠진 LLM의 과학적 솔루션 생성을 cross-domain 유추(analogical reasoning, AR)로 풀어 다양성·신규성·도메인 성능을 동시에 끌어올리는 새로운 솔루션 생성 기법 제안.

🎯 핵심 기여도

💡 핵심 아이디어

LLM의 mode collapse는 같은 도메인 내에서 가능성을 좁히기 때문이다. cross-domain analogy를 통해 다른 분야 문제의 관계 구조를 매핑해오면, 동일 도메인에서는 떠올리지 못할 신규 접근을 체계적으로 탐색할 수 있다.

🔬 기술적 접근법

📊 주요 결과

💭 의의 및 한계

**의의**: LLM을 단순 답변 생성기를 넘어 과학적 발견의 창의성 보조 도구로 격상시키는 일반 기법 제시. **한계**: 평가가 생물의학 도메인에 편중되며, analogy의 품질·타당성 검증을 위한 전문가 개입이 여전히 필요.

🚀 실용적 활용