Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.11661 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916031633555456 |
|---|---|
| author | Jang, Yunhui Zhu, Lu Fawkes, Jake Denton, Alisandra Kaye Beaini, Dominique Noutahi, Emmanuel |
| author_facet | Jang, Yunhui Zhu, Lu Fawkes, Jake Denton, Alisandra Kaye Beaini, Dominique Noutahi, Emmanuel |
| contents | Large language models (LLMs) have recently gained significant attention as a promising approach to accelerate scientific discovery. However, their application in open-ended scientific domains such as biology remains limited, primarily due to the lack of factually grounded and actionable explanations. To address this, we introduce a structured explanation formalism for virtual cells that represents biological reasoning as mechanistic action graphs, enabling systematic verification and falsification. Building upon this, we propose VCR-Agent, a multi-agent framework that integrates biologically grounded knowledge retrieval with a verifier-based filtering approach to generate and validate mechanistic reasoning autonomously. Using this framework, we release VC-TRACES dataset, which consists of verified mechanistic explanations derived from the Tahoe-100M atlas. Empirically, we demonstrate that training with these explanations improves factual precision and provides a more effective supervision signal for downstream gene expression prediction. These results underscore the importance of reliable mechanistic reasoning for virtual cells, achieved through the synergy of multi-agent and rigorous verification. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_11661 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Towards Autonomous Mechanistic Reasoning in Virtual Cells Jang, Yunhui Zhu, Lu Fawkes, Jake Denton, Alisandra Kaye Beaini, Dominique Noutahi, Emmanuel Machine Learning Artificial Intelligence Large language models (LLMs) have recently gained significant attention as a promising approach to accelerate scientific discovery. However, their application in open-ended scientific domains such as biology remains limited, primarily due to the lack of factually grounded and actionable explanations. To address this, we introduce a structured explanation formalism for virtual cells that represents biological reasoning as mechanistic action graphs, enabling systematic verification and falsification. Building upon this, we propose VCR-Agent, a multi-agent framework that integrates biologically grounded knowledge retrieval with a verifier-based filtering approach to generate and validate mechanistic reasoning autonomously. Using this framework, we release VC-TRACES dataset, which consists of verified mechanistic explanations derived from the Tahoe-100M atlas. Empirically, we demonstrate that training with these explanations improves factual precision and provides a more effective supervision signal for downstream gene expression prediction. These results underscore the importance of reliable mechanistic reasoning for virtual cells, achieved through the synergy of multi-agent and rigorous verification. |
| title | Towards Autonomous Mechanistic Reasoning in Virtual Cells |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2604.11661 |