MAT-Cell: A Multi-Agent Tree-Structured Reasoning Framework for Batch-Level Single-Cell Annotation
Yehui Yang, Zelin Zang, Changxi Chi, Jingbo Zhou, Xienan Zheng + 7 more
TLDR
MAT-Cell is a neuro-symbolic, multi-agent framework that uses tree-structured reasoning for robust single-cell annotation, outperforming SOTA.
Key contributions
- Reframes single-cell annotation into constructive, verifiable proof generation, moving beyond black-box classification.
- Injects symbolic biological constraints via adaptive RAG to ground neural reasoning and reduce transcriptomic noise.
- Employs dialectic verification with rebuttal agents to audit reasoning paths and enforce logical consistency.
- Significantly outperforms SOTA models on large-scale and cross-species benchmarks, robust in challenging scenarios.
Why it matters
Existing single-cell annotation methods face issues like poor generalization and spurious associations. MAT-Cell resolves these by integrating biological priors and a rigorous multi-agent verification process, leading to more accurate and verifiable cell state identification crucial for biological discovery.
Original Abstract
Automated cellular reasoning faces a core dichotomy: supervised methods fall into the Reference Trap and fail to generalize to out-of-distribution cell states, while large language models (LLMs), without grounded biological priors, suffer from a Signal-to-Noise Paradox that produces spurious associations. We propose MAT-Cell, a neuro-symbolic reasoning framework that reframes single-cell analysis from black-box classification into constructive, verifiable proof generation. MAT-Cell injects symbolic constraints through adaptive Retrieval-Augmented Generation (RAG) to ground neural reasoning in biological axioms and reduce transcriptomic noise. It further employs a dialectic verification process with homogeneous rebuttal agents to audit and prune reasoning paths, forming syllogistic derivation trees that enforce logical consistency.Across large-scale and cross-species benchmarks, MAT-Cell significantly outperforms state-of-the-art (SOTA) models and maintains robust per-formance in challenging scenarios where baselinemethods severely degrade. Code is available at https://gith ub.com/jiangliu91/MAT-Cell-A-Mul ti-Agent-Tree-Structured-Reasoni ng-Framework-for-Batch-Level-Sin gle-Cell-Annotation.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.