← 研踪 PaperTrailarXiv:2607.18777 · v2
PertReason: A Knowledge-Grounded Benchmark and Framework for Cell-State-Conditioned Mechanistic Reasoning of Perturbation Effects
首次提交 2026-07-21 14:59
版本更新 2026-11-12 03:43
雷达观测 2026-11-13 14:16
摘要
Evaluating machine learning in scientific domains requires separating correct predictions from correct reasons under realistic distribution shifts. We introduce PertReason, a knowledge-grounded approach to evaluating and improving cell-state-conditioned reasoning about perturbation effects. Our framework comprises PertReasonQA, a benchmark that provides evaluation and supervision, and PertReasonLM, a reference large language model that demonstrates how such supervision supports reasoning-centric training. PertReasonQA tests whether models can generate mechanistically faithful explanations while remaining robust to complex shifts, such as new cells, unseen perturbations, cross-modal or combinatorial extrapolations. PertReasonQA combines single-cell genetic and chemical perturbation data across multiple cellular contexts with knowledge graphs, and dynamically conditions pathways on cell-specific basal states to avoid generic memorization. Evaluations of state-of-the-art models reveal systematic gaps between outcome prediction and mechanistic reasoning. As a reference probe of the benchmark, PertReasonLM is trained through supervised fine-tuning and reinforcement learning using mechanistic supervision from PertReasonQA. PertReasonLM improves predictive accuracy and reduces answer-reason disagreement by grounding rationales in context-specific pathways. Together, we provide a diagnostic framework for exposing and mitigating failures in faithful reasoning in data-rich scientific systems. We release our datasets to support further research at https://dongkwan-kim.github.io/PertReason/.
作者备注：NeurIPS 2026 E&D Camera Ready; 43 pages; Dongkwan Kim and Yiming Gao contributed equally to this work
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-12 03:43 · 最新观测版本
PertReason: A Knowledge-Grounded Benchmark and Framework for Cell-State-Conditioned Mechanistic Reasoning of Perturbation Effects
Evaluating machine learning in scientific domains requires separating correct predictions from correct reasons under realistic distribution shifts. We introduce PertReason, a knowledge-grounded approach to evaluating and improving cell-state-conditioned reasoning about perturbation effects. Our framework comprises PertReasonQA, a benchmark that provides evaluation and supervision, and PertReasonLM, a reference large language model that demonstrates how such supervision supports reasoning-centric training. PertReasonQA tests whether models can generate mechanistically faithful explanations while remaining robust to complex shifts, such as new cells, unseen perturbations, cross-modal or combinatorial extrapolations. PertReasonQA combines single-cell genetic and chemical perturbation data across multiple cellular contexts with knowledge graphs, and dynamically conditions pathways on cell-specific basal states to avoid generic memorization. Evaluations of state-of-the-art models reveal systematic gaps between outcome prediction and mechanistic reasoning. As a reference probe of the benchmark, PertReasonLM is trained through supervised fine-tuning and reinforcement learning using mechanistic supervision from PertReasonQA. PertReasonLM improves predictive accuracy and reduces answer-reason disagreement by grounding rationales in context-specific pathways. Together, we provide a diagnostic framework for exposing and mitigating failures in faithful reasoning in data-rich scientific systems. We release our datasets to support further research at https://dongkwan-kim.github.io/PertReason/.
查看此版本解读 →查看此版本原文 ↗