The paper introduces OmicsBench, a new reasoning benchmark for multi‑omics sequences that includes 1,160 expert‑validated questions across DNA regulation, RNA processing, and protein function tasks, requiring traceable evidence chains. Evaluation of 17 large language models shows that scientific LLMs, while more accurate in classification, often lack valid evidence, suggesting shortcut learning. To address this, the authors propose tool‑augmented on‑policy distillation (TA‑OPD), a post‑training method that improves both evidence grounding and predictive performance across five Qwen3.5 models of varying sizes.
By Jie Ying, Zhefan Wang, Zihong Chen, Zhengqing Li, Jinzhe Li, Gang Li, Jian Liu, Fang Hu, Tao Luo, Zhonghang Yuan, Wanli Ouyang, Stan Z. Li, Fan Yang, Nanqing Dong
arXiv:2608. 07437v1 Announce Type: new Abstract: Reliable hypothesis testing is the foundation of many empirical scientific claims.
By Jiacheng Miao, Jin Mu, Guanhua Chen, James Zou
arXiv:2606. 01042v1 Announce Type: cross Abstract: Perturbation experiments are central to understanding cellular mechanisms, but remain costly and sparse, motivating prediction of gene expression responses for unobserved conditions.
By Xinyu Yuan, Xixian Liu, Jianan Zhao, Yashi Zhang, Hongyu Guo, Jian Tang
arXiv:2510. 17532v2 Announce Type: replace-cross Abstract: Predicting cancer treatment outcomes requires models that are both accurate and interpretable, particularly in the presence of heterogeneous clinical data.
By Raghu Vamshi Hemadri, Geetha Krishna Guruju, Kristi Topollai, Anna Ewa Choromanska
The study evaluates whether biological reasoning models actually use their biological inputs by testing six models on DNA, protein, and single‑cell tasks. By perturbing one biological input while keeping others fixed, the authors find that many models (e.g., Evo2, ESM3, BioReason, BioReason‑Pro) rely primarily on textual information, with minimal impact from the biological representations. In contrast, models like ChatNT, Prot2Text‑V2, CellWhisperer, and Cell2Sentence‑Scale show greater dependence on their biological inputs, yet overall accuracy gains do not consistently reflect increased biological input contribution.
By Ada Fang, Nikitha Thoduguli, Lukas Fesser, Hanlin Zhang, Sham M. Kakade, Marinka Zitnik
ForeSCITE is a framework that uses large language models to analyze temporal patterns in biomedical conclusions. It clusters published conclusions, estimates a direction that separates earlier from more recent findings within each cluster, and projects conclusion embeddings along these directions. The projected coordinates are decoded into natural‑language hypotheses, and retrospective evaluations show that ForeSCITE expands semantic coverage, introduces novel biomedical entity pairs, and identifies plausible CAR‑T hypotheses that experts deem biologically relevant.
By Chia-Hsuan Chang, Brian Ondov, Jiang Lv, Lan Jiang, Yunmai Wang, Haerin Song, Xueqing Peng, Qiaozhu Mei, Xiaolei Su, Hua Xu
The paper introduces HypoKG, a unified biochemical knowledge graph built from KEGG, Rhea, and UniProt, and uses it to benchmark 13,200 biomedical hypotheses generated by six large language models (LLMs). By varying the biological information provided—source enzyme only, full biological path, or source and disease endpoint—the study finds that LLMs produce higher-scoring hypotheses when given minimal information, but these are less evidence‑grounded. When supplied with the full biological path, the models generate hypotheses that align more closely with known mechanistic relationships, a phenomenon the authors term evidence‑disciplined reasoning, which is confirmed by shuffling intermediate path steps.
"whyItMatters":"The study demonstrates that knowledge graphs can both uncover novel disease–enzyme pairs and guide LLMs to reason more accurately from evidence, improving the reliability of AI‑generated biomedical hypotheses."
By Dominic Okonkwo, Adetayo Okunoye, Ismailcem Budak Arpinar
arXiv:2606. 10279v1 Announce Type: new Abstract: Supervised fine-tuning with synthetic rationale data is widely assumed to improve language model performance on clinical prediction tasks by teaching models not just what to predict but why.
By Buxin Su, Bingxuan Li, Cheng Qian, Yiwei Wang, Jin Jin, Bingxin Zhao
AI co-scientists have expanded the use of large language models for generating scientific hypotheses and guiding experimental investigation. Here we investigate whether temporal patterns in the semant...
arXiv:2608. 15669v1 Announce Type: new Abstract: Scientific discovery often involves optimising expensive-to-evaluate objectives over vast, structured, and open-ended hypothesis spaces, such as molecules, protein sequences, and computer programs.
By Zhongwei Yu, Yan Song, Xue Yan, Anjie Liu, Xingyu Lu, Yihang Chen, Huichi Zhou, Siyuan Guo, Luoyang Sun, Sihan Chen, Xiangning Yu, Jun Wang
Supervised fine-tuning with synthetic rationale data is widely assumed to improve language model performance on clinical prediction tasks by teaching models not just what to predict but why. We test this assumption on five-year Alzheimer's disease and related dementias (ADRD) prediction from longitudinal health histories.
The paper introduces PRISMS, a framework that uses expert pairwise rankings of varying fidelity to curate scientific designs without relying on data-intensive regression models. By escalating queries from lower- to higher-fidelity rankers based on Fisher-information, PRISMS improves discovery recall and reduces the number of screening rounds compared to regression-only and non‑escalated ranking methods. In optimization tasks, PRISMS outperforms Bayesian optimization by achieving higher hypervolume.
By Kevin Tirta Wijaya, Alston Lo, Michael Sun, Wojciech Matusik, Vahid Babaei