# OmicsLM preprint reports gains in interpreting gene expression measurements

_Published Tuesday, October 27, 2026 at 9:05 AM EDT · Science, AI · Latest · Tier 2 — Notable_

Researchers report that OmicsLM, a 9B multimodal language model, outperformed evaluated alternatives across all six variants of a benchmark testing interpretation and comparison of gene expression measurements. The preprint describes a model that combines compact numerical representations of biological samples with natural-language instructions, allowing multiple samples in the same context.

Training included 20 billion tokens of continued pretraining and 5.21 million supervised conversations. The researchers introduced GEO-OmicsQA, a benchmark containing 3,000 questions drawn from real studies. Tests favored continuous numerical representations over text-only gene lists, although OmicsLM led on three of four held-out annotation tasks.

## Sources

- [cs.AI updates on arXiv.org](https://arxiv.org/abs/2605.06728)

---
Canonical: https://techandbusiness.org/newswire/LIvZMiQOL1bWDvrQWSvwD-
Published: 2026-10-27T13:05:59.672Z
Story chronology: 2026-10-27T04:00:00.000Z
Retrieved: 2026-10-27T16:35:56.493Z
Publisher: Tech & Business (techandbusiness.org)