Nucleotide dependency analysis turns genomic language models into functional element detectors
jmschreiber91 · x · 2026-10-21
A Nature Genetics paper introduces nucleotide dependencies, quantifying how substituting one genomic position changes attributions across others, enabling discovery of functional elements from genomic language models (gLMs). The team also shipped FastISM, which speeds up saturation mutagenesis for models with incomplete-receptive-field layers (e.g., convolutions), now the default with automatic fallback. Open access, 36k accesses and 29 citations already.
More from Research
- UC Berkeley releases Actor-Gatherer framework decoupling reasoning from context management — adityagp · 2026-10-21
- UW × NVIDIA paper: Compositional RL trains a 27B LLM to be the best teammate in multi-model systems — shangbinfeng · 2026-10-21
- C-RL details: LOO baseline cancels out in GRPO, no extra rollouts needed — shangbinfeng · 2026-10-21
- C-RL-trained 27B model becomes strongest collaborator in 21 of 24 settings — shangbinfeng · 2026-10-21
- Solo RL can make LLMs worse teammates: Shapley-based Compositional RL trains for collaboration — shangbinfeng · 2026-10-21
- Solo RL can make LLMs better alone but worse together, Compositional RL paper finds — shangbinfeng · 2026-10-21