Protein modeling thread says the real question is small MSA vs. big model
sokrypton · x · 2026-07-22
The post argues that the real comparison is small MSA vs. big model in protein modeling.
It says the biggest gains over pLMs appear in cases where there are very few sequences available. For large protein families, the practical baseline is often simply downloading related MSAs from AFDB and stacking your sequence on top.
The attached table highlights examples where MSA Pairformer / ESMC shows the largest ΔP@L improvements alongside the corresponding MSA depth values, suggesting that sequence context availability strongly shapes where the method helps most.
Related event: Protein Modeling: Small MSA vs Large Language Models(2 posts)→
More from Research
- HarnessMM cuts agent tokens per task by up to 88% while raising success rates — omarsar0 · 2026-10-26
- Ben Recht: The world needs a truly open language model, and it's easier to build than you think — beenwrekt · 2026-10-26
- New Blog Breaks Black-Box Distillation Into Five Practical Steps — chrisrohlf · 2026-10-26
- DetectorScope turns black-box AI text detectors into auditable ones with sparse autoencoders — _akhaliq · 2026-10-26
- OPST paper shows LLMs self-improve with zero human data, guided only by topic descriptions — _akhaliq · 2026-10-26
- Tencent's HiSTEVE 1.0: hierarchical VLA agent plays Minecraft with native keyboard and mouse, hits 85% SOTA — _akhaliq · 2026-10-26