AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Reverse Distillation: Disentangling and Scaling Protein Language Model Representations

Darius Catrina, Christian Bepler, Samuel Sledzieski, Rohit Singh

ICLR 2026 · Large language models › Language, safety and interpretability · approach: Sign and cross-lingual translation

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
reverse-distill ESM-2 15B on smaller modelsESM-2 15B · ProteinGym 1 mut (28 datasets)Spearman: 0.899 -> 0.904Improved
reverse-distill ESM-2 3BESM-2 3B · ProteinGym 1 mut (28 datasets)Spearman: 0.881 -> 0.893Improved
reverse-distill ESM-2 650MESM-2 650M · ProteinGym 2 mut (28 datasets)Spearman: 0.672 -> 0.688Improved
reverse-distill ESM-2 3BESM-2 3B · ProteinGym >2 mut set, 1 mut (6 datasets)Spearman: 0.591 -> 0.574Hurt
reverse-distill ESM-2 15BESM-2 15B · R2/R1AUPR: 0.368 -> 0.468Improved
reverse-distill ESM-2 650MESM-2 650M · MIBAUPR: 0.881 -> 0.855Hurt
reverse-distill ESM-2 650MESM-2 650M · LOCAUPR: 0.709 -> 0.669Hurt
reverse-distill ESM-2 3BESM-2 3B · SSP Q3AUPR: 0.791 -> 0.816Improved
use progressive chain rather than direct 8M->650M distillation (ablation)rd.650M · ProteinGym ARGR_ECOLISpearman: 0.849 -> 0.858Improved
use progressive chain rather than direct 650M->3B distillation (ablation)rd.3B · ProteinGym ARGR_ECOLISpearman: 0.859 -> 0.873Improved
use principal component regression with Johnstone threshold instead of OLS (ablation)rd models · ProteinGym (37 datasets)Spearman win rate: PCR wins 58.8% (650M), 60.3% (3B), 63.2% (15B)Improved
reverse distillation vs concatenation + PCA (ablation)rd.15B · ProteinGym (37 datasets)Spearman win rate: wins 66.7% (650M), 75.8% (3B), 62.1% (15B)Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy