AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Scaling Atomistic Protein Binder Design with Generative Pretraining and Test-Time Compute

Kieran Didi, Zuobai Zhang, Guoqing Zhou, Danny Reidenbach, Zhonglin Cao, Sooyoung Cha, Tomas Geffner, Christian Dallago, Jian Tang, Michael M. Bronstein, Martin Steinegger, Emine Kucukbenli and 2 more

ICLR 2026 · Probabilistic and scientific modeling › Molecules and proteins · approach: Generative protein structure design · 12 citations (OpenAlex)

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
add Teddymer synthetic dimer data to training (ablation)Complexa · protein binder benchmarkunique successesImproved
add global translation noise to the flow-matching objective (ablation)Complexa · protein binder benchmarkunique successes / binder placementImproved
add f_ipAE reward to beam search (ablation)Complexa beam search · protein targets# unique successes: 77.00 -> 83.36Improved
add interface H-bond reward to beam search (ablation)Complexa beam search · protein targetsavg # interface H-bonds: 5.271 -> 7.154Improved
combine f_ipAE and f_H-Bond rewards (ablation)Complexa beam search · protein targets# unique successes: 83.36 -> 86.26Improved
use structured search (beam search, FKS, MCTS) instead of best-of-N (ablation)Complexa inference-time scaling · hard protein targetsunique success rate vs GPU hoursImproved
use simple best-of-N instead of structured search (ablation)Complexa inference-time scaling · easy protein targetsunique success rate vs GPU hoursNo clear effect
initialize BindCraft from Complexa samples (G&H)BindCraft · easy protein targetssearch speedImproved
initialize BindCraft from Complexa samples (G&H)BindCraft · hard protein targetssearch speedNo clear effect
add ProteinMPNN re-design of sequences (ablation)Complexa · protein targetsavg # unique successes: 9.10 -> 14.4Improved
fix interface residues during MPNN re-design (ablation)Complexa · protein targetsavg # unique successes: 13.6 -> 14.4Mixed
extend inference-time search horizon beyond 100 GPU hoursComplexa · TNF-alpha, H1, IL17Aunique successes: 15 for TNF-alpha, 7 for H1, 1 for IL17AImproved

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