AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Multimodal Protein Language Models for Enzyme Kinetic Parameters: From Substrate Recognition to Conformational Adaptation

Fei Wang, Xinye Zheng, Kun Li, Yanyan Wei, Yuxin Liu, Ganpeng Hu, Tong Bao, Jingwen Yang

CVPR 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Medical vision-language models · 3 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 MRCA substrate cross-attention to the PLM (ablation)ESM2-3B PLM baseline · Ki predictionR2: 0.49 -> 0.51Improved
add G-MoE geometry-aware experts after MRCA (ablation)ERBA · Ki predictionR2: 0.51 -> 0.54Improved
add ESDA distribution alignment after G-MoE (ablation)ERBA · Ki predictionR2: 0.54 -> 0.61Improved
use MRCA instead of concat and self-attention fusion (ablation)ERBA · Ki predictionR2: 0.47 -> 0.61Improved
use G-MoE instead of standard MoE (ablation)ERBA · Ki predictionR2: 0.50 -> 0.61Improved
use ESDA objective instead of plain L2 loss (ablation)ERBA · Ki predictionR2: 0.48 -> 0.61Improved
use MRCA instead of concat and self-attention fusion (ablation)ERBA · Ki predictionRMSE: 1.44 -> 1.26Improved
fuse enzyme, then substrate, then structure instead of enzyme, structure, substrate (ablation)ERBA · Ki predictionR2: 37.8% R2 improvement, 20% RMSE reductionImproved
attach ERBA to ESM2-3B backboneESM2-3B · kcat predictionR2: 0.41 -> 0.54Improved
attach ERBA to ESM2-3B backboneESM2-3B · Km predictionR2: 0.43 -> 0.61Improved
attach ERBA to ESM2-8M backboneESM2-8M · kcat predictionR2: 0.19 -> 0.27Improved
attach ERBA to Ankh3-5.7B backboneAnkh3-5.7B · Ki predictionR2: 0.46 -> 0.57Improved

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