AI Paper Atlas

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

TRIBE: TRImodal Brain Encoder for whole-brain fMRI response prediction

Stéphane d'Ascoli, Jérémy Rapin, Yohann Benchetrit, Hubert Banville, Jean-Remi King

ICLR 2026 · Vision-language and multimodal › Human and embodied modeling · approach: fMRI visual decoding

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 video modality to text-only encoder (single to multimodal) (ablation)TRIBE · Validation encoding scoremean Pearson correlation: 0.22 -> 0.30 (text+video)Improved
add third modality (audio) to best bimodal model (ablation)TRIBE · Validation encoding scoremean Pearson correlation: 0.30 -> 0.31Improved
use multimodal instead of best unimodal encoder in primary visual cortex (ablation)TRIBE · Primary visual cortex parcelsencoding scoreHurt
use multimodal instead of best unimodal encoder in associative cortices (ablation)TRIBE · Prefrontal and parieto-occipito-temporal parcelsencoding score: up to 30% increaseImproved
train a multi-subject model instead of separate per-subject models (ablation)TRIBE · Validation encoding scoremean Pearson correlation: 0.29 -> 0.31Improved
add the transformer encoder over time (ablation)TRIBE · Validation encoding scoremean Pearson correlation: 0.23 -> 0.31Improved
increase number of training sessions (ablation)TRIBE · Multimodal encoder scalingencoding scoreImproved
increase LLM context length for text features up to 1024 words (ablation)TRIBE text-only · Text encoder scalingencoding scoreImproved
evaluate out-of-distribution instead of in-distribution (ablation)TRIBE · Friends S7 vs OOD moviesmean Pearson correlation: 0.3195 -> 0.2146Hurt

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