The Platonic brain bridge hypothesis is proposed: omni models, multimodal large language models that process video, audio and text jointly, converge on brain-like representations usable in both directions, and human brain organization is a usable architectural prior for multimodal large language models.
Abstract
Multimodal large language models predict brain activity, but brain alignment has been a measurement, not a design tool. We propose the Platonic brain bridge hypothesis: omni models, multimodal large language models that process video, audio and text jointly, converge on brain-like representations usable in both directions. From model to brain, brain-likeness of seven omni models is stable across participants, rises with every input channel in three bases, and our encoders lead the Algonauts 2025 out-of-distribution leaderboard. From brain to model, Brain-MoE fixes the expert partition of a frozen base to the seven networks of human cortex, trains experts on network-labelled Brain-AVQA questions, raises held-out accuracy in all 15 model-benchmark pairs by 6.42 percentage points on average and exceeds capacity-matched random experts in 14. Brain-Scope localizes the correspondence to sparse features whose removal weakens brain prediction. Human brain organization is therefore a usable architectural prior for multimodal large language models.
Large language models (LLMs) can predict human brain activity across a variety of brain regions during natural language comprehension. Typically, however, LLM-brain alignment is measured using one model for different regions of the brain, and then model performance is summarized across regions. This one-model-fits-all...
Zhivar Sourati, Meng-Hong Wu, Nona Ghazizadeh et al.· 0 citations
Alignment and causation dissociate: brain-aligned heads contribute to performance, but their removal is substantially less disruptive than removal of heads selected via attribution patching, and FV-ranked removal is substantially more disruptive than brain-ranked removal.
Christopher Pinier, Gustaw Opielka, Hannes Rosenbusch et al.· 0 citations
This retrospectively re-decoded 3,373 P300-speller selections from 47 people with amyotrophic lateral sclerosis, reconstructing a neural posterior and combining it with 25 language priors ranging from 5-grams to 46.7-billion-parameter models to reveal how strongly a fused brain-computer interface decision depends on it...
A. Gorenshtein, M. Omar, E. Jia et al.· medRxiv· 1 citation
Electroencephalography (EEG) foundation models are often pretrained with a fixed channel vocabulary or a limited set of montages, making transfer difficult when electrode layouts change. We propose CortexBridge, a lightweight adapter that combines EEG features with electrode and atlas coordinates to map arbitrary monta...
Jia-Zhen Hong, Xiao-Tian Zhou, Zi-Hao Ding et al.· 1 citation
Encoding models offer a principled framework for linking computational representations of language to neural activity, but most electroencephalography (EEG) evidence for brain–language alignment comes from tightly controlled, word-by-word reading paradigms. Whether such alignment is detectable during naturalistic readi...
Haorui Sun, D. Jangraw· bioRxiv· 0 citations
Related blog posts
Microsoft Research Blog· microsoft.comNov 16, 2026
AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduNov 13, 2026