BrainWideBench is presented, a benchmark for evaluating across-animal transfer on multi-region neural recordings, built on the International Brain Laboratory Brainwide Map dataset of neural and behavioral recordings spanning 276 brain regions from 139 mice performing a sensory-guided decision-making task.
Abstract
Advances in large-scale neural recording have made it possible to collect data across many animals and distributed brain regions, raising the question of whether this scale can be exploited to learn general-purpose neural representations transferable across diverse downstream tasks. Yet, progress toward this goal has been limited by fragmented evaluation protocols and a narrow focus on individual task domains. Here, we present BrainWideBench, a benchmark for evaluating across-animal transfer on multi-region neural recordings, built on the International Brain Laboratory Brainwide Map dataset of neural and behavioral recordings spanning 276 brain regions from 139 mice performing a sensory-guided decision-making task. The benchmark is organized around three complementary task suites that evaluate whether learned representations support downstream decoding of behavior, can predict masked or future neural activity, and can recover biologically meaningful anatomical organization. With this benchmark, we systematically evaluate pretraining methods across transfer settings, including finetuning on downstream objectives and zero-shot generalization to unseen animals. Our results confirm pretraining improves performance over matched single-session baselines, but we show current methods exhibit heterogeneity in transfer capabilities: gains depend strongly on the alignment between pretraining objectives and downstream tasks. No single approach performs uniformly well across all three suites, and most methods are designed to only address a subset of them. Together, these findings suggest that learning representations that jointly generalize across behavior, dynamics, and anatomy remains an open challenge. By providing a unified and reproducible evaluation suite, BrainWideBench establishes a framework for measuring progress toward general-purpose models of the mouse brain. Code is available at this link.
This work proposes MAPA, an otherwise vanilla masked autoencoder with two spatial encodings, an anatomical region embedding and a relative positional encoding that together enable it to learn neural representations that transfer to unseen subjects and across various tasks.
Ben-Ting Tang, Z. Spalding, G. Cogan· arXiv.org· 0 citations
iMINDBench is introduced, an iEEG Multi-Institution Neural Decoding Benchmark that evaluates models on a shared suite of fifteen decoding tasks across three naturalistic movie-watching datasets and finds that the evaluated pretrained systems generally outperform baselines within their respective preprocessing tracks, w...
Geeling Chau, Saba Hashemi, Yonghyeon Gwon et al.· arXiv.org· 0 citations
Findings support organizing fMRI pretraining and adaptation by measured learning relations rather than treating domains and tasks as independent flat sets.
Jun-Feng Xia, Wen-Hao Ye, Jun-Xiang Zhang et al.· 2 citations
NeuroGate, a framework for cross-regional neural signal translation that recovers an unrecorded region’s waveform from a recorded one, is introduced, demonstrating two applications: quantifying how recording redundancy bounds the information from each added source, and a reusable perturbation protocol for verifying dir...
Ali Zareh, Rufeyda Yağcı, M. K. Özdemir· bioRxiv· 0 citations
NeuroPB is introduced, a framework that scales neural decoding by transferring knowledge from pretrained behavioral representations and achieves performance comparable to pretraining on macaque trajectories, demonstrating that transferable kinematic structure is shared across biological and artificial models.
Lu-Yao Jin, Yonghao Song, Huan Zhao et al.· 0 citations
iBrain is introduced, a unified foundation model that jointly learns from iEEG and spiking activity that consistently outperforms single-signal pretraining baselines and achieves state-of-the-art performance on multiple benchmarks.
Ying Chen, Ti-Ou Wang, Zhi-Feng Yue· 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