发表机构
Northwestern University; University of Cambridge; Johns Hopkins University; Columbia University; University of California, Los Angeles; Stanford University（西北大学; 剑桥大学; 约翰霍普金斯大学; 哥伦比亚大学; 加州大学洛杉矶分校; 斯坦福大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
NeuroLens利用JEPA框架的自监督模型，从慢性神经记录中学习去噪潜在表征，改善决策和语义解码，支持跨会话泛化与少样本适应，为研究神经表征动态提供新范式。
AI 中文摘要
理解神经活动如何表征高阶认知以及这些表征如何随时间演变，长期以来一直是神经科学的核心追求。然而，当前的分析工具难以区分慢性神经记录中的表征可塑性与记录不稳定性。在此，我们引入NeuroLens（神经语义的潜在嵌入），一种基于联合嵌入预测架构（JEPA）框架的自监督模型，从慢性神经记录中学习去噪的、语义信息丰富的潜在变量。自适应编码器将变化的神经群体映射到共同的潜在空间，而时间预测器学习支持未来潜在状态预测的结构。通过在潜在空间中进行预测，NeuroLens捕获了时间上可预测的结构，并降低了对瞬态、记录特异性变异的敏感性。在小鼠和人类的慢性皮层内数据中，学习到的表征改善了对决策和语义任务变量的解码。多日预训练能够泛化到未来的会话，快速少样本适应未见过的神经群体，并且随时间推移的解码稳定性优于最先进的基线。总之，这些结果确立了NeuroLens作为研究神经表征在学习过程中及长时间尺度上如何变化的新范式。
英文摘要
Understanding how neural activity represents higher-order cognition and how these representations evolve over time has long been a central pursuit in neuroscience. However, current analytical tools cannot easily distinguish representational plasticity from recording instability in chronic neural recordings. Here, we introduce NeuroLens (Latent Embeddings of Neural Semantics), a self-supervised model based on the Joint-Embedding Predictive Architecture (JEPA) framework that learns denoised, semantically informative latents from chronic neural recordings. An adaptive encoder maps changing neural populations into a common latent space, while a temporal predictor learns structure that supports prediction of future latent states. By predicting in latent space, NeuroLens captures temporally predictable structure and reduces sensitivity to transient, recording-specific variability. Across chronic intracortical data in mice and humans, the learned representations improve decoding of decision-making and semantic task variables. Multi-day pretraining enables generalization to future sessions, rapid few-shot adaptation to unseen neural populations, and more stable decoding over time than state-of-the-art baselines. Together, these results establish NeuroLens as a new paradigm for studying how neural representations change during learning and over long timescales.
