简洁导航，深度对齐：2025年Mouse vs. AI比赛的获胜方案
Navigating Simply, Aligning Deeply: Winning Solutions for Mouse vs. AI 2025
AI总结：
本文提出了一种简洁的架构和深度的ResNet-like架构，分别在视觉鲁棒性和神经对齐方面取得优异成绩，挑战了传统模型复杂性假设。
摘要：
视觉鲁棒性和神经对齐仍然是开发能够匹配生物视觉系统的人工智能代理的关键挑战。我们展示了Team HCMUS_TheFangs在NeurIPS 2025 Mouse vs. AI：鲁棒视觉觅食比赛两个赛道中的获胜方法。在第一赛道（视觉鲁棒性）中，我们证明了通过结合架构简洁性和针对性组件，可以获得更优的泛化能力，最终得分为95.4%，使用轻量化的两层CNN，通过门控线性单元和观察归一化进行增强。在第二赛道（神经对齐）中，我们开发了一种深度的类似于ResNet的架构，具有16个卷积层和基于GLU的门控机制，实现了具有1780万参数的顶级神经预测性能。我们对十个在60000到1140000步之间训练的模型检查点进行系统分析，发现训练时间与性能之间存在非单调关系，最佳结果出现在约200000步时。通过全面的消融研究和失败案例分析，我们提供了关于为什么更简单的架构在视觉鲁棒性方面表现优异，而更深层次的模型随着容量增加在神经对齐方面表现更好的见解。我们的结果挑战了关于视觉运动学习中模型复杂性的传统假设，并为开发鲁棒、生物启发的视觉代理提供了实用指导。
英文摘要：
Visual robustness and neural alignment remain critical challenges in developing artificial agents that can match biological vision systems. We present the winning approaches from Team HCMUS_TheFangs for both tracks of the NeurIPS 2025 Mouse vs. AI: Robust Visual Foraging Competition. For Track 1 (Visual Robustness), we demonstrate that architectural simplicity combined with targeted components yields superior generalization, achieving 95.4% final score with a lightweight two-layer CNN enhanced by Gated Linear Units and observation normalization. For Track 2 (Neural Alignment), we develop a deep ResNet-like architecture with 16 convolutional layers and GLU-based gating that achieves top-1 neural prediction performance with 17.8 million parameters. Our systematic analysis of ten model checkpoints trained between 60K to 1.14M steps reveals that training duration exhibits a non-monotonic relationship with performance, with optimal results achieved around 200K steps. Through comprehensive ablation studies and failure case analysis, we provide insights into why simpler architectures excel at visual robustness while deeper models with increased capacity achieve better neural alignment. Our results challenge conventional assumptions about model complexity in visuomotor learning and offer practical guidance for developing robust, biologically-inspired visual agents.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
