研究人员推出CogScale，一个旨在高效评估AI架构序列处理能力的新基准。该基准包含14个可扩展的合成任务，允许在进行大量训练之前快速验证新设计。使用CogScale进行的初步评估测试了包括GRU、LSTM、Mamba和Transformer变体在内的七种不同架构，涵盖了各种参数预算和难度级别。 AI
影响 能够更快地迭代和验证用于序列数据处理的新型AI架构。
排序理由 该集群包含一篇介绍用于评估AI模型的新基准的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2605.19758v1 Announce Type: cross Abstract: The ability to maintain and manipulate information over time is a fundamental aspect of living beings and Artificial Intelligence. While modern models have achieved remarkable success in tasks like natural language processing, eva…
The ability to maintain and manipulate information over time is a fundamental aspect of living beings and Artificial Intelligence. While modern models have achieved remarkable success in tasks like natural language processing, evaluating the capacity of novel architectures to pro…