按年被提到
20242024 · 2 次2026
大模型与语言模型›长上下文与测试时计算·arXiv:2401.17505 · cs.LG

Arrows of Time for Large Language Models

Vassilis Papadopoulos, Jérémie Wenger, Clément Hongler · 2024
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2024-02-02 · 最近 2024-03-18
2不同的人
2活跃月份
2讨论串
5被引(OpenAlex)
arXiv 摘要(候选层不写导读)

We study the probabilistic modeling performed by Autoregressive Large Language Models (LLMs) through the angle of time directionality, addressing a question first raised in (Shannon, 1951). For large enough models, we empirically find a time asymmetry in their ability to learn natural language: a difference in the average log-perplexity when trying to predict the next token versus when trying to predict the previous one. This difference is at the same time subtle and very consistent across various modalities (language, model size, training time, ...). Theoretically, this is surprising: from an information-theoretic point of view, there should be no such difference. We provide a theoretical framework to explain how such an asymmetry can appear from sparsity and computational complexity considerations, and outline a number of perspectives opened by our results.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

没有评论直接链到它;下面是它被提交的讨论串里回复最多的评论(不计入统计)
讨论串评论nyoncore · 2024-02

“Isn't it obvious that since LLM are trained to predict the next word they do better than to predict the previous one?”

Arrows of Time for Large Language Models原评论 ↗
被提交为 story · 2 次