按年被提到
20222022 · 1 次2026
大模型与语言模型›长上下文与测试时计算·arXiv:2203.08913 · cs.LG

Memorizing Transformers

Yuhuai Wu, Markus N. Rabe, DeLesley Hutchins, Christian Szegedy · 2022
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2022-05-20 · 最近 2024-04-27
2不同的人
2活跃月份
2讨论串
90被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Language models typically need to be trained or finetuned in order to acquire new knowledge, which involves updating their weights. We instead envision language models that can simply read and memorize new data at inference time, thus acquiring new knowledge immediately. In this work, we extend language models with the ability to memorize the internal representations of past inputs. We demonstrate that an approximate kNN lookup into a non-differentiable memory of recent (key, value) pairs improves language modeling across various benchmarks and tasks, including generic webtext (C4), math papers (arXiv), books (PG-19), code (Github), as well as formal theorems (Isabelle). We show that the performance steadily improves when we increase the size of memory up to 262K tokens. On benchmarks including code and mathematics, we find that the model is capable of making use of newly defined functions and theorems during test time.

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

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论cs702 · 2024-04

“No. RAG is about finding relevant documents/paragraphs (via KNN lookups of their embeddings) and then inserting those documents/paragraphs into the input context, as sequences of input tokens. What I'm talking about is different: https://arxiv.org/abs/2203.08913”

What can LLMs never do?原评论 ↗
被提交为 story · 1 次