Recursively Summarizing Enables Long-Term Dialogue Memory in LLMs

Re:Recursively Summarizing Enables Long-Term Dialogue Memory... by yieldcrv on Saturday September 02, 2023 @01:43PM (#37363724)

7 people got together and wrote this paper

Can someone summarize why this is different/better/more important than what we’ve seen 4 months ago with AutoGPT or even longer ago with the guy who got ChatGPT to make a compression algorithm that other ChatGPT sessions could read and resume conversations from

[ Parent | on HN ]

Re:Recursively Summarizing Enables Long-Term Dialogue Memory... by codefreakxff on Saturday September 02, 2023 @02:04PM (#37363924)

I haven't read it yet to be certain, but my impression is that rather than using lossless compression of entire chats, this uses lossy summarization to get the gist of chats. There will be tradeoffs between the two methods, hopefully this paper covers that.

[ Parent | on HN ]

Re:Recursively Summarizing Enables Long-Term Dialogue Memory... by keskival on Saturday September 02, 2023 @03:16PM (#37364582)

It seems to be exactly the same as what everyone has been doing for a long time already, and they fail in citing existing implementations.

[ Parent | on HN ]