人工智能2023 · 分享于 Hacker News
Lost in the Middle: How Language Models Use Long Contexts (2023)
探索《Lost in the Middle: How Language Models Use Long Contexts (2023)》的原文与 Hacker News 社区讨论，查看来源并加入阅读清单。
从论文，走向讨论
看看社区如何讨论这个想法。
来自 Hacker News 的原始讨论。不同的视角、疑问，以及值得继续思考的线索。
最多展示 100 条相关讨论。论文可能出现在评论中，并非原帖投稿内容。
在讨论中发现的引用
按 HN 条目顺序展示最近的 50 条引用来源。
Ytosh ·
Codestral Mamba
485 赞同138 评论
Yaraghuvanshi ·
LLMs and the Harry Potter problem
65 赞同61 评论
Ygzer0 ·
GPT-4 Update: 32K Context Window Now for All Users
53 赞同26 评论
YspIrr ·
Handbook.md shows that long policy documents do not reliably govern agents
45 赞同19 评论
Yxinweihe ·
Show HN: TraceRoot – Open-source agentic debugging for distributed services
40 赞同22 评论
Ygk1 ·
You probably don’t need to fine-tune LLMs
16 赞同0 评论
Ydavidbarker ·
Introducing ChatGPT Enterprise
10 赞同5 评论
Ywilliamzeng0 ·
Launch HN: Sweep (YC S23) – A bot to create simple PRs in your codebase
9 赞同1 评论
Yja3k ·
Benchmarking GPT-4 Turbo – A Cautionary Tale
9 赞同0 评论
Yshishirpatil ·
MemGPT – LLMs with self-editing memory for unbounded context
6 赞同0 评论
Ybobvanluijt ·
Explainer Video – Lost in the Middle: How Language Models Use Long Context
4 赞同3 评论
Yaverylamp ·
Llama 32K Context Released by Together AI
3 赞同0 评论
Ykevin_hu ·
New models and developer products announced at OpenAI DevDay
3 赞同0 评论
Yrobertnishihara ·
A Comprehensive Guide for Building Rag-Based LLM Applications
2 赞同0 评论
YHeidaradar ·
Microsoft says GPT 3.5 has 20B parameters?
2 赞同0 评论
Ystuhlmueller ·
Build a search engine, not a vector DB
2 赞同0 评论
Ywslh ·
Lost in the Middle: How Language Models Use Long Contexts (2023)
2 赞同0 评论
Ygk1 ·
Lost in the Middle: How Language Models Use Long Contexts
1 赞同0 评论
Yjojodoi ·
Language models use long contexts
1 赞同0 评论
Yandrewon ·
What's new in Llama 2 and how to run it locally
1 赞同0 评论
YAmazingTurtle ·
OpenAI reduces Codex Model Context Size from 372k to 272k
1 赞同0 评论