人工智能2023 · 分享于 Hacker News
TinyStories: How Small Can Language Models Be and Still Speak Coherent English? (2023)
探索《TinyStories: How Small Can Language Models Be and Still Speak Coherent English? (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。
评论 #48617826 评论 #46204399 评论 #45996794 评论 #44901522 评论 #43517006 评论 #42869696 评论 #42427668 评论 #40533310 评论 #40533223 评论 #39052269 评论 #38977781 评论 #38897069 评论 #38834619 评论 #38420613 评论 #37937778 评论 #37838192 评论 #37607055 评论 #37484535 评论 #37183350 评论 #37090486 评论 #37078173 评论 #36896851 评论 #36634336 评论 #36541072 评论 #36514945 评论 #36513491 评论 #36489718 评论 #36414128 评论 #36413829 评论 #36316413 评论 #36316394 评论 #36314942 评论 #36106853 评论 #36042949 评论 #36009904 评论 #36002346
Ytimsuchanek ·
OpenAI says it has evidence DeepSeek used its model to train competitor
747 赞同1,541 评论
Yingve ·
What's the strongest AI model you can train on a laptop in five minutes?
571 赞同194 评论
Ygpjt ·
LLM from scratch, part 28 – training a base model from scratch on an RTX 3090
539 赞同121 评论
Ylappa ·
Phi-4: Microsoft's Newest Small Language Model Specializing in Complex Reasoning
439 赞同143 评论
Yjnord ·
“Imprecise” language models are smaller, speedier, and nearly as accurate
302 赞同169 评论
Ytzury ·
TinyStories: How Small Can Language Models Be and Still Speak Coherent English? (2023)
218 赞同104 评论
Ysimonw ·
Stuff we figured out about AI in 2023
210 赞同132 评论
YJPLeRouzic ·
Ask HN: How are Markov chains so different from tiny LLMs?
204 赞同189 评论
Ymaxrmk ·
Launch HN: Talc AI (YC S23) – Test Sets for AI
132 赞同47 评论
Ydanboarder ·
Ten Noteworthy AI Research Papers of 2023
128 赞同19 评论
YPaulHoule ·
Low responsiveness of ML models to critical or deteriorating health conditions
92 赞同53 评论
Ygeorgehill ·
Phi-2: Self-Extend Boosts Performance, Extends Context to 8k Without Training
89 赞同17 评论
Yfoobarqux ·
Textbooks are all you need
23 赞同5 评论
Youtrun86 ·
Consortium launched to build the largest open LLM
18 赞同5 评论
Yelsewhen ·
Mechanical Turk workers are using AI to automate being human
17 赞同0 评论
Y1wheel ·
Do Machine Learning Models Memorize or Generalize?
7 赞同0 评论
Yatg_abhishek ·
Outperforming LLMs with less training data and smaller model sizes
7 赞同0 评论
Y__vec__ ·
The tide is shifting: 1.3B outperforms 7B Llama 2
4 赞同2 评论
Ypat-jay ·
Apple bans ChatGPT use by employees, report says
3 赞同0 评论
Ytheafh ·
Some Neural Networks Learn Language Like Humans
3 赞同0 评论
Yhackandthink ·
Modern language models refute Chomsky’s approach to language
3 赞同1 评论
Ysurprisetalk ·
MonadGPT – What would have happened if ChatGPT was invented in the 17th century?
3 赞同0 评论
Ytosh ·
How Small Can Language Models Be and Still Speak Coherent English?
2 赞同0 评论
Ylordylord ·
Lojban: Constructed language to eliminate ambiguity from communication
2 赞同0 评论
Yda39a3ee ·
The Darwinian argument for worrying about AI
2 赞同0 评论
Ytuxguy ·
How to scale LLMs better with an alternative to transformers
2 赞同0 评论
Ylorepieri ·
Artificial General Intelligence – A gentle introduction
2 赞同0 评论
Y082349872349872 ·
Superintelligence: The Idea That Eats Smart People
1 赞同0 评论
YDanielKehoe ·
LLM Powered Autonomous Agents
1 赞同0 评论
Yandsoitis ·
Integrated Information Theory
1 赞同0 评论
Yog_kalu ·
Expanding Transformer size without losing function or starting from scratch
1 赞同0 评论
Yrbanffy ·
Disney Packed Big Emotion into a Little Robot
1 赞同0 评论
Ysurprisetalk ·
The 100k Whys of AI
1 赞同0 评论