人工智能2017 · 分享于 Hacker News
Deep Neural Networks – Attention is all you need (No recurrence or convolutions)
探索《Deep Neural Networks – Attention is all you need (No recurrence or convolutions)》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 #48873508 评论 #48822131 评论 #48822006 评论 #48436763 评论 #48382277 评论 #48215641 评论 #48059139 评论 #47992246 评论 #47987089 评论 #47943346 评论 #47930406 评论 #47487097 评论 #47462584 评论 #47341224 评论 #47171292 评论 #46318907 评论 #45658232 评论 #45532233 评论 #45392281 评论 #44867417 评论 #44610990 评论 #44240447 评论 #43970816 评论 #43200834 帖子正文 #43111112 评论 #42950308 评论 #42929946 评论 #42877874 评论 #42790725 评论 #42536345 评论 #41678118 评论 #41648422 帖子正文 #41491707 评论 #40987860 评论 #40893841 评论 #40861443 评论 #40527679 评论 #40398434 评论 #39869345 评论 #39868798 评论 #39033916 评论 #38735854 评论 #38696726 评论 #38642103 评论 #38453371 评论 #38397988 评论 #38309078 评论 #38052662 评论 #37795107 帖子正文 #37614690
Ytkellogg ·
S1: A $6 R1 competitor?
851 赞同416 评论
Ysixhobbits ·
Claude Code is all you need
851 赞同504 评论
Ygoing_ham ·
DreamFusion: Text-to-3D using 2D Diffusion
833 赞同201 评论
Ymmq ·
Yann LeCun, Geoffrey Hinton and Yoshua Bengio win Turing Award
773 赞同146 评论
Yluizfelberti ·
Ask HN: What's the best paper you've read in 2020?
705 赞同192 评论
Ypseudolus ·
DOJ accuses Visa of monopoly that affects price of 'nearly everything’
610 赞同464 评论
Ytamnd ·
LLMs can get "brain rot"
473 赞同293 评论
Yamrrs ·
SmolGPT: A minimal PyTorch implementation for training a small LLM from scratch
434 赞同55 评论
Yyigitdemirag ·
Better Language Models and Their Implications
426 赞同130 评论
Ythrow0101c ·
AI capex is so big that it's affecting economic statistics
365 赞同355 评论
Ydeafcalculus ·
Exploring LSTMs
349 赞同44 评论
Yredbell ·
The great software quality collapse or, how we normalized catastrophe
314 赞同253 评论
Ysh_tomer ·
Ilya Sutskever: “If you learn all of these, you’ll know 90% of what matters”
313 赞同110 评论
Yisof4ult ·
How AlphaChip transformed computer chip design
301 赞同194 评论
Ymakaimc ·
Building an end-to-end Speech Recognition model in PyTorch
288 赞同33 评论
Yusefulposter ·
Don't post generated/AI-edited comments. HN is for conversation between humans.
287 赞同86 评论
Yandrew3726 ·
Transformer: A Novel Neural Network Architecture for Language Understanding
280 赞同19 评论
YGupie ·
I asked GPT-3 for the question to “42”
276 赞同229 评论
Yepoch_100 ·
OpenAI Releases Largest GPT-2 Text Generation Model
266 赞同158 评论
Ysmhx ·
Biological Function Emerges from Unsupervised Learning on 250M Protein Sequences
251 赞同49 评论
Yskanderbm ·
MDMA reopens child-like “critical periods” in the brain
245 赞同134 评论
Ytracyhenry ·
Regularization is all you need: simple neural nets can excel on tabular data
216 赞同93 评论
Ymrcgnc ·
Markov Chains Explained Visually (2014)
213 赞同26 评论
Ychmaynard ·
How the Brain Creates a Timeline of the Past
188 赞同23 评论
Yjalammar ·
Show HN: Ecco – See what your NLP language model is “thinking”
185 赞同34 评论
Yyoloswagins ·
Extracting Structured Data from Recipes Using Conditional Random Fields (2015)
167 赞同20 评论
Ycalepayson ·
Darwin Machines
164 赞同82 评论
Ydebdut ·
The Illustrated Transformer (2018)
162 赞同11 评论
Yeunos ·
Tensor Product Attention Is All You Need
160 赞同104 评论
Ymilomg ·
T5Gemma 2: The next generation of encoder-decoder models
149 赞同28 评论
Yjocker12 ·
Machine Learning Can't Handle Long-Term Time-Series Data
148 赞同64 评论
YTonyTrapp ·
Windows 11 calls a zip file a 'postcode file' in UK English
139 赞同47 评论
Ysachaarbonel ·
Building a semantic search engine in Rust
136 赞同27 评论
Ylaurex ·
Can Demis Hassabis save Google?
133 赞同127 评论
Ywaxpancake ·
The Bitcoin whitepaper is hidden in every copy of macOS
130 赞同37 评论
Yocean_moist ·
TransMLA: Multi-head latent attention is all you need
123 赞同32 评论
Ynshm ·
Microsoft’s New MT-DNN Outperforms Google BERT
121 赞同38 评论
Yhhs ·
Why do we remember some life moments but not others?
119 赞同58 评论
Yhexhowells ·
Formal Algorithms for Transformers
106 赞同13 评论
Yt55 ·
ArXiv LaTeX Cleaner: Clean the LaTeX code of your paper to submit to ArXiv
103 赞同42 评论
Ythm ·
Is A.I. The Death of I.P.?
100 赞同297 评论
YZeljkoS ·
Attention and Augmented Recurrent Neural Networks
99 赞同4 评论
Ymoondistance ·
New AI Training Technique Is Drastically Faster, Says Google
84 赞同38 评论
Yelbasti ·
Transformer Attention is off by one
79 赞同10 评论
Yprotomok ·
DLVM: A modern compiler framework for neural network DSLs
75 赞同17 评论
Ybeefman ·
The potential of transformers in reinforcement learning
70 赞同23 评论
Yfrozenseven ·
Reinforcement Pre-Training
70 赞同18 评论
Ysamwillis ·
Anthropic - Introducing 100K Token Context Windows, Around 75,000 Words
69 赞同24 评论
Yfuzztester ·
Trump sides with Musk on support for H-1B visas for foreign tech workers
67 赞同109 评论
Ybelter ·
How to install Windows 11 on a device that does not meet Windows 11 requirements
65 赞同72 评论
YSai_ ·
Ask HN: Can someone ELI5 Transformers and the “Attention is all we need” paper
59 赞同19 评论
Ycbeach ·
Microsoft’s new text-to-speech model can duplicate anyone's voice in 3 seconds
49 赞同33 评论
Ysonabinu ·
Transformer: A Neural Network Architecture for Language Understanding (2017)
29 赞同3 评论
Yfoobarqux ·
Textbooks are all you need
23 赞同5 评论
Yyarapavan ·
You Don't Know Jack about Application Performance
23 赞同6 评论
Yflipped ·
Dirtyfrag: Universal Linux LPE
23 赞同0 评论
YnoAI ·
We aren't close to creating a rapidly self-improving AI
22 赞同7 评论
Yjerpint ·
Show HN: ArXiv-txt, LLM-friendly ArXiv papers
22 赞同11 评论
Ydanboarder ·
Strangely Accurate AI Predictions from Blurry Medical Scans Alarms Researchers
21 赞同13 评论
Yrerx ·
The Illustrated Transformer
17 赞同2 评论
Ybehnamoh ·
If you replace the "x" with a "5" in an ArXiv URL, you get an HTML version
13 赞同6 评论
Yalok-g ·
Deep Neural Networks – Attention is all you need (No recurrence or convolutions)
12 赞同0 评论
Yveerd ·
Scaffolded LLMs as natural language computers
11 赞同1 评论
Ytomcam ·
What Are Transformer Models and How Do They Work?
11 赞同1 评论
Yidibidiartists ·
Attention Is All You Need (Neural Networks)
8 赞同3 评论
Ystock_toaster ·
Apple Sues OpenAI for Stealing Trade Secrets to Build AI Hardware
8 赞同0 评论
YEGreg ·
Google internally developed chatbots like ChatGPT years ago
7 赞同2 评论
Ythinxer ·
Sparks of Artificial General Intelligence: Early Experiments with GPT-4
6 赞同1 评论
Ymelvinmelih ·
Show HN: Turn Any ArXiv Paper into a 200-Page Prerequisite Reading Book
6 赞同3 评论
YBluestein ·
Attention Is All You Need
6 赞同2 评论
Ymfiguiere ·
Microsoft and OpenAI Working on ChatGPT-Powered Bing in Challenge to Google
5 赞同2 评论
Ysh_tomer ·
Gitlab AI is going head to head with GitHub Copilot
5 赞同0 评论
Ytbruckner ·
Gemini Pro achieves accuracy slightly inferior to GPT 3.5 Turbo
5 赞同0 评论
Yshcheklein ·
Attention Is All You Need
4 赞同0 评论
Ymariuz ·
Understanding and Coding the Self-Attention Mechanism of Large Language Models
4 赞同0 评论
Ymetalwhale ·
Open source solution replicates ChatGPT training process
4 赞同0 评论
Ykdavis ·
Attention Is All You Need
3 赞同0 评论
Yjonbaer ·
Attention Is All You Need
3 赞同0 评论
Ykordlessagain ·
Attention Is All You Need
3 赞同0 评论
Ywebmaven ·
'I Worked on Google's AI. My Fears Are Coming True'
3 赞同1 评论
Yjostmey ·
Revealing example of self-attention, the building block of transformer AI models
3 赞同0 评论
Yryanworl ·
S3 Express Is All You Need
3 赞同0 评论
Yandriosr ·
Attention is all you need paper is offline due to invalid certificate
3 赞同0 评论
Ysimonebrunozzi ·
Attention is all you need (2017)
3 赞同0 评论
Yshaunistyping ·
Open Weights Kill the Moat
3 赞同0 评论
Ynicklovescode ·
Attention is All You Need by Google Brain
2 赞同1 评论
Y0x1997 ·
Attention Is All You Need
2 赞同0 评论
Ymax_ ·
Attention Is All You Need [pdf]
2 赞同0 评论
Yalan-stark ·
What were the 40 research papers on the list Ilya Sutskever gave John Carmack?
2 赞同0 评论
Ynadermx ·
The paper that made ChatGPT possible
2 赞同0 评论
Yvemgar ·
AI / ML / LLM / Transformer Models Timeline
2 赞同0 评论
Ysrijan4 ·
ChatGPT: It has no inner monologue or meta-cognition
2 赞同1 评论
Yrasbt ·
Why the Original Transformer LLM Figure Is Wrong, and Other Interesting Tidbits
2 赞同0 评论
Ymarban ·
The AI Boom Has Silicon Valley on Another Manic Quest to Change the World
2 赞同0 评论
YAnon84 ·
What Is a Transformer Model?
2 赞同0 评论
Ysergiotapia ·
Ask HN: Why did all major companies release AI at the same time?
2 赞同6 评论
Ypasswordoops ·
About That OpenAI "Breakthrough"
2 赞同1 评论
YRyanShook ·
Attention Is All You Need
2 赞同0 评论
Yiamskeole ·
How Will OpenAI Compete?
2 赞同0 评论
Ychistev ·
Ask HN: What are the greatest discoveries in the last few years?
2 赞同0 评论