# paper-reading

**Repository Path**: greener/paper-reading

## Basic Information

- **Project Name**: paper-reading
- **Description**: 李沐《论文精读》项目国内镜像
- **Primary Language**: Unknown
- **License**: Apache-2.0
- **Default Branch**: main
- **Homepage**: https://github.com/mli/paper-reading
- **GVP Project**: No

## Statistics

- **Stars**: 1
- **Forks**: 1
- **Created**: 2022-01-01
- **Last Updated**: 2022-11-10

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# 深度学习论文精读

## 录制完成的论文    

| 日期 | 标题 | 封面 | 时长 | 视频（播放数） |
| --: | -- | -- | --: | -- |
| 7/8/22 | 如何讲好故事、故事里的论点【[研究的艺术](https://press.uchicago.edu/ucp/books/book/chicago/C/bo23521678.html)·三】| <img src="imgs/craft_research_p3.jpg" width="200px"/> | 43:56 ||
| 7/1/22 | 明白问题的重要性【[研究的艺术](https://press.uchicago.edu/ucp/books/book/chicago/C/bo23521678.html)·二】| <img src="imgs/craft_research_p2.jpg" width="200px"/> | 1:03:40 |[![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV11S4y1v7S2)](https://www.bilibili.com/video/BV11S4y1v7S2/)|
| 6/24/22 | 跟读者建立联系【[研究的艺术](https://press.uchicago.edu/ucp/books/book/chicago/C/bo23521678.html)·一】 | <img src="imgs/craft_research_p1.jpg" width="200px"/> | 45:01 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1hY411T7vy)](https://www.bilibili.com/video/BV1hY411T7vy/) |
| 6/17/22 | [Zero](https://arxiv.org/pdf/1910.02054.pdf) 逐段精读 | <img src="imgs/zero.jpg" width="200px"/> | 52:21 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1tY411g7ZT)](https://www.bilibili.com/video/BV1tY411g7ZT/) |
| 6/10/22 | [DETR](https://arxiv.org/pdf/2005.12872.pdf) 逐段精读 | <img src="imgs/detr.jpg" width="200px"/> | 54:22 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1GB4y1X72R)](https://www.bilibili.com/video/BV1GB4y1X72R/) |
| 6/3/22 | [Megatron LM](https://arxiv.org/pdf/1909.08053.pdf) 逐段精读 | <img src="imgs/megatron_lm.jpg" width="200px"/> | 56:07 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1nB4y1R7Yz)](https://www.bilibili.com/video/BV1nB4y1R7Yz/) |
| 5/27/22 | [GPipe](https://proceedings.neurips.cc/paper/2019/file/093f65e080a295f8076b1c5722a46aa2-Paper.pdf) 逐段精读 | <img src="imgs/gpipe.jpg" width="200px"/> | 58:47 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1v34y1E7zu)](https://www.bilibili.com/video/BV1v34y1E7zu/) |
| 5/5/22 | [Pathways](https://arxiv.org/pdf/2203.12533.pdf) 逐段精读 | <img src="imgs/pathways.jpg" width="200px"/> | 1:02:13 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1xB4y1m7Xi)](https://www.bilibili.com/video/BV1xB4y1m7Xi/) |
| 4/28/22 | [视频理解论文串讲](https://arxiv.org/pdf/2012.06567.pdf)（下） | <img src="imgs/video-survey-p2.jpg" width="200px"/> | 1:08:32 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV11Y411P7ep)](https://www.bilibili.com/video/BV11Y411P7ep/) <br />[![](https://img.shields.io/youtube/views/J2YC0-k57NM?style=social)](https://youtu.be/J2YC0-k57NM) |
| 4/21/22 | [参数服务器（Parameter Server）](https://www.usenix.org/system/files/conference/osdi14/osdi14-paper-li_mu.pdf) 逐段精读 | <img src="imgs/ps.jpg" width="200px"/> | 1:37:40 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1YA4y197G8)](https://www.bilibili.com/video/BV1YA4y197G8/) |
| 4/14/22 | [视频理解论文串讲](https://arxiv.org/pdf/2012.06567.pdf)（上） | <img src="imgs/video-survey-p1.jpg" width="200px"/> | 51:15 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1fL4y157yA)](https://www.bilibili.com/video/BV1fL4y157yA/) <br />[![](https://img.shields.io/youtube/views/gK7AGO6okhc?style=social)](https://youtu.be/gK7AGO6okhc) |
| 3/31/22 | [I3D](https://arxiv.org/pdf/1705.07750.pdf) 论文精读 | <img src="imgs/i3d.jpg" width="200px"/> | 52:31 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1tY4y1p7hq)](https://www.bilibili.com/video/BV1tY4y1p7hq/) <br />[![](https://img.shields.io/youtube/views/9lIkKiAn6uE?style=social)](https://youtu.be/9lIkKiAn6uE) |
| 3/24/22 | 斯坦福 2022 年 [AI 指数报告](https://aiindex.stanford.edu/wp-content/uploads/2022/03/2022-AI-Index-Report_Master.pdf) 精读 | <img src="imgs/ai_index_22.jpg" width="200px"/> | 1:19:56 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1s44y1N7eu)](https://www.bilibili.com/video/BV1s44y1N7eu/) <br />[![](https://img.shields.io/youtube/views/K8h_xjQ6ufY?style=social)](https://youtu.be/K8h_xjQ6ufY) |
| 3/17/22 | [AlphaCode](https://storage.googleapis.com/deepmind-media/AlphaCode/competition_level_code_generation_with_alphacode.pdf) 论文精读 | <img src="imgs/alphacode.jpg" width="200px"/> | 44:00 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1ab4y1s7rc)](https://www.bilibili.com/video/BV1ab4y1s7rc/) <br />[![](https://img.shields.io/youtube/views/t8Gzkca9pW4?style=social)](https://youtu.be/t8Gzkca9pW4) |
| 3/10/22 | [OpenAI Codex](https://arxiv.org/pdf/2107.03374.pdf) 论文精读 | <img src="imgs/codex.jpg" width="200px"/> | 47:58 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1iY41137Zi)](https://www.bilibili.com/video/BV1iY41137Zi/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1490959755963666432)](https://www.zhihu.com/zvideo/1490959755963666432)<br />[![](https://img.shields.io/youtube/views/oZriUGkQSNM?style=social)](https://youtu.be/oZriUGkQSNM) |
| 3/3/22 | [GPT](https://s3-us-west-2.amazonaws.com/openai-assets/research-covers/language-unsupervised/language_understanding_paper.pdf), [GPT-2](https://d4mucfpksywv.cloudfront.net/better-language-models/language_models_are_unsupervised_multitask_learners.pdf), [GPT-3](https://arxiv.org/abs/2005.14165) 精读 | <img src="imgs/gpt3.jpg" width="200px"/> | 1:29:58 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1AF411b7xQ)](https://www.bilibili.com/video/BV1AF411b7xQ/)<br />[![](https://img.shields.io/youtube/views/t70Bl3w7bxY?style=social)](https://youtu.be/t70Bl3w7bxY) |
| 2/24/22 | [Two-Stream](https://proceedings.neurips.cc/paper/2014/file/00ec53c4682d36f5c4359f4ae7bd7ba1-Paper.pdf) 逐段精读 |  <img src="imgs/twostream.jpg" width="200px"/> | 52:57 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1mq4y1x7RU)](https://www.bilibili.com/video/BV1mq4y1x7RU/)<br />[![](https://img.shields.io/youtube/views/vuqwKP2iDe0?style=social)](https://youtu.be/vuqwKP2iDe0) |
| 2/10/22 | [CLIP](https://openai.com/blog/clip/) 逐段精读 | <img src="imgs/clip.jpg" width="200px"/> | 1:38:25 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1SL4y1s7LQ)](https://www.bilibili.com/video/BV1SL4y1s7LQ/)<br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1475706654562299904)](https://www.zhihu.com/zvideo/1475706654562299904) <br />[![](https://img.shields.io/youtube/views/OZF1t_Hieq8?style=social)](https://youtu.be/OZF1t_Hieq8) |
| 2/6/22 | 你（被）吐槽过[论文不够 novel](https://perceiving-systems.blog/en/post/novelty-in-science) 吗？| <img src="imgs/novelty.jpg" width="200px"/> | 14:11 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1ea41127Bq)](https://www.bilibili.com/video/BV1ea41127Bq/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1475719090198876161)](https://www.zhihu.com/zvideo/1475719090198876161) |
| 1/23/22 | [AlphaFold 2](https://www.nature.com/articles/s41586-021-03819-2.pdf) 精读 | <img src="imgs/alphafold_2.jpg" width="200px"/> |  1:15:28 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1oR4y1K7Xr)](https://www.bilibili.com/video/BV1oR4y1K7Xr/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1469132410537717760)](https://www.zhihu.com/zvideo/1469132410537717760)  <br />[![](https://img.shields.io/youtube/views/Oy3OCoGUr-w?style=social)](https://youtu.be/Oy3OCoGUr-w) |
| 1/18/22 | 如何判断（你自己的）研究工作的价值 | <img src="imgs/research_value.jpg" width="200px"/> |  9:59 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1oL411c7Us)](https://www.bilibili.com/video/BV1oL411c7Us/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1475716940051869696)](https://www.zhihu.com/zvideo/1475716940051869696) |
| 1/15/22 | [Swin Transformer](https://arxiv.org/pdf/2103.14030.pdf) 精读 | <img src="imgs/swin_transformer.jpg" width="200px"/> | 1:00:21 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV13L4y1475U)](https://www.bilibili.com/video/BV13L4y1475U/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1466282983652691968)](https://www.zhihu.com/zvideo/1466282983652691968)   <br />[![](https://img.shields.io/youtube/views/luP3-Fs0QCo?style=social)](https://youtu.be/luP3-Fs0QCo) |
| 1/7/22 | [指导数学直觉](https://www.nature.com/articles/s41586-021-04086-x.pdf) | <img src="imgs/math_conj.jpg" width="200px"/> | 52:51 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1YZ4y1S72j)](https://www.bilibili.com/video/BV1YZ4y1S72j/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1464060386375299072)](https://www.zhihu.com/zvideo/1464060386375299072)  <br />[![](https://img.shields.io/youtube/views/czFGjvhtss8?style=social)](https://youtu.be/czFGjvhtss8) |
| 1/5/22 | AlphaFold 2 预告 | <img src="imgs/alphafold_2_preview.jpg" width="200px"/> | 03:28 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1Eu411U7Te)](https://www.bilibili.com/video/BV1Eu411U7Te/) |
| 12/20/21 | [对比学习](#contrastive_learning)论文综述 | <img src="imgs/contrastive.jpg" width="200px"/> | 1:32:01 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV19S4y1M7hm)](https://www.bilibili.com/video/BV19S4y1M7hm/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1460828005077164032)](https://www.zhihu.com/zvideo/1460828005077164032)  <br />[![](https://img.shields.io/youtube/views/1pvxufGRuW4?style=social)](https://www.youtube.com/watch?v=1pvxufGRuW4) |
| 12/15/21 | [MoCo](https://arxiv.org/pdf/1911.05722.pdf) 逐段精读 | <img src="imgs/mocov1.jpg" width="200px"/> | 1:24:11 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1C3411s7t9)](https://www.bilibili.com/video/BV1C3411s7t9/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1454723120678936576)](https://www.zhihu.com/zvideo/1454723120678936576)   <br />[![](https://img.shields.io/youtube/views/1pvxufGRuW4?style=social)](https://www.youtube.com/watch?v=1pvxufGRuW4) |
| 12/9/21 | 如何找研究想法 1 | <img src="imgs/mae_idea.jpg" width="200px"/> | 5:34 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1qq4y1z7F2)](https://www.bilibili.com/video/BV1qq4y1z7F2/) |
| 12/8/21 | [MAE](https://arxiv.org/pdf/2111.06377.pdf) 逐段精读 | <img src="imgs/mae.jpg" width="200px"/> | 47:04 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1sq4y1q77t)](https://www.bilibili.com/video/BV1sq4y1q77t/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1452458167968251904)](https://www.zhihu.com/zvideo/1452458167968251904)  <br />[![](https://img.shields.io/youtube/views/mYlX2dpdHHM?style=social)](https://youtu.be/mYlX2dpdHHM) |
| 11/29/21 | [ViT](https://arxiv.org/pdf/2010.11929.pdf) 逐段精读 | <img src="imgs/vit.jpg" width="200px"/> | 1:11:30 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV15P4y137jb)](https://www.bilibili.com/video/BV15P4y137jb/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1449195245754380288)](https://www.zhihu.com/zvideo/1449195245754380288)  <br />[![](https://img.shields.io/youtube/views/FRFt3x0bO94?style=social)](https://youtu.be/FRFt3x0bO94) |
| 11/18/21 | [BERT](https://arxiv.org/abs/1810.04805) 逐段精读 | <img src="imgs/bert.jpg" width="200px"/> | 45:49  | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1PL411M7eQ)](https://www.bilibili.com/video/BV1PL411M7eQ/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1445340200976785408)](https://www.zhihu.com/zvideo/1445340200976785408)  <br />[![](https://img.shields.io/youtube/views/ULD3uIb2MHQ?style=social)](https://youtu.be/ULD3uIb2MHQ) |
| 11/9/21 | [GAN](https://papers.nips.cc/paper/2014/file/5ca3e9b122f61f8f06494c97b1afccf3-Paper.pdf) 逐段精读 | <img src="imgs/gan.jpg" width="200px"/> | 46:16  | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1rb4y187vD)](https://www.bilibili.com/video/BV1rb4y187vD/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1442091389241159681)](https://www.zhihu.com/zvideo/1442091389241159681)  <br />[![](https://img.shields.io/youtube/views/g_0HtlrLiDo?style=social)](https://www.youtube.com/watch?v=g_0HtlrLiDo) |
| 11/3/21 | 零基础多图详解 [图神经网络](https://distill.pub/2021/gnn-intro/)（GNN/GCN） | <img src="imgs/gnn.jpg" width="200px"/> | 1:06:19 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1iT4y1d7zP)](https://www.bilibili.com/video/BV1iT4y1d7zP/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1439540657619087360)](https://www.zhihu.com/zvideo/1439540657619087360)  <br />[![](https://img.shields.io/youtube/views/sejA2PtCITw?style=social)](https://youtu.be/sejA2PtCITw) |
| 10/27/21 | [Transformer](https://arxiv.org/abs/1706.03762) 逐段精读<br> （视频中提到的文献 [^transformer]) |<img src="imgs/transformer.jpg" width="200px"/> | 1:27:05 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1pu411o7BE)](https://www.bilibili.com/video/BV1pu411o7BE/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1437034536677404672)](https://www.zhihu.com/zvideo/1437034536677404672)  <br />[![](https://img.shields.io/youtube/views/nzqlFIcCSWQ?style=social)](https://youtu.be/nzqlFIcCSWQ) |
| 10/22/21 | [ResNet](https://arxiv.org/abs/1512.03385) 论文逐段精读 | <img src="imgs/resnet-2.jpg" width="200px"/> | 53:46 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1P3411y7nn)](https://www.bilibili.com/video/BV1P3411y7nn/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1434795406001180672)](https://www.zhihu.com/zvideo/1434795406001180672)  <br />[![](https://img.shields.io/youtube/views/pWMnzCX4cwQ?style=social)](https://www.youtube.com/watch?v=pWMnzCX4cwQ) |
| 10/21/21 | 撑起计算机视觉半边天的 [ResNet](https://arxiv.org/abs/1512.03385) | <img src="imgs/resnet-1.jpg" width="200px"/> | 11:50 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1Fb4y1h73E)](https://www.bilibili.com/video/BV1Fb4y1h73E/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1434787226101751808)](https://www.zhihu.com/zvideo/1434787226101751808)  <br />[![](https://img.shields.io/youtube/views/NnSldWhSqvY?style=social)](https://www.youtube.com/watch?v=NnSldWhSqvY) |
| 10/15/21 | [AlexNet](https://papers.nips.cc/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf) 论文逐段精读 | <img src="imgs/alexnet-2.jpg" width="200px"/> | 55:21 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1hq4y157t1)](https://www.bilibili.com/video/BV1hq4y157t1/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1432354207483871232)](https://www.zhihu.com/zvideo/1432354207483871232)  <br />[![](https://img.shields.io/youtube/views/zjnxu8KUYKA?style=social)](https://www.youtube.com/watch?v=zjnxu8KUYKA) |
| 10/14/21 | 9年后重读深度学习奠基作之一：[AlexNet](https://papers.nips.cc/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf) | <img src="imgs/alexnet-1.jpg" width="200px"/> | 19:59 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1ih411J7Kz)](https://www.bilibili.com/video/BV1ih411J7Kz/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1432155856322920448)](https://www.zhihu.com/zvideo/1432155856322920448)  <br />[![](https://img.shields.io/youtube/views/vdYH0fE6thY?style=social)](https://www.youtube.com/watch?v=vdYH0fE6thY) |
| 10/06/21 | 如何读论文 | <img src="imgs/read-paper.jpg" width="200px"/> | 06:39 | [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1H44y1t75x)](https://www.bilibili.com/video/BV1H44y1t75x/) <br />[![zhihu](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=zhihu&query=video.play_count&url=https://www.zhihu.com/api/v4/zvideos/1428973951632969728)](https://www.zhihu.com/zvideo/1428973951632969728)  <br />[![](https://img.shields.io/youtube/views/txjl_Q4jCyQ?style=social)](https://www.youtube.com/watch?v=txjl_Q4jCyQ&list=PLFXJ6jwg0qW-7UM8iUTj3qKqdhbQULP5I&index=1) |

[^transformer]: 1 [斯坦福100+作者的200+页综述](https://arxiv.org/abs/2108.07258)，2 [对LayerNorm的新研究](https://arxiv.org/pdf/1911.07013.pdf)，3 [对Attention在Transformer里面作用的研究](https://arxiv.org/abs/2103.03404)


## 所有论文

包括已经录制完成和之后将要介绍的论文。选取的原则是10年内深度学习里有影响力文章（必读文章），或者近期比较有意思的文章。当然这十年里重要的工作太多了，不可能一一过一遍。在选取的时候我会偏向一些之前 [直播课](https://c.d2l.ai/zh-v2/) 中没讲到过的。 欢迎大家在 [讨论区](https://github.com/mli/paper-reading/discussions) 里提供建（点）议（歌）。

总论文数 67，录制完成数 32

（这里引用采用的是 semanticscholar，是因为它提供 [API](https://api.semanticscholar.org/api-docs/graph#operation/get_graph_get_paper) 可以自动获取，不用手动更新。）

### 计算机视觉 - CNN


| 已录制 | 年份 | 名字                                                         | 简介                 | 引用 |
| ------ | ---- | ------------------------------------------------------------ | -------------------- | ------------------------------------------------------------ |
| ✅      | 2012 | [AlexNet](https://papers.nips.cc/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf) | 深度学习热潮的奠基作                   | [![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fabd1c342495432171beb7ca8fd9551ef13cbd0ff%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/ImageNet-classification-with-deep-convolutional-Krizhevsky-Sutskever/abd1c342495432171beb7ca8fd9551ef13cbd0ff) |
| | 2014 | [VGG](https://arxiv.org/pdf/1409.1556.pdf) | 使用 3x3 卷积构造更深的网络                   | [![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Feb42cf88027de515750f230b23b1a057dc782108%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Very-Deep-Convolutional-Networks-for-Large-Scale-Simonyan-Zisserman/eb42cf88027de515750f230b23b1a057dc782108) |
| | 2014 | [GoogleNet](https://arxiv.org/pdf/1409.4842.pdf) | 使用并行架构构造更深的网络                   | [![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fe15cf50aa89fee8535703b9f9512fca5bfc43327%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Going-deeper-with-convolutions-Szegedy-Liu/e15cf50aa89fee8535703b9f9512fca5bfc43327) |
|  ✅  | 2015 |  [ResNet](https://arxiv.org/pdf/1512.03385.pdf) | 构建深层网络都要有的残差连接。               |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F2c03df8b48bf3fa39054345bafabfeff15bfd11d%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Deep-Residual-Learning-for-Image-Recognition-He-Zhang/2c03df8b48bf3fa39054345bafabfeff15bfd11d)  |
|  | 2017 | [MobileNet](https://arxiv.org/pdf/1704.04861.pdf) | 适合终端设备的小CNN                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F3647d6d0f151dc05626449ee09cc7bce55be497e%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/MobileNets%3A-Efficient-Convolutional-Neural-Networks-Howard-Zhu/3647d6d0f151dc05626449ee09cc7bce55be497e)  |
| | 2019 | [EfficientNet](https://arxiv.org/pdf/1905.11946.pdf) | 通过架构搜索得到的CNN                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F4f2eda8077dc7a69bb2b4e0a1a086cf054adb3f9%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/EfficientNet%3A-Rethinking-Model-Scaling-for-Neural-Tan-Le/4f2eda8077dc7a69bb2b4e0a1a086cf054adb3f9)  |
| | 2021 |  [Non-deep networks](https://arxiv.org/pdf/2110.07641.pdf) | 让不深的网络也能在ImageNet刷到SOTA                   | [![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F0d7f6086772079bc3e243b7b375a9ca1a517ba8b%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Non-deep-Networks-Goyal-Bochkovskiy/0d7f6086772079bc3e243b7b375a9ca1a517ba8b) |

### 计算机视觉 - Transformer

| 已录制 | 年份 | 名字                                                         | 简介                 | 引用 |
| ------ | ---- | ------------------------------------------------------------ | -------------------- | ------------------------------------------------------------ |
| ✅ | 2020 | [ViT](https://arxiv.org/pdf/2010.11929.pdf) | Transformer杀入CV界                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F7b15fa1b8d413fbe14ef7a97f651f47f5aff3903%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/An-Image-is-Worth-16x16-Words%3A-Transformers-for-at-Dosovitskiy-Beyer/7b15fa1b8d413fbe14ef7a97f651f47f5aff3903)  |
| ✅ | 2021 |  [CLIP](https://openai.com/blog/clip/) | 图片和文本之间的对比学习                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F6f870f7f02a8c59c3e23f407f3ef00dd1dcf8fc4%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Learning-Transferable-Visual-Models-From-Natural-Radford-Kim/6f870f7f02a8c59c3e23f407f3ef00dd1dcf8fc4)  |
| ✅ | 2021 | [Swin Transformer](https://arxiv.org/pdf/2103.14030.pdf) | 多层次的Vision Transformer                   | [![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fc8b25fab5608c3e033d34b4483ec47e68ba109b7%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Swin-Transformer%3A-Hierarchical-Vision-Transformer-Liu-Lin/c8b25fab5608c3e033d34b4483ec47e68ba109b7) |
| | 2021 | [MLP-Mixer](https://arxiv.org/pdf/2105.01601.pdf) | 使用MLP替换self-attention            |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F2def61f556f9a5576ace08911496b7c7e4f970a4%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/MLP-Mixer%3A-An-all-MLP-Architecture-for-Vision-Tolstikhin-Houlsby/2def61f556f9a5576ace08911496b7c7e4f970a4)  |
| ✅ | 2021 | [MAE](https://arxiv.org/pdf/2111.06377.pdf) | BERT的CV版             |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fc1962a8cf364595ed2838a097e9aa7cd159d3118%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Masked-Autoencoders-Are-Scalable-Vision-Learners-He-Chen/c1962a8cf364595ed2838a097e9aa7cd159d3118)  |

### 生成模型

| 已录制 | 年份 | 名字                                              | 简介         | 引用 |
| ------ | ---- | ------------------------------------------------- | ------------ | ------------------------------------------------------------ |
|  ✅ | 2014 | [GAN](https://papers.nips.cc/paper/2014/file/5ca3e9b122f61f8f06494c97b1afccf3-Paper.pdf) | 生成模型的开创工作                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F54e325aee6b2d476bbbb88615ac15e251c6e8214%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Generative-Adversarial-Nets-Goodfellow-Pouget-Abadie/54e325aee6b2d476bbbb88615ac15e251c6e8214)  |
|  | 2015 | [DCGAN](https://arxiv.org/pdf/1511.06434.pdf) | 使用CNN的GAN          |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F8388f1be26329fa45e5807e968a641ce170ea078%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Unsupervised-Representation-Learning-with-Deep-Radford-Metz/8388f1be26329fa45e5807e968a641ce170ea078)  |
|  | 2016 | [pix2pix](https://arxiv.org/pdf/1611.07004.pdf) |           |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F8acbe90d5b852dadea7810345451a99608ee54c7%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Image-to-Image-Translation-with-Conditional-Isola-Zhu/8acbe90d5b852dadea7810345451a99608ee54c7)  |
|  | 2016 | [SRGAN](https://arxiv.org/pdf/1609.04802.pdf) | 图片超分辨率          |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fdf0c54fe61f0ffb9f0e36a17c2038d9a1964cba3%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Photo-Realistic-Single-Image-Super-Resolution-Using-Ledig-Theis/df0c54fe61f0ffb9f0e36a17c2038d9a1964cba3)  |
|  | 2017 | [WGAN](https://arxiv.org/abs/1701.07875) | 训练更加容易          |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F2f85b7376769473d2bed56f855f115e23d727094%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Wasserstein-GAN-Arjovsky-Chintala/2f85b7376769473d2bed56f855f115e23d727094)  |
|  | 2017 | [CycleGAN](https://arxiv.org/abs/1703.10593) |           |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fc43d954cf8133e6254499f3d68e45218067e4941%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Unpaired-Image-to-Image-Translation-Using-Networks-Zhu-Park/c43d954cf8133e6254499f3d68e45218067e4941)  |
|  | 2018 | [StyleGAN](https://arxiv.org/abs/1812.04948) |           |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fceb2ebef0b41e31c1a21b28c2734123900c005e2%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/A-Style-Based-Generator-Architecture-for-Generative-Karras-Laine/ceb2ebef0b41e31c1a21b28c2734123900c005e2)  |
| | 2019 | [StyleGAN2](https://arxiv.org/pdf/1912.04958.pdf) |        |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Ff3e3d1f86a534a3654d0ee263142e44f4e2c61e9%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Analyzing-and-Improving-the-Image-Quality-of-Karras-Laine/f3e3d1f86a534a3654d0ee263142e44f4e2c61e9)  |
| | 2020 | [DDPM](https://arxiv.org/pdf/2006.11239.pdf) | Diffusion Models   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F289db3be7bf77e06e75541ba93269de3d604ac72%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Denoising-Diffusion-Probabilistic-Models-Ho-Jain/289db3be7bf77e06e75541ba93269de3d604ac72)  |
| | 2021 | [Improved DDPM](https://arxiv.org/pdf/2102.09672.pdf) | 改进的 DDPM   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fde18baa4964804cf471d85a5a090498242d2e79f%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Improved-Denoising-Diffusion-Probabilistic-Models-Nichol-Dhariwal/de18baa4964804cf471d85a5a090498242d2e79f)  |
| | 2021 | [Guided Diffusion Models](https://arxiv.org/pdf/2105.05233.pdf) | 号称超越 GAN  |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F64ea8f180d0682e6c18d1eb688afdb2027c02794%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Diffusion-Models-Beat-GANs-on-Image-Synthesis-Dhariwal-Nichol/64ea8f180d0682e6c18d1eb688afdb2027c02794)  |
| | 2021 | [StyleGAN3](https://arxiv.org/pdf/2106.12423.pdf) |        |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fc1ff08b59f00c44f34dfdde55cd53370733a2c19%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Alias-Free-Generative-Adversarial-Networks-Karras-Aittala/c1ff08b59f00c44f34dfdde55cd53370733a2c19)  |

### 计算机视觉 - Object Detection

| 已录制 | 年份 | 名字                                              | 简介         | 引用 |
| ------ | ---- | ------------------------------------------------- | ------------ | ------------------------------------------------------------ |
|        | 2014 | [R-CNN](https://arxiv.org/pdf/1311.2524v5.pdf)    | Two-stage             |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F2f4df08d9072fc2ac181b7fced6a245315ce05c8%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/2f4df08d9072fc2ac181b7fced6a245315ce05c8)  |
|        | 2015 | [Fast R-CNN](http://arxiv.org/abs/1504.08083v2)   |                       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F7ffdbc358b63378f07311e883dddacc9faeeaf4b%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/7ffdbc358b63378f07311e883dddacc9faeeaf4b)  |
|        | 2015 | [Faster R-CNN](http://arxiv.org/abs/1506.01497v3) |                       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F424561d8585ff8ebce7d5d07de8dbf7aae5e7270%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/424561d8585ff8ebce7d5d07de8dbf7aae5e7270)  |
|        | 2016 | [SSD](http://arxiv.org/abs/1512.02325v5)          | Single stage          |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F4d7a9197433acbfb24ef0e9d0f33ed1699e4a5b0%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/4d7a9197433acbfb24ef0e9d0f33ed1699e4a5b0)  |
|        | 2016 | [YOLO](http://arxiv.org/abs/1506.02640v5)         |                       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Ff8e79ac0ea341056ef20f2616628b3e964764cfd%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/f8e79ac0ea341056ef20f2616628b3e964764cfd)  |
|        | 2017 | [Mask R-CNN](http://arxiv.org/abs/1703.06870v3)   |                       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fea99a5535388196d0d44be5b4d7dd02029a43bb2%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/ea99a5535388196d0d44be5b4d7dd02029a43bb2)  |
|        | 2017 | [YOLOv2](http://arxiv.org/abs/1612.08242v1)       |                       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F7d39d69b23424446f0400ef603b2e3e22d0309d6%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/7d39d69b23424446f0400ef603b2e3e22d0309d6)  |
|        | 2018 | [YOLOv3](http://arxiv.org/abs/1804.02767v1)       |                       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fe4845fb1e624965d4f036d7fd32e8dcdd2408148%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/e4845fb1e624965d4f036d7fd32e8dcdd2408148)  |
|        | 2019 | [CenterNet](https://arxiv.org/pdf/1904.07850.pdf) | Anchor free           |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F6a2e2fd1b5bb11224daef98b3fb6d029f68a73f2%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Objects-as-Points-Zhou-Wang/6a2e2fd1b5bb11224daef98b3fb6d029f68a73f2)  |
|        | 2020 | [DETR](https://arxiv.org/pdf/2005.12872.pdf)      | Transformer           |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F962dc29fdc3fbdc5930a10aba114050b82fe5a3e%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/End-to-End-Object-Detection-with-Transformers-Carion-Massa/962dc29fdc3fbdc5930a10aba114050b82fe5a3e)  |

<a name="contrastive_learning"></a>

### 计算机视觉 - 对比学习


| 已录制 | 年份 | 名字                                                         | 简介                 | 引用 |
| ------ | ---- | ------------------------------------------------------------ | -------------------- | ------------------------------------------------------------ |
| ✅      | 2018 | [InstDisc](https://arxiv.org/pdf/1805.01978.pdf) | 提出实例判别和memory bank做对比学习                  |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F155b7782dbd713982a4133df3aee7adfd0b6b304%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Unsupervised-Feature-Learning-via-Non-parametric-Wu-Xiong/155b7782dbd713982a4133df3aee7adfd0b6b304)  |
| ✅      | 2018 | [CPC](https://arxiv.org/pdf/1807.03748.pdf) | 对比预测编码，图像语音文本强化学习全都能做                   | [![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fb227f3e4c0dc96e5ac5426b85485a70f2175a205%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Representation-Learning-with-Contrastive-Predictive-Oord-Li/b227f3e4c0dc96e5ac5426b85485a70f2175a205) |
| ✅      | 2019 | [InvaSpread](https://arxiv.org/pdf/1904.03436.pdf) | 一个编码器的端到端对比学习                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fe4bde6fe33b6c2cf9d1647ac0b041f7d1ba29c5b%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Unsupervised-Embedding-Learning-via-Invariant-and-Ye-Zhang/e4bde6fe33b6c2cf9d1647ac0b041f7d1ba29c5b)  |
| ✅  | 2019 |  [CMC](https://arxiv.org/pdf/1906.05849.pdf) | 多视角下的对比学习               |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F97f4d09175705be4677d675fa27e55defac44800%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Contrastive-Multiview-Coding-Tian-Krishnan/97f4d09175705be4677d675fa27e55defac44800)  |
| ✅ | 2019 | [MoCov1](https://arxiv.org/pdf/1911.05722.pdf) | 无监督训练效果也很好                   | [![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fec46830a4b275fd01d4de82bffcabe6da086128f%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Momentum-Contrast-for-Unsupervised-Visual-Learning-He-Fan/ec46830a4b275fd01d4de82bffcabe6da086128f) |
|  ✅ | 2020 |  [SimCLRv1](https://arxiv.org/pdf/2002.05709.pdf) |  简单的对比学习 (数据增强 + MLP head + 大batch训练久)                  |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F34733eaf66007516347a40ad5d9bbe1cc9dacb6b%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/A-Simple-Framework-for-Contrastive-Learning-of-Chen-Kornblith/34733eaf66007516347a40ad5d9bbe1cc9dacb6b)  |
|  ✅ | 2020 | [MoCov2](https://arxiv.org/pdf/2003.04297.pdf) | MoCov1 + improvements from SimCLRv1                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fa1b8a8df281bbaec148a897927a49ea47ea31515%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Improved-Baselines-with-Momentum-Contrastive-Chen-Fan/a1b8a8df281bbaec148a897927a49ea47ea31515)  |
|  ✅ | 2020 |  [SimCLRv2](https://arxiv.org/pdf/2006.10029.pdf) | 大的自监督预训练模型很适合做半监督学习                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F3e7f5f4382ac6f9c4fef6197dd21abf74456acd1%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Big-Self-Supervised-Models-are-Strong-Learners-Chen-Kornblith/3e7f5f4382ac6f9c4fef6197dd21abf74456acd1)  |
| ✅  | 2020 |  [BYOL](https://arxiv.org/pdf/2006.07733.pdf) | 不需要负样本的对比学习                   | [![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F38f93092ece8eee9771e61c1edaf11b1293cae1b%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Bootstrap-Your-Own-Latent%3A-A-New-Approach-to-Grill-Strub/38f93092ece8eee9771e61c1edaf11b1293cae1b) |
|  ✅ | 2020 |  [SWaV](https://arxiv.org/pdf/2006.09882.pdf) | 聚类对比学习                   | [![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F10161d83d29fc968c4612c9e9e2b61a2fc25842e%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Unsupervised-Learning-of-Visual-Features-by-Cluster-Caron-Misra/10161d83d29fc968c4612c9e9e2b61a2fc25842e) |
|  ✅ | 2020 |  [SimSiam](https://arxiv.org/pdf/2011.10566.pdf) | 化繁为简的孪生表征学习                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F0e23d2f14e7e56e81538f4a63e11689d8ac1eb9d%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Exploring-Simple-Siamese-Representation-Learning-Chen-He/0e23d2f14e7e56e81538f4a63e11689d8ac1eb9d)  |
| ✅ | 2021 | [MoCov3](https://arxiv.org/pdf/2104.02057.pdf) | 如何更稳定的自监督训练ViT                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F739ceacfafb1c4eaa17509351b647c773270b3ae%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/An-Empirical-Study-of-Training-Self-Supervised-Chen-Xie/739ceacfafb1c4eaa17509351b647c773270b3ae)  |
|  ✅ | 2021 |  [DINO](https://arxiv.org/pdf/2104.14294.pdf) | transformer加自监督在视觉也很香                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fad4a0938c48e61b7827869e4ac3baffd0aefab35%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Emerging-Properties-in-Self-Supervised-Vision-Caron-Touvron/ad4a0938c48e61b7827869e4ac3baffd0aefab35)  |


### 自然语言处理 - Transformer

| 已录制 | 年份 | 名字                                                         | 简介                 | 引用 |
| ------ | ---- | ------------------------------------------------------------ | -------------------- | ------------------------------------------------------------ |
| ✅ | 2017 | [Transformer](https://arxiv.org/abs/1706.03762) | 继MLP、CNN、RNN后的第四大类架构                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F204e3073870fae3d05bcbc2f6a8e263d9b72e776%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Attention-is-All-you-Need-Vaswani-Shazeer/204e3073870fae3d05bcbc2f6a8e263d9b72e776)  |
| ✅ | 2018 | [GPT](https://s3-us-west-2.amazonaws.com/openai-assets/research-covers/language-unsupervised/language_understanding_paper.pdf) | 使用 Transformer 解码器来做预训练               |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fcd18800a0fe0b668a1cc19f2ec95b5003d0a5035%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Improving-Language-Understanding-by-Generative-Radford-Narasimhan/cd18800a0fe0b668a1cc19f2ec95b5003d0a5035)  |
| ✅ | 2018 | [BERT](https://arxiv.org/abs/1810.04805) | Transformer一统NLP的开始                  |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fdf2b0e26d0599ce3e70df8a9da02e51594e0e992%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/BERT%3A-Pre-training-of-Deep-Bidirectional-for-Devlin-Chang/df2b0e26d0599ce3e70df8a9da02e51594e0e992)  |
| ✅ | 2019 | [GPT-2](https://d4mucfpksywv.cloudfront.net/better-language-models/language_models_are_unsupervised_multitask_learners.pdf)  |  更大的 GPT 模型，朝着zero-shot learning迈了一大步             |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F9405cc0d6169988371b2755e573cc28650d14dfe%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Language-Models-are-Unsupervised-Multitask-Learners-Radford-Wu/9405cc0d6169988371b2755e573cc28650d14dfe)  |
| ✅ | 2020 |  [GPT-3](https://arxiv.org/abs/2005.14165) | 100倍更大的 GPT-2，few-shot learning效果显著                  |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F6b85b63579a916f705a8e10a49bd8d849d91b1fc%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Language-Models-are-Few-Shot-Learners-Brown-Mann/6b85b63579a916f705a8e10a49bd8d849d91b1fc)  |

### 系统

| 已录制 | 年份 | 名字                                                         | 简介                 | 引用 |
| ------ | ---- | ------------------------------------------------------------ | -------------------- | ------------------------------------------------------------ |
|  ✅ |  2014 | [参数服务器](https://www.usenix.org/system/files/conference/osdi14/osdi14-paper-li_mu.pdf) | 支持千亿参数的传统机器学习模型       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F0de0c3240bda7972bd0a3c8369ebc4b4f2e4f9c2%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Scaling-Distributed-Machine-Learning-with-the-Li-Andersen/0de0c3240bda7972bd0a3c8369ebc4b4f2e4f9c2)  |
|   | 2018 | [GPipe](https://proceedings.neurips.cc/paper/2019/file/093f65e080a295f8076b1c5722a46aa2-Paper.pdf) | 流水线（Pipeline）并行      |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fc18663fea10c8a303d045fd2c1f33cacf9b73ca3%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/GPipe%3A-Efficient-Training-of-Giant-Neural-Networks-Huang-Cheng/c18663fea10c8a303d045fd2c1f33cacf9b73ca3)  |
|  | 2019 | [Megatron-LM](https://arxiv.org/pdf/1909.08053.pdf) | 张量（Tensor）并行      |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F8323c591e119eb09b28b29fd6c7bc76bd889df7a%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Megatron-LM%3A-Training-Multi-Billion-Parameter-Using-Shoeybi-Patwary/8323c591e119eb09b28b29fd6c7bc76bd889df7a) |
| | 2019 | [Zero](https://arxiv.org/pdf/1910.02054.pdf) | 参数分片      |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F00c957711b12468cb38424caccdf5291bb354033%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/ZeRO%3A-Memory-optimizations-Toward-Training-Trillion-Rajbhandari-Rasley/00c957711b12468cb38424caccdf5291bb354033)  |
| ✅ |  2022 | [Pathways](https://arxiv.org/pdf/2203.12533.pdf) |  将Jax拓展到上千TPU核上       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F512e9aa873a1cd7eb61ce1f25ca7df6acb7e2352%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Pathways%3A-Asynchronous-Distributed-Dataflow-for-ML-Barham-Chowdhery/512e9aa873a1cd7eb61ce1f25ca7df6acb7e2352)  |

### 图神经网络

| 已录制 | 年份 | 名字                                                         | 简介                 | 引用 |
| ------ | ---- | ------------------------------------------------------------ | -------------------- | ------------------------------------------------------------ |
|  ✅ |  2021 | [图神经网络介绍](https://distill.pub/2021/gnn-intro/) | GNN的可视化介绍                  |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F2c0e0440882a42be752268d0b64243243d752a74%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/A-Gentle-Introduction-to-Graph-Neural-Networks-S%C3%A1nchez-Lengeling-Reif/2c0e0440882a42be752268d0b64243243d752a74)  |

### 优化算法

| 已录制 | 年份 | 名字                                                         | 简介                 | 引用 |
| ------ | ---- | ------------------------------------------------------------ | -------------------- | ------------------------------------------------------------ |
| | 2014 | [Adam](https://arxiv.org/abs/1412.6980) | 深度学习里最常用的优化算法之一                   |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fa6cb366736791bcccc5c8639de5a8f9636bf87e8%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Adam%3A-A-Method-for-Stochastic-Optimization-Kingma-Ba/a6cb366736791bcccc5c8639de5a8f9636bf87e8)  |
| | 2016 |  [为什么超大的模型泛化性不错](https://arxiv.org/abs/1611.03530)   |               |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F54ddb00fa691728944fd8becea90a373d21597cf%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Understanding-deep-learning-requires-rethinking-Zhang-Bengio/54ddb00fa691728944fd8becea90a373d21597cf)  |
| | 2017 | [为什么Momentum有效](https://distill.pub/2017/momentum/) | Distill的可视化介绍            |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F3e8ccf9d3d843c9855c5d76ab66d3e775384da72%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Why-Momentum-Really-Works-Goh/3e8ccf9d3d843c9855c5d76ab66d3e775384da72)  |


### 新领域应用


| 已录制 | 年份 | 名字                                                         | 简介                 | 引用 |
| ------ | ---- | ------------------------------------------------------------ | -------------------- | ------------------------------------------------------------ |
| | 2016 | [AlphaGo](https://storage.googleapis.com/deepmind-media/alphago/AlphaGoNaturePaper.pdf) | 强化学习出圈                 |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F846aedd869a00c09b40f1f1f35673cb22bc87490%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Mastering-the-game-of-Go-with-deep-neural-networks-Silver-Huang/846aedd869a00c09b40f1f1f35673cb22bc87490)  |
| | 2020 | [AlphaFold](https://discovery.ucl.ac.uk/id/eprint/10089234/1/343019_3_art_0_py4t4l_convrt.pdf) | 赢得比赛的的蛋白质3D结构预测 |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F3a083d843f891b3574494c385699c21766ce8b7a%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Improved-protein-structure-prediction-using-from-Senior-Evans/3a083d843f891b3574494c385699c21766ce8b7a)  |
| ✅ | 2021 | [AlphaFold 2](https://www.nature.com/articles/s41586-021-03819-2.pdf) | 原子级别精度的蛋白质3D结构预测       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Fdc32a984b651256a8ec282be52310e6bd33d9815%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Highly-accurate-protein-structure-prediction-with-Jumper-Evans/dc32a984b651256a8ec282be52310e6bd33d9815)  |
| ✅ | 2021 | [Codex](https://arxiv.org/pdf/2107.03374.pdf) | 使用注释生成代码       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Facbdbf49f9bc3f151b93d9ca9a06009f4f6eb269%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Evaluating-Large-Language-Models-Trained-on-Code-Chen-Tworek/acbdbf49f9bc3f151b93d9ca9a06009f4f6eb269)  |
| ✅ | 2021 | [指导数学直觉](https://www.nature.com/articles/s41586-021-04086-x.pdf) | 分析不同数学物体之前的联系来帮助发现新定理         |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2Ff672b8fb430606fee0bb368f16603531ce1e90c4%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Advancing-mathematics-by-guiding-human-intuition-AI-Davies-Velickovic/f672b8fb430606fee0bb368f16603531ce1e90c4)  |
| ✅ | 2022 | [AlphaCode](https://storage.googleapis.com/deepmind-media/AlphaCode/competition_level_code_generation_with_alphacode.pdf) | 媲美一般程序员的编程解题水平       |[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F5cbe278b65a81602a864184bbca37de91448a5f5%3Ffields%3DcitationCount)](https://www.semanticscholar.org/paper/Competition-Level-Code-Generation-with-AlphaCode-Li-Choi/5cbe278b65a81602a864184bbca37de91448a5f5)  |