{"id":"arxiv:2312.17742","title":"Learning Vision from Models Rivals Learning Vision from Data","date":"2023-12-28","section":"Data, training, optimization","quality":0.170732,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/","source_url":"https://arxiv.org/abs/2312.17742","text_available":true,"authors":["Yonglong Tian","Lijie Fan","Kaifeng Chen","Dina Katabi"],"outgoing":[{"source":"arxiv:2312.17742","target":"arxiv:2002.05202","evidence":"id","id":"arxiv:2002.05202","title":"GLU Variants Improve Transformer","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/e052a2974593d29355ab45c301766ac6f5545ac1bc81d605e78d3adc71e9e2e4/"},{"source":"arxiv:2312.17742","target":"arxiv:2006.09882","evidence":"title","id":"arxiv:2006.09882","title":"Unsupervised Learning of Visual Features by Contrasting Cluster Assignments","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/6601018611a96dff6f5324f2e4f39f62484917cf4d8f8a8d5baa7f4be1deb433/"},{"source":"arxiv:2312.17742","target":"arxiv:2104.14294","evidence":"title","id":"arxiv:2104.14294","title":"Emerging Properties in Self-Supervised Vision Transformers","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/4372186d141ca514b08e93bb3b1e26fdbcba9ce62187fb772c14bf223160c06d/"},{"source":"arxiv:2312.17742","target":"arxiv:2111.07832","evidence":"id","id":"arxiv:2111.07832","title":"iBOT: Image BERT Pre-Training with Online Tokenizer","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/2bd0269c0872fc518db65c2ca98dd30b5d38701f79b0a72fec2ed49648ba589f/"},{"source":"arxiv:2312.17742","target":"arxiv:2304.07193","evidence":"id","id":"arxiv:2304.07193","title":"DINOv2: Learning Robust Visual Features without Supervision","url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/b15933596df35497024556ae88bfa696f3b0e8372b678194f04f1bfb9f3be553/"}],"incoming":[],"text_parts":[{"start":0,"end":15681,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-1.html"},{"start":15681,"end":31371,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-2.html"},{"start":31371,"end":38789,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html"}],"contents":[{"title":"Learning Vision from Models Rivals Learning Vision from Data","offset":0,"level":1,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-1.html#text"},{"title":"Abstract","offset":64,"level":6,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-1.html#text"},{"title":"1 Introduction","offset":997,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-1.html#text"},{"title":"2 Related Works","offset":7094,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-1.html#text"},{"title":"3 Approach","offset":10496,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-1.html#text"},{"title":"3.1 Synthesizing captions","offset":11469,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-1.html#text"},{"title":"3.2 Synthesizing Images","offset":14519,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-1.html#text"},{"title":"3.3 Representation Learning","offset":15191,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-1.html#text"},{"title":"3.4 Implementation","offset":18825,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-2.html#text"},{"title":"4 Experiment","offset":20689,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-2.html#text"},{"title":"4.1 Study different components","offset":20877,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-2.html#text"},{"title":"4.2 Scaling up","offset":25748,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-2.html#text"},{"title":"4.3 Further analysis","offset":29483,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-2.html#text"},{"title":"5 Discussions and Conclusion","offset":32479,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html#text"},{"title":"Appendix A Concept Sampling","offset":34653,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html#text"},{"title":"Appendix B Implementation Details","offset":35220,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html#text"},{"title":"B.1 Pre-training","offset":35258,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html#text"},{"title":"B.2 ImageNet linear probing","offset":35656,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html#text"},{"title":"B.3 End-to-End ImageNet fine-tuning","offset":36293,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html#text"},{"title":"B.4 Semantic segmentation on ADE20k","offset":36557,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html#text"},{"title":"B.5 Fine-grained linear classification","offset":36985,"level":3,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html#text"},{"title":"Appendix C In-context Learning Examples","offset":37665,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html#text"},{"title":"Instructions for reporting errors","offset":37825,"level":2,"url":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/part-3.html#text"}],"text_code_points":38789,"source_markdown":"https://vinnik-dmitry07.github.io/favorite-papers/agents/papers/71dc1683019904602d643d16348a76b98c556b924ca964e3949d8d6420dc4402/source.md"}