{"about":{"non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","site":"https://codewithpapers.app","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/why-can-gpt-learn-in-context-language-models","title":"Why Can GPT Learn In-Context? Language Models Implicitly Perform Gradient Descent as Meta-Optimizers","arxiv_id":"2212.10559","date":"2022-12-20","proceeding":null,"authors":["Damai Dai","Yutao Sun","Li Dong","Yaru Hao","Shuming Ma","Zhifang Sui","Furu Wei"],"abstract":"Large pretrained language models have shown surprising in-context learning (ICL) ability. With a few demonstration input-label pairs, they can predict the label for an unseen input without parameter updates. Despite the great success in performance, its working mechanism still remains an open question. In this paper, we explain language models as meta-optimizers and understand in-context learning as implicit finetuning. Theoretically, we figure out that Transformer attention has a dual form of gradient descent. On top of it, we understand ICL as follows: GPT first produces meta-gradients according to the demonstration examples, and then these meta-gradients are applied to the original GPT to build an ICL model. We comprehensively compare the behaviors of in-context learning and explicit finetuning on real tasks to provide empirical evidence that supports our understanding. Experimental results show that in-context learning behaves similarly to explicit finetuning from multiple perspectives. Inspired by the dual form between Transformer attention and gradient descent, we design a momentum-based attention by analogy with gradient descent with momentum. The improved performance over vanilla attention further supports our understanding from another perspective, and more importantly, shows the potential to utilize our understanding for future model design. The code is available at \\url{https://aka.ms/icl}.","url_abs":"https://arxiv.org/abs/2212.10559v3","url_pdf":"https://arxiv.org/pdf/2212.10559v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"why-can-gpt-learn-in-context-language-models","repo_url":"https://github.com/microsoft/lmops","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"jax","reach":null}],"tasks":[{"task_slug":"in-context-learning","task_name":"In-Context Learning"},{"task_slug":"open-question","task_name":"Open-Ended Question Answering"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"discriminative-fine-tuning","method_name":"Discriminative Fine-Tuning"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt","method_name":"GPT"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2212.10559","atlas_url":"https://app.syntology.ai/?focus=2212.10559","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2212.10559"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2212.10559\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2212.10559"},"call":"get_concepts_for_paper(arxiv_id=\"2212.10559\")"}],"by_repo":[{"repo":"microsoft/lmops","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0,"harvested_for_other_papers":5,"harvested_for":["2212.09611","2305.03495","2309.09530","2402.03898","2406.14491"],"harvested_for_calls":[{"arxiv_id":"2212.09611","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2212.09611"},"call":"get_harvested_code_for_paper(arxiv_id=\"2212.09611\")"},{"arxiv_id":"2305.03495","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2305.03495"},"call":"get_harvested_code_for_paper(arxiv_id=\"2305.03495\")"},{"arxiv_id":"2309.09530","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2309.09530"},"call":"get_harvested_code_for_paper(arxiv_id=\"2309.09530\")"},{"arxiv_id":"2402.03898","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2402.03898"},"call":"get_harvested_code_for_paper(arxiv_id=\"2402.03898\")"},{"arxiv_id":"2406.14491","tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2406.14491"},"call":"get_harvested_code_for_paper(arxiv_id=\"2406.14491\")"}]}],"by_repo_harvested_for_is":"harvested_for: the arXiv ids (as Syntology's graph holds them) of every paper whose samples Syntology's graph links to this repository, in arXiv-id order, complete; the page names the first 3 and says how many more. A link only: it does not say the papers are the same work, that one copied the other, or that the code implements this paper's method; harvested_for_calls gives the get_harvested_code_for_paper call for each arXiv-shaped id among them","by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"2cb21f1cde7d5e399053f5c8fe62792a0691ff94252666263923241fc2fc7dfb","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}