{"about":{"site":"https://codewithpapers.app","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.","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"},"url":"/paper/competition-dynamics-shape-algorithmic-phases","title":"Competition Dynamics Shape Algorithmic Phases of In-Context Learning","arxiv_id":"2412.01003","date":"2024-12-01","proceeding":null,"authors":["Core Francisco Park","Ekdeep Singh Lubana","Itamar Pres","Hidenori Tanaka"],"abstract":"In-Context Learning (ICL) has significantly expanded the general-purpose nature of large language models, allowing them to adapt to novel tasks using merely the inputted context. This has motivated a series of papers that analyze tractable synthetic domains and postulate precise mechanisms that may underlie ICL. However, the use of relatively distinct setups that often lack a sequence modeling nature to them makes it unclear how general the reported insights from such studies are. Motivated by this, we propose a synthetic sequence modeling task that involves learning to simulate a finite mixture of Markov chains. As we show, models trained on this task reproduce most well-known results on ICL, hence offering a unified setting for studying the concept. Building on this setup, we demonstrate we can explain a model's behavior by decomposing it into four broad algorithms that combine a fuzzy retrieval vs. inference approach with either unigram or bigram statistics of the context. These algorithms engage in a competition dynamics to dominate model behavior, with the precise experimental conditions dictating which algorithm ends up superseding others: e.g., we find merely varying context size or amount of training yields (at times sharp) transitions between which algorithm dictates the model behavior, revealing a mechanism that explains the transient nature of ICL. In this sense, we argue ICL is best thought of as a mixture of different algorithms, each with its own peculiarities, instead of a monolithic capability. This also implies that making general claims about ICL that hold universally across all settings may be infeasible.","url_abs":"https://arxiv.org/abs/2412.01003v4","url_pdf":"https://arxiv.org/pdf/2412.01003v4.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":"competition-dynamics-shape-algorithmic-phases","repo_url":"https://github.com/cfpark00/markov-mixtures","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"in-context-learning","task_name":"In-Context Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2412.01003","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2412.01003"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/cfpark00/markov-mixtures","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1,"unverified":6},"by_repo_kind":{"official":{"samples":7,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"abd7e936a4a2330d","entry":"kl_std_normal","repo":"cfpark00/markov-mixtures","repo_kind":"official","path":"mltools/models/model_tools.py","file_url":"https://github.com/cfpark00/markov-mixtures/blob/HEAD/mltools/models/model_tools.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"abd7e936a4a2330d"}},{"code_sha256_prefix":"d50b1048ff431a29","entry":"get_sqrt_steps","repo":"cfpark00/markov-mixtures","repo_kind":"official","path":"mltools/ml_utils.py","file_url":"https://github.com/cfpark00/markov-mixtures/blob/HEAD/mltools/ml_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d50b1048ff431a29"}},{"code_sha256_prefix":"21cbbf4fd70bbff6","entry":"load_config","repo":"cfpark00/markov-mixtures","repo_kind":"official","path":"utils.py","file_url":"https://github.com/cfpark00/markov-mixtures/blob/HEAD/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"21cbbf4fd70bbff6"}},{"code_sha256_prefix":"705d2e9c15d29c8c","entry":"npattern","repo":"cfpark00/markov-mixtures","repo_kind":"official","path":"utils.py","file_url":"https://github.com/cfpark00/markov-mixtures/blob/HEAD/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"705d2e9c15d29c8c"}},{"code_sha256_prefix":"8925f576f84f1269","entry":"to_np","repo":"cfpark00/markov-mixtures","repo_kind":"official","path":"mltools/ml_utils.py","file_url":"https://github.com/cfpark00/markov-mixtures/blob/HEAD/mltools/ml_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8925f576f84f1269"}},{"code_sha256_prefix":"36d847cba1e96c54","entry":"train","repo":"cfpark00/markov-mixtures","repo_kind":"official","path":"mltools/ml_utils.py","file_url":"https://github.com/cfpark00/markov-mixtures/blob/HEAD/mltools/ml_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"36d847cba1e96c54"}},{"code_sha256_prefix":"a65fb5d84616e942","entry":"xor2","repo":"cfpark00/markov-mixtures","repo_kind":"official","path":"utils.py","file_url":"https://github.com/cfpark00/markov-mixtures/blob/HEAD/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a65fb5d84616e942"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}