{"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/why-in-context-learning-transformers-are","title":"Fine-tuned In-Context Learning Transformers are Excellent Tabular Data Classifiers","arxiv_id":"2405.13396","date":"2024-05-22","proceeding":null,"authors":["Felix den Breejen","Sangmin Bae","Stephen Cha","Se-Young Yun"],"abstract":"The recently introduced TabPFN pretrains an In-Context Learning (ICL) transformer on synthetic data to perform tabular data classification. In this work, we extend TabPFN to the fine-tuning setting, resulting in a significant performance boost. We also discover that fine-tuning enables ICL-transformers to create complex decision boundaries, a property regular neural networks do not have. Based on this observation, we propose to pretrain ICL-transformers on a new forest dataset generator which creates datasets that are unrealistic, but have complex decision boundaries. TabForest, the ICL-transformer pretrained on this dataset generator, shows better fine-tuning performance when pretrained on more complex datasets. Additionally, TabForest outperforms TabPFN on some real-world datasets when fine-tuning, despite having lower zero-shot performance due to the unrealistic nature of the pretraining datasets. By combining both dataset generators, we create TabForestPFN, an ICL-transformer that achieves excellent fine-tuning performance and good zero-shot performance.","url_abs":"https://arxiv.org/abs/2405.13396v2","url_pdf":"https://arxiv.org/pdf/2405.13396v2.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-in-context-learning-transformers-are","repo_url":"https://github.com/felixdenbreejen/tabforestpfn","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":[{"method_slug":"tabpfn","method_name":"TABPFN"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2405.13396","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2405.13396"}},"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/felixdenbreejen/tabforestpfn","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":4},"by_repo_kind":{"official":{"samples":4,"ran":4,"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":"b9ae30298a39499f","entry":"get_optimizer","repo":"felixdenbreejen/tabforestpfn","repo_kind":"official","path":"tabularbench/core/get_optimizer.py","file_url":"https://github.com/felixdenbreejen/tabforestpfn/blob/HEAD/tabularbench/core/get_optimizer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b9ae30298a39499f"}},{"code_sha256_prefix":"65a13579cf86d425","entry":"get_scheduler","repo":"felixdenbreejen/tabforestpfn","repo_kind":"official","path":"tabularbench/core/get_scheduler.py","file_url":"https://github.com/felixdenbreejen/tabforestpfn/blob/HEAD/tabularbench/core/get_scheduler.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"65a13579cf86d425"}},{"code_sha256_prefix":"6a352f7e9bba806b","entry":"make_standard_dataset_split","repo":"felixdenbreejen/tabforestpfn","repo_kind":"official","path":"tabularbench/core/dataset_split.py","file_url":"https://github.com/felixdenbreejen/tabforestpfn/blob/HEAD/tabularbench/core/dataset_split.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6a352f7e9bba806b"}},{"code_sha256_prefix":"18fbf7cb632b4863","entry":"make_stratified_dataset_split","repo":"felixdenbreejen/tabforestpfn","repo_kind":"official","path":"tabularbench/core/dataset_split.py","file_url":"https://github.com/felixdenbreejen/tabforestpfn/blob/HEAD/tabularbench/core/dataset_split.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"18fbf7cb632b4863"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}