{"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/erasing-the-bias-fine-tuning-foundation","title":"Erasing the Bias: Fine-Tuning Foundation Models for Semi-Supervised Learning","arxiv_id":"2405.11756","date":"2024-05-20","proceeding":null,"authors":["Kai Gan","Tong Wei"],"abstract":"Semi-supervised learning (SSL) has witnessed remarkable progress, resulting in the emergence of numerous method variations. However, practitioners often encounter challenges when attempting to deploy these methods due to their subpar performance. In this paper, we present a novel SSL approach named FineSSL that significantly addresses this limitation by adapting pre-trained foundation models. We identify the aggregated biases and cognitive deviation problems inherent in foundation models, and propose a simple yet effective solution by imposing balanced margin softmax and decoupled label smoothing. Through extensive experiments, we demonstrate that FineSSL sets a new state of the art for SSL on multiple benchmark datasets, reduces the training cost by over six times, and can seamlessly integrate various fine-tuning and modern SSL algorithms. The source code is available at https://github.com/Gank0078/FineSSL.","url_abs":"https://arxiv.org/abs/2405.11756v1","url_pdf":"https://arxiv.org/pdf/2405.11756v1.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":"erasing-the-bias-fine-tuning-foundation","repo_url":"https://github.com/gank0078/finessl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[],"methods":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2405.11756","atlas_url":"https://app.syntology.ai/?focus=2405.11756","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2405.11756"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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":"deterministic:regex_extraction","url":"https://github.com/Gank0078/FineSSL","reach":{"status":"ok"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/mlfoundations/open_","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gank0078/finessl","reach":{"status":"ok"}}],"summary":{"ran":14,"ran_draft_wrong":5,"unverified":5},"by_repo_kind":{"official":{"samples":24,"ran":19,"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":24,"samples":[{"code_sha256_prefix":"8a507254368b1ac4","entry":"AdaptFormer","repo":"gank0078/finessl","repo_kind":"official","path":"model.py","file_url":"https://github.com/gank0078/finessl/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8a507254368b1ac4"}},{"code_sha256_prefix":"3df5cb5be41b7d33","entry":"Adapter","repo":"gank0078/finessl","repo_kind":"official","path":"model.py","file_url":"https://github.com/gank0078/finessl/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3df5cb5be41b7d33"}},{"code_sha256_prefix":"c6e7c3451c85404f","entry":"AutoContrast","repo":"Gank0078/FineSSL","repo_kind":"official","path":"datasets/randaugment.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/datasets/randaugment.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c6e7c3451c85404f"}},{"code_sha256_prefix":"0d1a8d6ce2015fc7","entry":"Brightness","repo":"Gank0078/FineSSL","repo_kind":"official","path":"datasets/randaugment.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/datasets/randaugment.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0d1a8d6ce2015fc7"}},{"code_sha256_prefix":"32cb3f5760fa593e","entry":"Color","repo":"Gank0078/FineSSL","repo_kind":"official","path":"datasets/randaugment.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/datasets/randaugment.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"32cb3f5760fa593e"}},{"code_sha256_prefix":"43c4c5b948a24c2a","entry":"LoRA","repo":"gank0078/finessl","repo_kind":"official","path":"model.py","file_url":"https://github.com/gank0078/finessl/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"43c4c5b948a24c2a"}},{"code_sha256_prefix":"c4e927462d3c6255","entry":"SSF","repo":"gank0078/finessl","repo_kind":"official","path":"model.py","file_url":"https://github.com/gank0078/finessl/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c4e927462d3c6255"}},{"code_sha256_prefix":"8809b24309edd1b4","entry":"VPT","repo":"gank0078/finessl","repo_kind":"official","path":"model.py","file_url":"https://github.com/gank0078/finessl/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8809b24309edd1b4"}},{"code_sha256_prefix":"dd2afb67421effa9","entry":"ViT_Head","repo":"gank0078/finessl","repo_kind":"official","path":"model.py","file_url":"https://github.com/gank0078/finessl/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dd2afb67421effa9"}},{"code_sha256_prefix":"ec713fa2011796da","entry":"balanced_softmax_loss","repo":"Gank0078/FineSSL","repo_kind":"official","path":"utils/losses.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/utils/losses.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ec713fa2011796da"}},{"code_sha256_prefix":"98f385d847636a3e","entry":"basic_clean","repo":"Gank0078/FineSSL","repo_kind":"official","path":"clip/simple_tokenizer.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/clip/simple_tokenizer.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"98f385d847636a3e"}},{"code_sha256_prefix":"d39d467199ab65dc","entry":"compute_adjustment_by_py","repo":"Gank0078/FineSSL","repo_kind":"official","path":"trainer.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/trainer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d39d467199ab65dc"}},{"code_sha256_prefix":"fb866335bba20e2b","entry":"compute_adjustment_list","repo":"Gank0078/FineSSL","repo_kind":"official","path":"datasets/data.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/datasets/data.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fb866335bba20e2b"}},{"code_sha256_prefix":"4dcf06ba28983266","entry":"focal_loss","repo":"Gank0078/FineSSL","repo_kind":"official","path":"utils/losses.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/utils/losses.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4dcf06ba28983266"}},{"code_sha256_prefix":"d919ae32e5e4e616","entry":"get_pairs","repo":"Gank0078/FineSSL","repo_kind":"official","path":"clip/simple_tokenizer.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/clip/simple_tokenizer.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d919ae32e5e4e616"}},{"code_sha256_prefix":"f6043e7798bcabbd","entry":"interleave","repo":"Gank0078/FineSSL","repo_kind":"official","path":"trainer.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/trainer.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f6043e7798bcabbd"}},{"code_sha256_prefix":"f3db85434e77ebab","entry":"logit_adjusted_loss","repo":"Gank0078/FineSSL","repo_kind":"official","path":"utils/losses.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/utils/losses.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f3db85434e77ebab"}},{"code_sha256_prefix":"e4dd766bcd0d9671","entry":"read_txt_as_list","repo":"Gank0078/FineSSL","repo_kind":"official","path":"datasets/data.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/datasets/data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e4dd766bcd0d9671"}},{"code_sha256_prefix":"9542161e9640b858","entry":"whitespace_clean","repo":"Gank0078/FineSSL","repo_kind":"official","path":"clip/simple_tokenizer.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/clip/simple_tokenizer.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9542161e9640b858"}},{"code_sha256_prefix":"83ed3ade497a3a83","entry":"ViT_Tuner","repo":"gank0078/finessl","repo_kind":"official","path":"model.py","file_url":"https://github.com/gank0078/finessl/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"83ed3ade497a3a83"}},{"code_sha256_prefix":"c47aa9e5b049a11d","entry":"build_model","repo":"Gank0078/FineSSL","repo_kind":"official","path":"clip/model.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/clip/model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c47aa9e5b049a11d"}},{"code_sha256_prefix":"da48cf51668411a0","entry":"compute_accuracy","repo":"Gank0078/FineSSL","repo_kind":"official","path":"utils/evaluator.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/utils/evaluator.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"da48cf51668411a0"}},{"code_sha256_prefix":"86ff4fcbd01d5f8a","entry":"load","repo":"Gank0078/FineSSL","repo_kind":"official","path":"clip/clip.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/clip/clip.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"86ff4fcbd01d5f8a"}},{"code_sha256_prefix":"5479cd4eeee14d78","entry":"transpose","repo":"Gank0078/FineSSL","repo_kind":"official","path":"datasets/data.py","file_url":"https://github.com/Gank0078/FineSSL/blob/HEAD/datasets/data.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5479cd4eeee14d78"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}