{"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/bridging-pre-trained-models-and-downstream","title":"Bridging Pre-trained Models and Downstream Tasks for Source Code Understanding","arxiv_id":"2112.02268","date":"2021-12-04","proceeding":null,"authors":["Deze Wang","Zhouyang Jia","Shanshan Li","Yue Yu","Yun Xiong","Wei Dong","Xiangke Liao"],"abstract":"With the great success of pre-trained models, the pretrain-then-finetune paradigm has been widely adopted on downstream tasks for source code understanding. However, compared to costly training a large-scale model from scratch, how to effectively adapt pre-trained models to a new task has not been fully explored. In this paper, we propose an approach to bridge pre-trained models and code-related tasks. We exploit semantic-preserving transformation to enrich downstream data diversity, and help pre-trained models learn semantic features invariant to these semantically equivalent transformations. Further, we introduce curriculum learning to organize the transformed data in an easy-to-hard manner to fine-tune existing pre-trained models. We apply our approach to a range of pre-trained models, and they significantly outperform the state-of-the-art models on tasks for source code understanding, such as algorithm classification, code clone detection, and code search. Our experiments even show that without heavy pre-training on code data, natural language pre-trained model RoBERTa fine-tuned with our lightweight approach could outperform or rival existing code pre-trained models fine-tuned on the above tasks, such as CodeBERT and GraphCodeBERT. This finding suggests that there is still much room for improvement in code pre-trained models.","url_abs":"https://arxiv.org/abs/2112.02268v2","url_pdf":"https://arxiv.org/pdf/2112.02268v2.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":"bridging-pre-trained-models-and-downstream","repo_url":"https://github.com/wangdeze18/DACL","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"clone-detection","task_name":"Clone Detection"},{"task_slug":"code-search","task_name":"Code Search"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bert","method_name":"BERT"},{"method_slug":"codebert","method_name":"CodeBERT"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-linear-decay","method_name":"Linear Warmup With Linear Decay"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"roberta","method_name":"RoBERTa"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2112.02268","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2112.02268"}},"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/wangdeze18/DACL","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":13},"by_repo_kind":{"official":{"samples":13,"ran":0,"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":"7d233b5a51ce6818","entry":"builddataset","repo":"wangdeze18/DACL","repo_kind":"official","path":"codesearch/convert_score.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/codesearch/convert_score.py","link_basis":"harvester_set","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":"7d233b5a51ce6818"}},{"code_sha256_prefix":"242cf4b7922b3210","entry":"calScores","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/code/bagging.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/code/bagging.py","link_basis":"harvester_set","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":"242cf4b7922b3210"}},{"code_sha256_prefix":"790aecc4340767b7","entry":"calculate_scores","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/evaluator/eva_MAP.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/evaluator/eva_MAP.py","link_basis":"harvester_set","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":"790aecc4340767b7"}},{"code_sha256_prefix":"83477337636fe464","entry":"calculate_scores","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/evaluator/evaluator.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/evaluator/evaluator.py","link_basis":"harvester_set","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":"83477337636fe464"}},{"code_sha256_prefix":"77b01d53fe1ab5d8","entry":"calculate_scores","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/evaluator/evaluator_score.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/evaluator/evaluator_score.py","link_basis":"harvester_set","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":"77b01d53fe1ab5d8"}},{"code_sha256_prefix":"9d6328379f9b5a8c","entry":"convert_examples_to_features","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/code/bagging.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/code/bagging.py","link_basis":"harvester_set","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":"9d6328379f9b5a8c"}},{"code_sha256_prefix":"4cfead0734041166","entry":"evaluate","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/code/bagging.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/code/bagging.py","link_basis":"harvester_set","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":"4cfead0734041166"}},{"code_sha256_prefix":"bedbea7ed3a8588a","entry":"extract_answers","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/evaluator/extract_answers.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/evaluator/extract_answers.py","link_basis":"harvester_set","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":"bedbea7ed3a8588a"}},{"code_sha256_prefix":"4f23a89db8afd4e3","entry":"linear","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/code/pacing_functions.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/code/pacing_functions.py","link_basis":"harvester_set","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":"4f23a89db8afd4e3"}},{"code_sha256_prefix":"f869934d6404c4bc","entry":"read_answers","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/evaluator/eva_MAP.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/evaluator/eva_MAP.py","link_basis":"harvester_set","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":"f869934d6404c4bc"}},{"code_sha256_prefix":"a82333e3033af699","entry":"read_predictions","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/evaluator/eva_MAP.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/evaluator/eva_MAP.py","link_basis":"harvester_set","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":"a82333e3033af699"}},{"code_sha256_prefix":"94cb9437e94901a0","entry":"root_2","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/code/pacing_functions.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/code/pacing_functions.py","link_basis":"harvester_set","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":"94cb9437e94901a0"}},{"code_sha256_prefix":"3055a8267da0a254","entry":"root_5","repo":"wangdeze18/DACL","repo_kind":"official","path":"algorithm_classification/code/pacing_functions.py","file_url":"https://github.com/wangdeze18/DACL/blob/HEAD/algorithm_classification/code/pacing_functions.py","link_basis":"harvester_set","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":"3055a8267da0a254"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}