{"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/unleashing-the-power-of-task-specific","title":"Task-Specific Directions: Definition, Exploration, and Utilization in Parameter Efficient Fine-Tuning","arxiv_id":"2409.01035","date":"2024-09-02","proceeding":null,"authors":["Chongjie Si","Zhiyi Shi","Shifan Zhang","Xiaokang Yang","Hanspeter Pfister","Wei Shen"],"abstract":"Large language models demonstrate impressive performance on downstream tasks, yet they require extensive resource consumption when fully fine-tuning all parameters. To mitigate this, Parameter Efficient Fine-Tuning (PEFT) strategies, such as LoRA, have been developed. In this paper, we delve into the concept of task-specific directions (TSDs), which are critical for transitioning large models from pretrained states to task-specific enhancements in PEFT. We propose a framework to clearly define these directions and explore their properties and practical utilization challenges. We then introduce a novel approach, LoRA-Dash, which aims to maximize the impact of TSDs during the fine-tuning process, thereby enhancing model performance on targeted tasks. Additionally, based on our exploration of TSD, we focus on an important issue in PEFT: the initialization of LoRA. While some works have pointed out the significance of initialization for LoRA's performance and proposed various strategies, these methods are often empirical and not task-specific. To address this issue, we propose LoRA-Init. Starting from TSD, we identify the directions that require the most adjustment during fine-tuning for downstream tasks. By initializing the matrices in LoRA with these directions, LoRA-Init significantly enhances LoRA's performance. Moreover, we can combine LoRA-Dash and LoRA-Init to create the final version of LoRA based on TSDs, which we refer to as LoRA-TSD. Extensive experiments have conclusively demonstrated the effectiveness of these methods, and in-depth analyses further reveal the underlying mechanisms behind their success.","url_abs":"https://arxiv.org/abs/2409.01035v4","url_pdf":"https://arxiv.org/pdf/2409.01035v4.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":"unleashing-the-power-of-task-specific","repo_url":"https://github.com/Chongjie-Si/Subspace-Tuning","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"jax","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"parameter-efficient-fine-tuning","task_name":"parameter-efficient fine-tuning"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2409.01035","atlas_url":"https://app.syntology.ai/?focus=2409.01035","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2409.01035"}},"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":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Chongjie-Si/Subspace-Tuning","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":1,"ran_draft_wrong":5,"unverified":4},"by_repo_kind":{"official":{"samples":10,"ran":6,"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":"2dc17f5e6fb39b3f","entry":"evaluate","repo":"Chongjie-Si/Subspace-Tuning","repo_kind":"official","path":"CR_MR/multi_dataset_eval.py","file_url":"https://github.com/Chongjie-Si/Subspace-Tuning/blob/HEAD/CR_MR/multi_dataset_eval.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2dc17f5e6fb39b3f"}},{"code_sha256_prefix":"56e674752db00649","entry":"generate_prompt","repo":"Chongjie-Si/Subspace-Tuning","repo_kind":"official","path":"CR_MR/finetune.py","file_url":"https://github.com/Chongjie-Si/Subspace-Tuning/blob/HEAD/CR_MR/finetune.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":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"56e674752db00649"}},{"code_sha256_prefix":"0dbc7fc6c518d0fb","entry":"generate_prompt","repo":"Chongjie-Si/Subspace-Tuning","repo_kind":"official","path":"CR_MR/generate.py","file_url":"https://github.com/Chongjie-Si/Subspace-Tuning/blob/HEAD/CR_MR/generate.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":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0dbc7fc6c518d0fb"}},{"code_sha256_prefix":"1c6b8be890b6118a","entry":"generate_prompt","repo":"Chongjie-Si/Subspace-Tuning","repo_kind":"official","path":"CR_MR/commonsense_evaluate.py","file_url":"https://github.com/Chongjie-Si/Subspace-Tuning/blob/HEAD/CR_MR/commonsense_evaluate.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":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1c6b8be890b6118a"}},{"code_sha256_prefix":"56a9466ff35b35bb","entry":"load_data","repo":"Chongjie-Si/Subspace-Tuning","repo_kind":"official","path":"CR_MR/commonsense_evaluate.py","file_url":"https://github.com/Chongjie-Si/Subspace-Tuning/blob/HEAD/CR_MR/commonsense_evaluate.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":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"56a9466ff35b35bb"}},{"code_sha256_prefix":"a4f5d0414a1cf9c6","entry":"translate_state_dict_key","repo":"Chongjie-Si/Subspace-Tuning","repo_kind":"official","path":"CR_MR/export_state_dict_checkpoint.py","file_url":"https://github.com/Chongjie-Si/Subspace-Tuning/blob/HEAD/CR_MR/export_state_dict_checkpoint.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":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a4f5d0414a1cf9c6"}},{"code_sha256_prefix":"415f46d8a341e28d","entry":"main","repo":"Chongjie-Si/Subspace-Tuning","repo_kind":"official","path":"CR_MR/commonsense_evaluate.py","file_url":"https://github.com/Chongjie-Si/Subspace-Tuning/blob/HEAD/CR_MR/commonsense_evaluate.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"415f46d8a341e28d"}},{"code_sha256_prefix":"4740c3e235eedfb4","entry":"main","repo":"Chongjie-Si/Subspace-Tuning","repo_kind":"official","path":"CR_MR/math_evaluate.py","file_url":"https://github.com/Chongjie-Si/Subspace-Tuning/blob/HEAD/CR_MR/math_evaluate.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4740c3e235eedfb4"}},{"code_sha256_prefix":"a2f49e13f6e13d6b","entry":"permute","repo":"Chongjie-Si/Subspace-Tuning","repo_kind":"official","path":"CR_MR/export_state_dict_checkpoint.py","file_url":"https://github.com/Chongjie-Si/Subspace-Tuning/blob/HEAD/CR_MR/export_state_dict_checkpoint.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a2f49e13f6e13d6b"}},{"code_sha256_prefix":"6e275f232bbfb9b6","entry":"unpermute","repo":"Chongjie-Si/Subspace-Tuning","repo_kind":"official","path":"CR_MR/export_state_dict_checkpoint.py","file_url":"https://github.com/Chongjie-Si/Subspace-Tuning/blob/HEAD/CR_MR/export_state_dict_checkpoint.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"6e275f232bbfb9b6"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}