{"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/delift-data-efficient-language-model","title":"DELIFT: Data Efficient Language model Instruction Fine Tuning","arxiv_id":"2411.04425","date":"2024-11-07","proceeding":null,"authors":["Ishika Agarwal","KrishnaTeja Killamsetty","Lucian Popa","Marina Danilevksy"],"abstract":"Fine-tuning large language models (LLMs) is essential for enhancing their performance on specific tasks but is often resource-intensive due to redundant or uninformative data. To address this inefficiency, we introduce DELIFT (Data Efficient Language model Instruction Fine-Tuning), a novel algorithm that systematically optimizes data selection across the three key stages of fine-tuning: (1) instruction tuning, (2) task-specific fine-tuning (e.g., reasoning, question-answering), and (3) continual fine-tuning (e.g., incorporating new data versions). Unlike existing methods that focus on single-stage optimization or rely on computationally intensive gradient calculations, DELIFT operates efficiently across all stages. Central to our approach is a pairwise utility metric that quantifies how beneficial a data sample is for improving the model's responses to other samples, effectively measuring the informational value relative to the model's current capabilities. By leveraging different submodular functions applied to this metric, DELIFT selects diverse and optimal subsets that are useful across all stages of fine-tuning. Experiments across various tasks and model scales demonstrate that DELIFT can reduce the fine-tuning data size by up to 70% without compromising performance, offering significant computational savings and outperforming existing methods in both efficiency and efficacy.","url_abs":"https://arxiv.org/abs/2411.04425v2","url_pdf":"https://arxiv.org/pdf/2411.04425v2.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":"delift-data-efficient-language-model","repo_url":"https://github.com/agarwalishika/delift","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"model","task_name":"model"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2411.04425","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2411.04425"}},"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/agarwalishika/delift","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":7},"by_repo_kind":{"official":{"samples":7,"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":"9466a06224c8446a","entry":"batch_inference","repo":"agarwalishika/delift","repo_kind":"official","path":"subset_selection/model_inference.py","file_url":"https://github.com/agarwalishika/delift/blob/HEAD/subset_selection/model_inference.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":"9466a06224c8446a"}},{"code_sha256_prefix":"40f1ecbbd3fde8d6","entry":"batch_inference_text","repo":"agarwalishika/delift","repo_kind":"official","path":"subset_selection/model_inference.py","file_url":"https://github.com/agarwalishika/delift/blob/HEAD/subset_selection/model_inference.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":"40f1ecbbd3fde8d6"}},{"code_sha256_prefix":"f4b1584d2d507091","entry":"fit_tsne","repo":"agarwalishika/delift","repo_kind":"official","path":"visualization/create_embeddings.py","file_url":"https://github.com/agarwalishika/delift/blob/HEAD/visualization/create_embeddings.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":"f4b1584d2d507091"}},{"code_sha256_prefix":"70f5e6ddf38ce6d3","entry":"get_matrix","repo":"agarwalishika/delift","repo_kind":"official","path":"visualization/create_embeddings.py","file_url":"https://github.com/agarwalishika/delift/blob/HEAD/visualization/create_embeddings.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":"70f5e6ddf38ce6d3"}},{"code_sha256_prefix":"846141342e3ff8ae","entry":"get_prompts_refs","repo":"agarwalishika/delift","repo_kind":"official","path":"visualization/data_object.py","file_url":"https://github.com/agarwalishika/delift/blob/HEAD/visualization/data_object.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":"846141342e3ff8ae"}},{"code_sha256_prefix":"1f96e190107daf8e","entry":"load_matrix","repo":"agarwalishika/delift","repo_kind":"official","path":"visualization/create_embeddings.py","file_url":"https://github.com/agarwalishika/delift/blob/HEAD/visualization/create_embeddings.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":"1f96e190107daf8e"}},{"code_sha256_prefix":"b495e87e1ef43502","entry":"parse_file_name","repo":"agarwalishika/delift","repo_kind":"official","path":"visualization/folder_names.py","file_url":"https://github.com/agarwalishika/delift/blob/HEAD/visualization/folder_names.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":"b495e87e1ef43502"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}