{"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/fedllm-bench-realistic-benchmarks-for","title":"FedLLM-Bench: Realistic Benchmarks for Federated Learning of Large Language Models","arxiv_id":"2406.04845","date":"2024-06-07","proceeding":null,"authors":["Rui Ye","Rui Ge","Xinyu Zhu","Jingyi Chai","Yaxin Du","Yang Liu","Yanfeng Wang","Siheng Chen"],"abstract":"Federated learning has enabled multiple parties to collaboratively train large language models without directly sharing their data (FedLLM). Following this training paradigm, the community has put massive efforts from diverse aspects including framework, performance, and privacy. However, an unpleasant fact is that there are currently no realistic datasets and benchmarks for FedLLM and previous works all rely on artificially constructed datasets, failing to capture properties in real-world scenarios. Addressing this, we propose FedLLM-Bench, which involves 8 training methods, 4 training datasets, and 6 evaluation metrics, to offer a comprehensive testbed for the FedLLM community. FedLLM-Bench encompasses three datasets (e.g., user-annotated multilingual dataset) for federated instruction tuning and one dataset (e.g., user-annotated preference dataset) for federated preference alignment, whose scale of client number ranges from 38 to 747. Our datasets incorporate several representative diversities: language, quality, quantity, instruction, length, embedding, and preference, capturing properties in real-world scenarios. Based on FedLLM-Bench, we conduct experiments on all datasets to benchmark existing FL methods and provide empirical insights (e.g., multilingual collaboration). We believe that our FedLLM-Bench can benefit the FedLLM community by reducing required efforts, providing a practical testbed, and promoting fair comparisons. Code and datasets are available at https://github.com/rui-ye/FedLLM-Bench.","url_abs":"https://arxiv.org/abs/2406.04845v1","url_pdf":"https://arxiv.org/pdf/2406.04845v1.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":"fedllm-bench-realistic-benchmarks-for","repo_url":"https://github.com/rui-ye/fedllm-bench","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"fedllm-bench-realistic-benchmarks-for","repo_url":"https://github.com/brains-group/openfedllm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"fedllm-bench-realistic-benchmarks-for","repo_url":"https://github.com/rui-ye/openfedllm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"federated-learning","task_name":"Federated Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2406.04845","atlas_url":"https://app.syntology.ai/?focus=2406.04845","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.04845"}},"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/rui-ye/fedllm-bench","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/rui-ye/openfedllm","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/brains-group/openfedllm","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":16,"unverified":3},"by_repo_kind":{"official":{"samples":11,"ran":9,"repositories":1},"listed":{"samples":8,"ran":7,"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":11,"samples":[{"code_sha256_prefix":"80dde39da1ba3708","entry":"compress","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"compression.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/compression.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":"80dde39da1ba3708"}},{"code_sha256_prefix":"89b2eb0c997d6cbb","entry":"gaussian_noise","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"federated_learning/fed_global.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/federated_learning/fed_global.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"89b2eb0c997d6cbb"}},{"code_sha256_prefix":"962adc31294b40bf","entry":"get_auxiliary_dict","repo":"rui-ye/openfedllm","repo_kind":"listed","path":"federated_learning/fed_utils.py","file_url":"https://github.com/rui-ye/openfedllm/blob/HEAD/federated_learning/fed_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"962adc31294b40bf"}},{"code_sha256_prefix":"2128b33c10829870","entry":"get_clients_this_round","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"federated_learning/fed_global.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/federated_learning/fed_global.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"2128b33c10829870"}},{"code_sha256_prefix":"c5824da57107fe8c","entry":"get_compressed_list","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"compression.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/compression.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":"c5824da57107fe8c"}},{"code_sha256_prefix":"9c102ddaa014f4d2","entry":"get_dataset_this_round","repo":"rui-ye/openfedllm","repo_kind":"listed","path":"federated_learning/split_dataset.py","file_url":"https://github.com/rui-ye/openfedllm/blob/HEAD/federated_learning/split_dataset.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9c102ddaa014f4d2"}},{"code_sha256_prefix":"93e0de8955732c63","entry":"get_model_config","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"config.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/config.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"93e0de8955732c63"}},{"code_sha256_prefix":"79b93353bd3c904e","entry":"get_model_config","repo":"rui-ye/openfedllm","repo_kind":"listed","path":"config.py","file_url":"https://github.com/rui-ye/openfedllm/blob/HEAD/config.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"79b93353bd3c904e"}},{"code_sha256_prefix":"38486baf4c608ec4","entry":"get_proxy_dict","repo":"rui-ye/openfedllm","repo_kind":"listed","path":"federated_learning/fed_utils.py","file_url":"https://github.com/rui-ye/openfedllm/blob/HEAD/federated_learning/fed_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"38486baf4c608ec4"}},{"code_sha256_prefix":"f632fd8875ef18c9","entry":"get_retry_time","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"evaluation/open_ended/gen_judge_bench.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/evaluation/open_ended/gen_judge_bench.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"f632fd8875ef18c9"}},{"code_sha256_prefix":"9250c783f987def8","entry":"get_training_args","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"config.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/config.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"9250c783f987def8"}},{"code_sha256_prefix":"83d9e1534d3a0be4","entry":"get_training_args","repo":"rui-ye/openfedllm","repo_kind":"listed","path":"config.py","file_url":"https://github.com/rui-ye/openfedllm/blob/HEAD/config.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"83d9e1534d3a0be4"}},{"code_sha256_prefix":"f4ab6660068cc1b9","entry":"global_aggregate","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"federated_learning/fed_global.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/federated_learning/fed_global.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"f4ab6660068cc1b9"}},{"code_sha256_prefix":"8fef5b18a4b5d83b","entry":"global_aggregate","repo":"rui-ye/openfedllm","repo_kind":"listed","path":"federated_learning/fed_global.py","file_url":"https://github.com/rui-ye/openfedllm/blob/HEAD/federated_learning/fed_global.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8fef5b18a4b5d83b"}},{"code_sha256_prefix":"5297d9d21f45c15a","entry":"remove_parent_directory_name","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"data_module.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/data_module.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5297d9d21f45c15a"}},{"code_sha256_prefix":"60c6c78eb9667d17","entry":"split_dataset","repo":"rui-ye/openfedllm","repo_kind":"listed","path":"federated_learning/split_dataset.py","file_url":"https://github.com/rui-ye/openfedllm/blob/HEAD/federated_learning/split_dataset.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"60c6c78eb9667d17"}},{"code_sha256_prefix":"174d0858d8cf33ca","entry":"get_conv_template","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"conversation.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/conversation.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":"174d0858d8cf33ca"}},{"code_sha256_prefix":"506ee6d4bab01412","entry":"get_conv_template","repo":"rui-ye/openfedllm","repo_kind":"listed","path":"utils/conversation.py","file_url":"https://github.com/rui-ye/openfedllm/blob/HEAD/utils/conversation.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":"506ee6d4bab01412"}},{"code_sha256_prefix":"5196732e9a84b6dc","entry":"load_compress_model","repo":"rui-ye/fedllm-bench","repo_kind":"official","path":"compression.py","file_url":"https://github.com/rui-ye/fedllm-bench/blob/HEAD/compression.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"5196732e9a84b6dc"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}