{"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/towards-fast-multilingual-llm-inference","title":"Towards Fast Multilingual LLM Inference: Speculative Decoding and Specialized Drafters","arxiv_id":"2406.16758","date":"2024-06-24","proceeding":null,"authors":["Euiin Yi","Taehyeon Kim","Hongseok Jeung","Du-Seong Chang","Se-Young Yun"],"abstract":"Large language models (LLMs) have revolutionized natural language processing and broadened their applicability across diverse commercial applications. However, the deployment of these models is constrained by high inference time in multilingual settings. To mitigate this challenge, this paper explores a training recipe of an assistant model in speculative decoding, which is leveraged to draft and-then its future tokens are verified by the target LLM. We show that language-specific draft models, optimized through a targeted pretrain-and-finetune strategy, substantially brings a speedup in inference time compared to the previous methods. We validate these models across various languages in inference time, out-of-domain speedup, and GPT-4o evaluation.","url_abs":"https://arxiv.org/abs/2406.16758v2","url_pdf":"https://arxiv.org/pdf/2406.16758v2.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":"towards-fast-multilingual-llm-inference","repo_url":"https://github.com/Kthyeon/Multilingual-SpecBench","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2406.16758","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.16758"}},"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/Kthyeon/Multilingual-SpecBench","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":4,"unverified":2},"by_repo_kind":{"official":{"samples":6,"ran":4,"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":"87ed72425ac0f9d8","entry":"baseline_forward","repo":"Kthyeon/Multilingual-SpecBench","repo_kind":"official","path":"evaluation/inference_baseline.py","file_url":"https://github.com/Kthyeon/Multilingual-SpecBench/blob/HEAD/evaluation/inference_baseline.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"87ed72425ac0f9d8"}},{"code_sha256_prefix":"3cd324610fed3cbe","entry":"loading_config","repo":"Kthyeon/Multilingual-SpecBench","repo_kind":"official","path":"data_processing/processing_utils.py","file_url":"https://github.com/Kthyeon/Multilingual-SpecBench/blob/HEAD/data_processing/processing_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3cd324610fed3cbe"}},{"code_sha256_prefix":"57aba3c21a2ebe44","entry":"lookahead_forward","repo":"Kthyeon/Multilingual-SpecBench","repo_kind":"official","path":"evaluation/inference_lookahead.py","file_url":"https://github.com/Kthyeon/Multilingual-SpecBench/blob/HEAD/evaluation/inference_lookahead.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"57aba3c21a2ebe44"}},{"code_sha256_prefix":"ccf07672a5bb126f","entry":"pld_forward","repo":"Kthyeon/Multilingual-SpecBench","repo_kind":"official","path":"evaluation/inference_pld.py","file_url":"https://github.com/Kthyeon/Multilingual-SpecBench/blob/HEAD/evaluation/inference_pld.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ccf07672a5bb126f"}},{"code_sha256_prefix":"e8d521071d16dadf","entry":"load_and_transform_json","repo":"Kthyeon/Multilingual-SpecBench","repo_kind":"official","path":"data_generation/convert_to_train.py","file_url":"https://github.com/Kthyeon/Multilingual-SpecBench/blob/HEAD/data_generation/convert_to_train.py","link_basis":"first_harvest_node","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":"e8d521071d16dadf"}},{"code_sha256_prefix":"7d1be98a44359d93","entry":"loading_dataset","repo":"Kthyeon/Multilingual-SpecBench","repo_kind":"official","path":"data_processing/processing_utils.py","file_url":"https://github.com/Kthyeon/Multilingual-SpecBench/blob/HEAD/data_processing/processing_utils.py","link_basis":"first_harvest_node","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":"7d1be98a44359d93"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}