{"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/do-we-truly-need-so-many-samples-multi-llm","title":"Do We Truly Need So Many Samples? Multi-LLM Repeated Sampling Efficiently Scales Test-Time Compute","arxiv_id":"2504.00762","date":"2025-04-01","proceeding":null,"authors":["Jianhao Chen","Zishuo Xun","Bocheng Zhou","Han Qi","Hangfan Zhang","Qiaosheng Zhang","Yang Chen","Wei Hu","Yuzhong Qu","Wanli Ouyang","Shuyue Hu"],"abstract":"This paper presents a simple, effective, and cost-efficient strategy to improve LLM performance by scaling test-time compute. Our strategy builds upon the repeated-sampling-then-voting framework, with a novel twist: incorporating multiple models, even weaker ones, to leverage their complementary strengths that potentially arise from diverse training data and paradigms. By using consistency as a signal, our strategy dynamically switches between models. Theoretical analysis highlights the efficiency and performance advantages of our strategy. Extensive experiments on six datasets demonstrate that our strategy not only outperforms self-consistency and state-of-the-art multi-agent debate approaches, but also significantly reduces inference costs. Additionally, ModelSwitch requires only a few comparable LLMs to achieve optimal performance and can be extended with verification methods, demonstrating the potential of leveraging multiple LLMs in the generation-verification paradigm.","url_abs":"https://arxiv.org/abs/2504.00762v4","url_pdf":"https://arxiv.org/pdf/2504.00762v4.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":"do-we-truly-need-so-many-samples-multi-llm","repo_url":"https://github.com/jianhaochen-nju/modelswitch","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2504.00762","atlas_url":"https://app.syntology.ai/?focus=2504.00762","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.00762"}},"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/jianhaochen-nju/modelswitch","reach":null}],"summary":{"ran_draft_wrong":2,"unverified":2},"by_repo_kind":{"official":{"samples":4,"ran":2,"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":4,"samples":[{"code_sha256_prefix":"20136da384a07b87","entry":"calculate_scores_with_weights","repo":"jianhaochen-nju/modelswitch","repo_kind":"official","path":"src/Evaluation.py","file_url":"https://github.com/jianhaochen-nju/modelswitch/blob/HEAD/src/Evaluation.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"20136da384a07b87"}},{"code_sha256_prefix":"5b10b79877f2037d","entry":"extract_via_string","repo":"jianhaochen-nju/modelswitch","repo_kind":"official","path":"src/Evaluation.py","file_url":"https://github.com/jianhaochen-nju/modelswitch/blob/HEAD/src/Evaluation.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5b10b79877f2037d"}},{"code_sha256_prefix":"6f911101e7e4e5b9","entry":"compute_correctness","repo":"jianhaochen-nju/modelswitch","repo_kind":"official","path":"src/Evaluation.py","file_url":"https://github.com/jianhaochen-nju/modelswitch/blob/HEAD/src/Evaluation.py","link_basis":"first_harvest_node","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":"6f911101e7e4e5b9"}},{"code_sha256_prefix":"1810436c16248ab1","entry":"metric","repo":"jianhaochen-nju/modelswitch","repo_kind":"official","path":"src/Model_switch.py","file_url":"https://github.com/jianhaochen-nju/modelswitch/blob/HEAD/src/Model_switch.py","link_basis":"first_harvest_node","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":"1810436c16248ab1"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}