{"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/prediction-powered-ranking-of-large-language","title":"Prediction-Powered Ranking of Large Language Models","arxiv_id":"2402.17826","date":"2024-02-27","proceeding":null,"authors":["Ivi Chatzi","Eleni Straitouri","Suhas Thejaswi","Manuel Gomez Rodriguez"],"abstract":"Large language models are often ranked according to their level of alignment with human preferences -- a model is better than other models if its outputs are more frequently preferred by humans. One of the popular ways to elicit human preferences utilizes pairwise comparisons between the outputs provided by different models to the same inputs. However, since gathering pairwise comparisons by humans is costly and time-consuming, it has become a common practice to gather pairwise comparisons by a strong large language model -- a model strongly aligned with human preferences. Surprisingly, practitioners cannot currently measure the uncertainty that any mismatch between human and model preferences may introduce in the constructed rankings. In this work, we develop a statistical framework to bridge this gap. Given a (small) set of pairwise comparisons by humans and a large set of pairwise comparisons by a model, our framework provides a rank-set -- a set of possible ranking positions -- for each of the models under comparison. Moreover, it guarantees that, with a probability greater than or equal to a user-specified value, the rank-sets cover the true ranking consistent with the distribution of human pairwise preferences asymptotically. Using pairwise comparisons made by humans in the LMSYS Chatbot Arena platform and pairwise comparisons made by three strong large language models, we empirically demonstrate the effectivity of our framework and show that the rank-sets constructed using only pairwise comparisons by the strong large language models are often inconsistent with (the distribution of) human pairwise preferences.","url_abs":"https://arxiv.org/abs/2402.17826v3","url_pdf":"https://arxiv.org/pdf/2402.17826v3.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":"prediction-powered-ranking-of-large-language","repo_url":"https://github.com/networks-learning/prediction-powered-ranking","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"chatbot","task_name":"Chatbot"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"large-language-model","task_name":"Large Language Model"},{"task_slug":"prediction","task_name":"Prediction"}],"methods":[{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2402.17826","atlas_url":"https://app.syntology.ai/?focus=2402.17826","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2402.17826"}},"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":"deterministic:regex_extraction","url":"https://github.com/Networks-Learning/prediction-powered-ranking","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/networks-learning/prediction-powered-ranking","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1,"ran":5,"unverified":2},"by_repo_kind":{"official":{"samples":8,"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":"cdbfedcc67b27ade","entry":"estimate","repo":"networks-learning/prediction-powered-ranking","repo_kind":"official","path":"src/estimate.py","file_url":"https://github.com/networks-learning/prediction-powered-ranking/blob/HEAD/src/estimate.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":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"cdbfedcc67b27ade"}},{"code_sha256_prefix":"261a0d1a62eaba83","entry":"find_ranksets","repo":"Networks-Learning/prediction-powered-ranking","repo_kind":"official","path":"src/ranksets.py","file_url":"https://github.com/Networks-Learning/prediction-powered-ranking/blob/HEAD/src/ranksets.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":"261a0d1a62eaba83"}},{"code_sha256_prefix":"89a90895358d82f5","entry":"generate_data","repo":"Networks-Learning/prediction-powered-ranking","repo_kind":"official","path":"src/synthetic.py","file_url":"https://github.com/Networks-Learning/prediction-powered-ranking/blob/HEAD/src/synthetic.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":"89a90895358d82f5"}},{"code_sha256_prefix":"1e46229d93713b67","entry":"get_common_instances","repo":"Networks-Learning/prediction-powered-ranking","repo_kind":"official","path":"src/data_process.py","file_url":"https://github.com/Networks-Learning/prediction-powered-ranking/blob/HEAD/src/data_process.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":"1e46229d93713b67"}},{"code_sha256_prefix":"c29e47426e6fd79a","entry":"input_json","repo":"Networks-Learning/prediction-powered-ranking","repo_kind":"official","path":"src/data_process.py","file_url":"https://github.com/Networks-Learning/prediction-powered-ranking/blob/HEAD/src/data_process.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":"c29e47426e6fd79a"}},{"code_sha256_prefix":"77d4e94ee72e5889","entry":"model_list","repo":"Networks-Learning/prediction-powered-ranking","repo_kind":"official","path":"src/data_process.py","file_url":"https://github.com/Networks-Learning/prediction-powered-ranking/blob/HEAD/src/data_process.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":"77d4e94ee72e5889"}},{"code_sha256_prefix":"a8295d36104f1f85","entry":"lhat_opt","repo":"Networks-Learning/prediction-powered-ranking","repo_kind":"official","path":"src/estimate.py","file_url":"https://github.com/Networks-Learning/prediction-powered-ranking/blob/HEAD/src/estimate.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":"a8295d36104f1f85"}},{"code_sha256_prefix":"db81315466a224b1","entry":"sample_avg","repo":"Networks-Learning/prediction-powered-ranking","repo_kind":"official","path":"src/estimate.py","file_url":"https://github.com/Networks-Learning/prediction-powered-ranking/blob/HEAD/src/estimate.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":"db81315466a224b1"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}