{"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/peering-through-preferences-unraveling","title":"Peering Through Preferences: Unraveling Feedback Acquisition for Aligning Large Language Models","arxiv_id":"2308.15812","date":"2023-08-30","proceeding":null,"authors":["Hritik Bansal","John Dang","Aditya Grover"],"abstract":"Aligning large language models (LLMs) with human values and intents critically involves the use of human or AI feedback. While dense feedback annotations are expensive to acquire and integrate, sparse feedback presents a structural design choice between ratings (e.g., score Response A on a scale of 1-7) and rankings (e.g., is Response A better than Response B?). In this work, we analyze the effect of this design choice for the alignment and evaluation of LLMs. We uncover an inconsistency problem wherein the preferences inferred from ratings and rankings significantly disagree 60% for both human and AI annotators. Our subsequent analysis identifies various facets of annotator biases that explain this phenomena, such as human annotators would rate denser responses higher while preferring accuracy during pairwise judgments. To our surprise, we also observe that the choice of feedback protocol also has a significant effect on the evaluation of aligned LLMs. In particular, we find that LLMs that leverage rankings data for alignment (say model X) are preferred over those that leverage ratings data (say model Y), with a rank-based evaluation protocol (is X/Y's response better than reference response?) but not with a rating-based evaluation protocol (score Rank X/Y's response on a scale of 1-7). Our findings thus shed light on critical gaps in methods for evaluating the real-world utility of language models and their strong dependence on the feedback protocol used for alignment. Our code and data are available at https://github.com/Hritikbansal/sparse_feedback.","url_abs":"https://arxiv.org/abs/2308.15812v3","url_pdf":"https://arxiv.org/pdf/2308.15812v3.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":"peering-through-preferences-unraveling","repo_url":"https://github.com/hritikbansal/sparse_feedback","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":{"syntology_url":"https://syntology.ai/paper/2308.15812","atlas_url":"https://app.syntology.ai/?focus=2308.15812","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2308.15812"}},"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/Hritikbansal/sparse_feedback","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hritikbansal/sparse_feedback","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":3},"by_repo_kind":{"official":{"samples":3,"ran":3,"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":"d15fb7c71a0a8362","entry":"create_instruction","repo":"Hritikbansal/sparse_feedback","repo_kind":"official","path":"utils/convert_data_for_ranking_rm.py","file_url":"https://github.com/Hritikbansal/sparse_feedback/blob/HEAD/utils/convert_data_for_ranking_rm.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d15fb7c71a0a8362"}},{"code_sha256_prefix":"3d46eceabcfa1dfa","entry":"custom_collate","repo":"Hritikbansal/sparse_feedback","repo_kind":"official","path":"src/rankings/utils.py","file_url":"https://github.com/Hritikbansal/sparse_feedback/blob/HEAD/src/rankings/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":"3d46eceabcfa1dfa"}},{"code_sha256_prefix":"828addff0128292d","entry":"custom_collate","repo":"Hritikbansal/sparse_feedback","repo_kind":"official","path":"src/ratings/utils.py","file_url":"https://github.com/Hritikbansal/sparse_feedback/blob/HEAD/src/ratings/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":"828addff0128292d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}