{"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/chatbench-from-static-benchmarks-to-human-ai","title":"ChatBench: From Static Benchmarks to Human-AI Evaluation","arxiv_id":"2504.07114","date":"2025-03-22","proceeding":null,"authors":["Serina Chang","Ashton Anderson","Jake M. Hofman"],"abstract":"With the rapid adoption of LLM-based chatbots, there is a pressing need to evaluate what humans and LLMs can achieve together. However, standard benchmarks, such as MMLU, measure LLM capabilities in isolation (i.e., \"AI-alone\"). Here, we design and conduct a user study to convert MMLU questions into user-AI conversations, by seeding the user with the question and having them carry out a conversation with the LLM to answer their question. We release ChatBench, a new dataset with AI-alone, user-alone, and user-AI data for 396 questions and two LLMs, including 144K answers and 7,336 user-AI conversations. We find that AI-alone accuracy fails to predict user-AI accuracy, with significant differences across multiple subjects (math, physics, and moral reasoning), and we analyze the user-AI conversations to provide insight into how they diverge from AI-alone benchmarks. Finally, we show that fine-tuning a user simulator on a subset of ChatBench improves its ability to estimate user-AI accuracies, increasing correlation on held-out questions by more than 20 points, creating possibilities for scaling interactive evaluation.","url_abs":"https://arxiv.org/abs/2504.07114v1","url_pdf":"https://arxiv.org/pdf/2504.07114v1.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":"chatbench-from-static-benchmarks-to-human-ai","repo_url":"https://github.com/serinachang5/interactive-eval","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"mmlu","task_name":"MMLU"},{"task_slug":"math","task_name":"Math"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2504.07114","atlas_url":"https://app.syntology.ai/?focus=2504.07114","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.07114"}},"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/serinachang5/interactive-eval","reach":null}],"summary":{"ran_draft_wrong":5,"unverified":2},"by_repo_kind":{"official":{"samples":7,"ran":5,"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":7,"samples":[{"code_sha256_prefix":"4c7d56046181f764","entry":"convert_conversation_to_str","repo":"serinachang5/interactive-eval","repo_kind":"official","path":"generate_conversations.py","file_url":"https://github.com/serinachang5/interactive-eval/blob/HEAD/generate_conversations.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":"4c7d56046181f764"}},{"code_sha256_prefix":"2cb702e21e393e8f","entry":"get_answers_from_checkpoints","repo":"serinachang5/interactive-eval","repo_kind":"official","path":"analyze_results.py","file_url":"https://github.com/serinachang5/interactive-eval/blob/HEAD/analyze_results.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":"2cb702e21e393e8f"}},{"code_sha256_prefix":"8689136dc08e3584","entry":"get_checkpoints","repo":"serinachang5/interactive-eval","repo_kind":"official","path":"analyze_results.py","file_url":"https://github.com/serinachang5/interactive-eval/blob/HEAD/analyze_results.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":"8689136dc08e3584"}},{"code_sha256_prefix":"8400472e53a0348d","entry":"get_data_from_final_storage","repo":"serinachang5/interactive-eval","repo_kind":"official","path":"analyze_results.py","file_url":"https://github.com/serinachang5/interactive-eval/blob/HEAD/analyze_results.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":"8400472e53a0348d"}},{"code_sha256_prefix":"2422e0b1229b167a","entry":"letter_only","repo":"serinachang5/interactive-eval","repo_kind":"official","path":"qa_reasoning.py","file_url":"https://github.com/serinachang5/interactive-eval/blob/HEAD/qa_reasoning.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"2422e0b1229b167a"}},{"code_sha256_prefix":"d6b0770a7fd138fb","entry":"copy_and_paste","repo":"serinachang5/interactive-eval","repo_kind":"official","path":"qa_reasoning.py","file_url":"https://github.com/serinachang5/interactive-eval/blob/HEAD/qa_reasoning.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":"d6b0770a7fd138fb"}},{"code_sha256_prefix":"dbd40b5329f4e864","entry":"extract_answer_from_response","repo":"serinachang5/interactive-eval","repo_kind":"official","path":"qa_reasoning.py","file_url":"https://github.com/serinachang5/interactive-eval/blob/HEAD/qa_reasoning.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":"dbd40b5329f4e864"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}