{"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/paperbench-evaluating-ai-s-ability-to","title":"PaperBench: Evaluating AI's Ability to Replicate AI Research","arxiv_id":"2504.01848","date":"2025-04-02","proceeding":null,"authors":["Giulio Starace","Oliver Jaffe","Dane Sherburn","James Aung","Jun Shern Chan","Leon Maksin","Rachel Dias","Evan Mays","Benjamin Kinsella","Wyatt Thompson","Johannes Heidecke","Amelia Glaese","Tejal Patwardhan"],"abstract":"We introduce PaperBench, a benchmark evaluating the ability of AI agents to replicate state-of-the-art AI research. Agents must replicate 20 ICML 2024 Spotlight and Oral papers from scratch, including understanding paper contributions, developing a codebase, and successfully executing experiments. For objective evaluation, we develop rubrics that hierarchically decompose each replication task into smaller sub-tasks with clear grading criteria. In total, PaperBench contains 8,316 individually gradable tasks. Rubrics are co-developed with the author(s) of each ICML paper for accuracy and realism. To enable scalable evaluation, we also develop an LLM-based judge to automatically grade replication attempts against rubrics, and assess our judge's performance by creating a separate benchmark for judges. We evaluate several frontier models on PaperBench, finding that the best-performing tested agent, Claude 3.5 Sonnet (New) with open-source scaffolding, achieves an average replication score of 21.0\\%. Finally, we recruit top ML PhDs to attempt a subset of PaperBench, finding that models do not yet outperform the human baseline. We \\href{https://github.com/openai/preparedness}{open-source our code} to facilitate future research in understanding the AI engineering capabilities of AI agents.","url_abs":"https://arxiv.org/abs/2504.01848v1","url_pdf":"https://arxiv.org/pdf/2504.01848v1.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":"paperbench-evaluating-ai-s-ability-to","repo_url":"https://github.com/openai/preparedness","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2504.01848","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.01848"}},"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/openai/preparedness","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":11},"by_repo_kind":{"official":{"samples":11,"ran":0,"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":"474c5c3c6f409c4b","entry":"award_rate_metric","repo":"openai/preparedness","repo_kind":"official","path":"project/evmbench/evmbench/bootstrap.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/evmbench/evmbench/bootstrap.py","link_basis":"harvester_set","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":"474c5c3c6f409c4b"}},{"code_sha256_prefix":"963a5cdd7d66721d","entry":"build_audit_image","repo":"openai/preparedness","repo_kind":"official","path":"project/evmbench/docker_build.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/evmbench/docker_build.py","link_basis":"harvester_set","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":"963a5cdd7d66721d"}},{"code_sha256_prefix":"2e39eb97480c77a0","entry":"build_base_image","repo":"openai/preparedness","repo_kind":"official","path":"project/evmbench/docker_build.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/evmbench/docker_build.py","link_basis":"harvester_set","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":"2e39eb97480c77a0"}},{"code_sha256_prefix":"5e75d45ff41e318c","entry":"find_audit_dirs","repo":"openai/preparedness","repo_kind":"official","path":"project/evmbench/docker_build.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/evmbench/docker_build.py","link_basis":"harvester_set","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":"5e75d45ff41e318c"}},{"code_sha256_prefix":"f64bac3d8b45c56f","entry":"get_model_context_window_length","repo":"openai/preparedness","repo_kind":"official","path":"project/common/preparedness_turn_completer/preparedness_turn_completer/utils.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/common/preparedness_turn_completer/preparedness_turn_completer/utils.py","link_basis":"harvester_set","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":"f64bac3d8b45c56f"}},{"code_sha256_prefix":"b526f0966186e191","entry":"is_content_part","repo":"openai/preparedness","repo_kind":"official","path":"project/common/preparedness_turn_completer/preparedness_turn_completer/type_helpers.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/common/preparedness_turn_completer/preparedness_turn_completer/type_helpers.py","link_basis":"harvester_set","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":"b526f0966186e191"}},{"code_sha256_prefix":"331d019270d5a5f8","entry":"is_image_download_timeout","repo":"openai/preparedness","repo_kind":"official","path":"project/common/preparedness_turn_completer/preparedness_turn_completer/utils.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/common/preparedness_turn_completer/preparedness_turn_completer/utils.py","link_basis":"harvester_set","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":"331d019270d5a5f8"}},{"code_sha256_prefix":"af25bf38f4ce76b9","entry":"is_text_part","repo":"openai/preparedness","repo_kind":"official","path":"project/common/preparedness_turn_completer/preparedness_turn_completer/type_helpers.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/common/preparedness_turn_completer/preparedness_turn_completer/type_helpers.py","link_basis":"harvester_set","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":"af25bf38f4ce76b9"}},{"code_sha256_prefix":"ef76b3b8bb566ab9","entry":"is_text_parts_list","repo":"openai/preparedness","repo_kind":"official","path":"project/common/preparedness_turn_completer/preparedness_turn_completer/type_helpers.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/common/preparedness_turn_completer/preparedness_turn_completer/type_helpers.py","link_basis":"harvester_set","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":"ef76b3b8bb566ab9"}},{"code_sha256_prefix":"b56afe9c46b4b30f","entry":"mean_metric","repo":"openai/preparedness","repo_kind":"official","path":"project/evmbench/evmbench/bootstrap.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/evmbench/evmbench/bootstrap.py","link_basis":"harvester_set","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":"b56afe9c46b4b30f"}},{"code_sha256_prefix":"9c66e4585a1675fd","entry":"success_rate_metric","repo":"openai/preparedness","repo_kind":"official","path":"project/evmbench/evmbench/bootstrap.py","file_url":"https://github.com/openai/preparedness/blob/HEAD/project/evmbench/evmbench/bootstrap.py","link_basis":"harvester_set","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":"9c66e4585a1675fd"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}