{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QF4DHV3V3LOT25DQVDF2647ZE7","short_pith_number":"pith:QF4DHV3V","schema_version":"1.0","canonical_sha256":"817833d775dadd3d7470a8cbaf73f927c0b85a28e7779cb388e5a4f85ca9466a","source":{"kind":"arxiv","id":"2410.07095","version":6},"attestation_state":"computed","paper":{"title":"MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aleksander M\\k{a}dry, Dane Sherburn, Evan Mays, Giulio Starace, James Aung, Jun Shern Chan, Kevin Liu, Leon Maksin, Lilian Weng, Neil Chowdhury, Oliver Jaffe, Tejal Patwardhan","submitted_at":"2024-10-09T17:34:27Z","abstract_excerpt":"We introduce MLE-bench, a benchmark for measuring how well AI agents perform at machine learning engineering. To this end, we curate 75 ML engineering-related competitions from Kaggle, creating a diverse set of challenging tasks that test real-world ML engineering skills such as training models, preparing datasets, and running experiments. We establish human baselines for each competition using Kaggle's publicly available leaderboards. We use open-source agent scaffolds to evaluate several frontier language models on our benchmark, finding that the best-performing setup--OpenAI's o1-preview wi"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2410.07095","kind":"arxiv","version":6},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-09T17:34:27Z","cross_cats_sorted":[],"title_canon_sha256":"4808a882ceec70a9b006e387a5a4ae7304b8cf998c4b6ce47d411243cfcd32ac","abstract_canon_sha256":"78f46afeabf2db76f56a21a7180c73b2d734f47ae536a6871ce359a462ad3ca3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:22.256596Z","signature_b64":"1znAovspyhiSPEOL/R96dfp4n6YPxit0qA2YoWDxjzkQQS7MBxrCWC/RQhywW3XpH89jdwY+klps/GpnhxkDDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"817833d775dadd3d7470a8cbaf73f927c0b85a28e7779cb388e5a4f85ca9466a","last_reissued_at":"2026-05-17T23:39:22.256025Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:22.256025Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aleksander M\\k{a}dry, Dane Sherburn, Evan Mays, Giulio Starace, James Aung, Jun Shern Chan, Kevin Liu, Leon Maksin, Lilian Weng, Neil Chowdhury, Oliver Jaffe, Tejal Patwardhan","submitted_at":"2024-10-09T17:34:27Z","abstract_excerpt":"We introduce MLE-bench, a benchmark for measuring how well AI agents perform at machine learning engineering. To this end, we curate 75 ML engineering-related competitions from Kaggle, creating a diverse set of challenging tasks that test real-world ML engineering skills such as training models, preparing datasets, and running experiments. We establish human baselines for each competition using Kaggle's publicly available leaderboards. We use open-source agent scaffolds to evaluate several frontier language models on our benchmark, finding that the best-performing setup--OpenAI's o1-preview wi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.07095","kind":"arxiv","version":6},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2410.07095","created_at":"2026-05-17T23:39:22.256100+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.07095v6","created_at":"2026-05-17T23:39:22.256100+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.07095","created_at":"2026-05-17T23:39:22.256100+00:00"},{"alias_kind":"pith_short_12","alias_value":"QF4DHV3V3LOT","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"QF4DHV3V3LOT25DQ","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"QF4DHV3V","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":56,"internal_anchor_count":35,"sample":[{"citing_arxiv_id":"2605.17637","citing_title":"WebGameBench: Requirement-to-Application Evaluation for Coding Agents via Browser-Native Games","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2503.21460","citing_title":"Large Language Model Agent: A Survey on Methodology, Applications and Challenges","ref_index":144,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22154","citing_title":"IdleSpec: Exploiting Idle Time via Speculative Planning for LLM Agents","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20306","citing_title":"WildRoadBench: A Wild Aerial Road-Damage Grounding Benchmark for Vision-Language Models and Autonomous Agents","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20690","citing_title":"Declarative Data Services: Structured Agentic Discovery for Composing Data Systems","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16616","citing_title":"MLReplicate: Benchmarking Autonomous Research Systems for Machine Learning Reproducibility","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17373","citing_title":"FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17439","citing_title":"DiagEval: Trajectory-Conditioned Diagnosis for Reliable Software Evaluation with GUI Agents","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17637","citing_title":"WebGameBench: Requirement-to-Application Evaluation for Coding Agents via Browser-Native Games","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19099","citing_title":"DecisionBench: A Benchmark for Emergent Delegation in Long-Horizon Agentic Workflows","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19156","citing_title":"How Far Are We From True Auto-Research?","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19447","citing_title":"What and When to Distill: Selective Hindsight Distillation for Multi-Turn Agents","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20025","citing_title":"AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaboration","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20086","citing_title":"What Do Evolutionary Coding Agents Evolve?","ref_index":59,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17439","citing_title":"DiagEval: Trajectory-Conditioned Diagnosis for Reliable Software Evaluation with GUI Agents","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15766","citing_title":"BioXArena: Benchmarking LLM Agents on Multi-Modal Biomedical Machine Learning Tasks","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15308","citing_title":"SMCEvolve: Principled Scientific Discovery via Sequential Monte Carlo Evolution","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2506.02387","citing_title":"VS-Bench: Evaluating VLMs for Strategic Abilities in Multi-Agent Environments","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2506.22598","citing_title":"RExBench: Can coding agents autonomously implement AI research extensions?","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2508.10177","citing_title":"KompeteAI: Accelerated Autonomous Multi-Agent System for End-to-End Pipeline Generation for Machine Learning Problems","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2509.06806","citing_title":"MachineLearningLM: Scaling Many-shot In-context Learning via Continued Pretraining","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2510.17795","citing_title":"What Makes AI Research Replicable? Executable Knowledge Graphs as Scientific Knowledge Representations","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2509.08827","citing_title":"A Survey of Reinforcement Learning for Large Reasoning Models","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2512.09629","citing_title":"End-to-end PDDL Planning with Hardcoded and Dynamic Agents","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2412.04984","citing_title":"Frontier Models are Capable of In-context Scheming","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QF4DHV3V3LOT25DQVDF2647ZE7","json":"https://pith.science/pith/QF4DHV3V3LOT25DQVDF2647ZE7.json","graph_json":"https://pith.science/api/pith-number/QF4DHV3V3LOT25DQVDF2647ZE7/graph.json","events_json":"https://pith.science/api/pith-number/QF4DHV3V3LOT25DQVDF2647ZE7/events.json","paper":"https://pith.science/paper/QF4DHV3V"},"agent_actions":{"view_html":"https://pith.science/pith/QF4DHV3V3LOT25DQVDF2647ZE7","download_json":"https://pith.science/pith/QF4DHV3V3LOT25DQVDF2647ZE7.json","view_paper":"https://pith.science/paper/QF4DHV3V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.07095&json=true","fetch_graph":"https://pith.science/api/pith-number/QF4DHV3V3LOT25DQVDF2647ZE7/graph.json","fetch_events":"https://pith.science/api/pith-number/QF4DHV3V3LOT25DQVDF2647ZE7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QF4DHV3V3LOT25DQVDF2647ZE7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QF4DHV3V3LOT25DQVDF2647ZE7/action/storage_attestation","attest_author":"https://pith.science/pith/QF4DHV3V3LOT25DQVDF2647ZE7/action/author_attestation","sign_citation":"https://pith.science/pith/QF4DHV3V3LOT25DQVDF2647ZE7/action/citation_signature","submit_replication":"https://pith.science/pith/QF4DHV3V3LOT25DQVDF2647ZE7/action/replication_record"}},"created_at":"2026-05-17T23:39:22.256100+00:00","updated_at":"2026-05-17T23:39:22.256100+00:00"}