{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:F56VSANHBNNTVA4ZABD32KKJG5","short_pith_number":"pith:F56VSANH","canonical_record":{"source":{"id":"2505.24785","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T16:46:29Z","cross_cats_sorted":[],"title_canon_sha256":"d275dbb160cb6a4ae01dfb54a9827a6ff5db99e2f8d7686b451536cc424b75e1","abstract_canon_sha256":"a624b365bfbe4d88270c3cfb7744070cde7be8f1b9324d5b9997f9607d89b461"},"schema_version":"1.0"},"canonical_sha256":"2f7d5901a70b5b3a83990047bd2949376737be190a3f30fc73f470faded8f29e","source":{"kind":"arxiv","id":"2505.24785","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.24785","created_at":"2026-07-05T11:14:09Z"},{"alias_kind":"arxiv_version","alias_value":"2505.24785v2","created_at":"2026-07-05T11:14:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24785","created_at":"2026-07-05T11:14:09Z"},{"alias_kind":"pith_short_12","alias_value":"F56VSANHBNNT","created_at":"2026-07-05T11:14:09Z"},{"alias_kind":"pith_short_16","alias_value":"F56VSANHBNNTVA4Z","created_at":"2026-07-05T11:14:09Z"},{"alias_kind":"pith_short_8","alias_value":"F56VSANH","created_at":"2026-07-05T11:14:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:F56VSANHBNNTVA4ZABD32KKJG5","target":"record","payload":{"canonical_record":{"source":{"id":"2505.24785","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T16:46:29Z","cross_cats_sorted":[],"title_canon_sha256":"d275dbb160cb6a4ae01dfb54a9827a6ff5db99e2f8d7686b451536cc424b75e1","abstract_canon_sha256":"a624b365bfbe4d88270c3cfb7744070cde7be8f1b9324d5b9997f9607d89b461"},"schema_version":"1.0"},"canonical_sha256":"2f7d5901a70b5b3a83990047bd2949376737be190a3f30fc73f470faded8f29e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:09.349803Z","signature_b64":"PW/2pCdni4LaZjTK369ZtIPYx5FobBXcMzFZU+sL9PTS0bdVRdP8/rd8REwPUjMzCDaakzjgAtCN/CIfxJrfCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f7d5901a70b5b3a83990047bd2949376737be190a3f30fc73f470faded8f29e","last_reissued_at":"2026-07-05T11:14:09.349284Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:09.349284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.24785","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:14:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"85PSFYOWWCu6gA83a7hJlVZyIpvYerMRgW2mG9nB3QNgO/nXWGNwE650X6NH03JvWAaGi+0AcKA06mLRx5EVAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T05:27:17.050650Z"},"content_sha256":"7f79c8957fe3c021983ddce442029f94db28901c25106cde5ff8c448c7a82b27","schema_version":"1.0","event_id":"sha256:7f79c8957fe3c021983ddce442029f94db28901c25106cde5ff8c448c7a82b27"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:F56VSANHBNNTVA4ZABD32KKJG5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"EXP-Bench: Can AI Conduct AI Research Experiments?","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Ang Chen, Jayanth Srinivasa, Jiachen Liu, Jiarong Xing, Jingjia Peng, Matei Zaharia, Mosharaf Chowdhury, Myungjin Lee, Patrick Tser Jern Kon, Qiuyi Ding, Xinyi Zhu, Yibo Huang, Yiming Qiu","submitted_at":"2025-05-30T16:46:29Z","abstract_excerpt":"Automating AI research holds immense potential for accelerating scientific progress, yet current AI agents struggle with the complexities of rigorous, end-to-end experimentation. We introduce EXP-Bench, a novel benchmark designed to systematically evaluate AI agents on complete research experiments sourced from influential AI publications. Given a research question and incomplete starter code, EXP-Bench challenges AI agents to formulate hypotheses, design and implement experimental procedures, execute them, and analyze results. To enable the creation of such intricate and authentic tasks with "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24785","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.24785/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:14:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Eahz0XRERkPh+81ss1v7jRxWarpHAcEZc1pmkG1ryO2dwQP9+/Po3oWE+8aQ8ZhszEOUxWbKTTfzcBnwwewqCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T05:27:17.051143Z"},"content_sha256":"d3e40507225fbcebf61a51818ab6e805a12e081b0bee7cec24dbf6ebe93232f7","schema_version":"1.0","event_id":"sha256:d3e40507225fbcebf61a51818ab6e805a12e081b0bee7cec24dbf6ebe93232f7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/F56VSANHBNNTVA4ZABD32KKJG5/bundle.json","state_url":"https://pith.science/pith/F56VSANHBNNTVA4ZABD32KKJG5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/F56VSANHBNNTVA4ZABD32KKJG5/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-15T05:27:17Z","links":{"resolver":"https://pith.science/pith/F56VSANHBNNTVA4ZABD32KKJG5","bundle":"https://pith.science/pith/F56VSANHBNNTVA4ZABD32KKJG5/bundle.json","state":"https://pith.science/pith/F56VSANHBNNTVA4ZABD32KKJG5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/F56VSANHBNNTVA4ZABD32KKJG5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:F56VSANHBNNTVA4ZABD32KKJG5","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"a624b365bfbe4d88270c3cfb7744070cde7be8f1b9324d5b9997f9607d89b461","cross_cats_sorted":[],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T16:46:29Z","title_canon_sha256":"d275dbb160cb6a4ae01dfb54a9827a6ff5db99e2f8d7686b451536cc424b75e1"},"schema_version":"1.0","source":{"id":"2505.24785","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.24785","created_at":"2026-07-05T11:14:09Z"},{"alias_kind":"arxiv_version","alias_value":"2505.24785v2","created_at":"2026-07-05T11:14:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24785","created_at":"2026-07-05T11:14:09Z"},{"alias_kind":"pith_short_12","alias_value":"F56VSANHBNNT","created_at":"2026-07-05T11:14:09Z"},{"alias_kind":"pith_short_16","alias_value":"F56VSANHBNNTVA4Z","created_at":"2026-07-05T11:14:09Z"},{"alias_kind":"pith_short_8","alias_value":"F56VSANH","created_at":"2026-07-05T11:14:09Z"}],"graph_snapshots":[{"event_id":"sha256:d3e40507225fbcebf61a51818ab6e805a12e081b0bee7cec24dbf6ebe93232f7","target":"graph","created_at":"2026-07-05T11:14:09Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2505.24785/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Automating AI research holds immense potential for accelerating scientific progress, yet current AI agents struggle with the complexities of rigorous, end-to-end experimentation. We introduce EXP-Bench, a novel benchmark designed to systematically evaluate AI agents on complete research experiments sourced from influential AI publications. Given a research question and incomplete starter code, EXP-Bench challenges AI agents to formulate hypotheses, design and implement experimental procedures, execute them, and analyze results. To enable the creation of such intricate and authentic tasks with ","authors_text":"Ang Chen, Jayanth Srinivasa, Jiachen Liu, Jiarong Xing, Jingjia Peng, Matei Zaharia, Mosharaf Chowdhury, Myungjin Lee, Patrick Tser Jern Kon, Qiuyi Ding, Xinyi Zhu, Yibo Huang, Yiming Qiu","cross_cats":[],"headline":"","license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T16:46:29Z","title":"EXP-Bench: Can AI Conduct AI Research Experiments?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24785","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:7f79c8957fe3c021983ddce442029f94db28901c25106cde5ff8c448c7a82b27","target":"record","created_at":"2026-07-05T11:14:09Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"a624b365bfbe4d88270c3cfb7744070cde7be8f1b9324d5b9997f9607d89b461","cross_cats_sorted":[],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T16:46:29Z","title_canon_sha256":"d275dbb160cb6a4ae01dfb54a9827a6ff5db99e2f8d7686b451536cc424b75e1"},"schema_version":"1.0","source":{"id":"2505.24785","kind":"arxiv","version":2}},"canonical_sha256":"2f7d5901a70b5b3a83990047bd2949376737be190a3f30fc73f470faded8f29e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2f7d5901a70b5b3a83990047bd2949376737be190a3f30fc73f470faded8f29e","first_computed_at":"2026-07-05T11:14:09.349284Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:09.349284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PW/2pCdni4LaZjTK369ZtIPYx5FobBXcMzFZU+sL9PTS0bdVRdP8/rd8REwPUjMzCDaakzjgAtCN/CIfxJrfCg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:09.349803Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.24785","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7f79c8957fe3c021983ddce442029f94db28901c25106cde5ff8c448c7a82b27","sha256:d3e40507225fbcebf61a51818ab6e805a12e081b0bee7cec24dbf6ebe93232f7"],"state_sha256":"739f7e2413318d9713cf228579a2c30ea319606b07b0f55a8ded75ffed0be8f3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yXVyTwYus+nFZ1OqVEz6HR0YwcYpYE5NLD6M4LPADsLwfazKe0kDoEWnaNr/jhRSaBrFyVYbOIL6HiIzO9ttDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T05:27:17.055605Z","bundle_sha256":"96f05167b327e04ad1bab7a4975781f868f07b8f635a7792d82affc8f8db7f21"}}