{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:PYGVHWY3S7OVTGY56PZPZOZP4Y","short_pith_number":"pith:PYGVHWY3","canonical_record":{"source":{"id":"2605.16616","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-15T20:35:32Z","cross_cats_sorted":[],"title_canon_sha256":"612be3bc31291556324c024c4c6ac5452c3dce269cc79881d377deb0bd733dec","abstract_canon_sha256":"11c9552c63d3a2d77f98537e85d15da516848cb5c6374b25c24fc00a79cd1e0b"},"schema_version":"1.0"},"canonical_sha256":"7e0d53db1b97dd599b1df3f2fcbb2fe60405ded792577748006c319425c2676b","source":{"kind":"arxiv","id":"2605.16616","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16616","created_at":"2026-05-20T00:02:32Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16616v1","created_at":"2026-05-20T00:02:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16616","created_at":"2026-05-20T00:02:32Z"},{"alias_kind":"pith_short_12","alias_value":"PYGVHWY3S7OV","created_at":"2026-05-20T00:02:32Z"},{"alias_kind":"pith_short_16","alias_value":"PYGVHWY3S7OVTGY5","created_at":"2026-05-20T00:02:32Z"},{"alias_kind":"pith_short_8","alias_value":"PYGVHWY3","created_at":"2026-05-20T00:02:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:PYGVHWY3S7OVTGY56PZPZOZP4Y","target":"record","payload":{"canonical_record":{"source":{"id":"2605.16616","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-15T20:35:32Z","cross_cats_sorted":[],"title_canon_sha256":"612be3bc31291556324c024c4c6ac5452c3dce269cc79881d377deb0bd733dec","abstract_canon_sha256":"11c9552c63d3a2d77f98537e85d15da516848cb5c6374b25c24fc00a79cd1e0b"},"schema_version":"1.0"},"canonical_sha256":"7e0d53db1b97dd599b1df3f2fcbb2fe60405ded792577748006c319425c2676b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:32.621735Z","signature_b64":"0pErT6ceXvgje3dSaJiXKG8TaguZmUFnnKLXgNo80fJltXNNyygc1aXhTmlxwwFlEYyfbckpITGrGu+ll3n6BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e0d53db1b97dd599b1df3f2fcbb2fe60405ded792577748006c319425c2676b","last_reissued_at":"2026-05-20T00:02:32.620922Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:32.620922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.16616","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:02:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bRQiegD4I+HXmn1d7n/X+rzoi7w4vV4ld5O+AONRzkMUeudPdhMN8UBQrnxWDtpfJDYvHe4QADxONphCQ0jTCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T06:16:35.753429Z"},"content_sha256":"e326bc29c6eea3e2972245065cd184d56f270b0d876aca9d3e5b3c267febf2d0","schema_version":"1.0","event_id":"sha256:e326bc29c6eea3e2972245065cd184d56f270b0d876aca9d3e5b3c267febf2d0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:PYGVHWY3S7OVTGY56PZPZOZP4Y","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MLReplicate: Benchmarking Autonomous Research Systems for Machine Learning Reproducibility","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Diyana Muhammed, Farhana Keya, Gollam Rabby, Sasi Kiran Gaddipati, S\\\"oren Auer","submitted_at":"2026-05-15T20:35:32Z","abstract_excerpt":"Autonomous research systems capable of generating complete scientific manuscripts have advanced rapidly, yet robust and realistic evaluation frameworks have failed to keep pace. To bridge this gap, we introduce MLReplicate, an end-to-end benchmark evaluating autonomous research systems on machine learning reproducibility. The benchmark was constructed from ICML 2025 outstanding papers reformulated into standardized input specifications and evaluated across 6 state-of-the-art research systems: AI SCIENTIST-V1, AI SCIENTIST-V2, AGENT LABORATORY, CYCLERESEARCHER, AI RESEARCHER, and TINY SCIENTIST"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.16616","kind":"arxiv","version":1},"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/2605.16616/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-19T19:21:56.780155Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.590805Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"67a0e49f81711bdfa644a87e14c4b8e77110527420f633a4aae03effe91542d3"},"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":"2026-05-20T00:02:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"96g4+p4WpNhHX/IqfnZyTJihVY8mv4wt9lm/T273uYxwoI3Xf3zUTiGBwNH4asZMTTIln9y/98IY/U1kYCpGAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T06:16:35.753951Z"},"content_sha256":"3d0f6e906672fddf673325f34b226f3244dd00b8233e043d398f918660901a4a","schema_version":"1.0","event_id":"sha256:3d0f6e906672fddf673325f34b226f3244dd00b8233e043d398f918660901a4a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PYGVHWY3S7OVTGY56PZPZOZP4Y/bundle.json","state_url":"https://pith.science/pith/PYGVHWY3S7OVTGY56PZPZOZP4Y/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PYGVHWY3S7OVTGY56PZPZOZP4Y/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":"2026-05-26T06:16:35Z","links":{"resolver":"https://pith.science/pith/PYGVHWY3S7OVTGY56PZPZOZP4Y","bundle":"https://pith.science/pith/PYGVHWY3S7OVTGY56PZPZOZP4Y/bundle.json","state":"https://pith.science/pith/PYGVHWY3S7OVTGY56PZPZOZP4Y/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PYGVHWY3S7OVTGY56PZPZOZP4Y/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:PYGVHWY3S7OVTGY56PZPZOZP4Y","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":"11c9552c63d3a2d77f98537e85d15da516848cb5c6374b25c24fc00a79cd1e0b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-15T20:35:32Z","title_canon_sha256":"612be3bc31291556324c024c4c6ac5452c3dce269cc79881d377deb0bd733dec"},"schema_version":"1.0","source":{"id":"2605.16616","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16616","created_at":"2026-05-20T00:02:32Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16616v1","created_at":"2026-05-20T00:02:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16616","created_at":"2026-05-20T00:02:32Z"},{"alias_kind":"pith_short_12","alias_value":"PYGVHWY3S7OV","created_at":"2026-05-20T00:02:32Z"},{"alias_kind":"pith_short_16","alias_value":"PYGVHWY3S7OVTGY5","created_at":"2026-05-20T00:02:32Z"},{"alias_kind":"pith_short_8","alias_value":"PYGVHWY3","created_at":"2026-05-20T00:02:32Z"}],"graph_snapshots":[{"event_id":"sha256:3d0f6e906672fddf673325f34b226f3244dd00b8233e043d398f918660901a4a","target":"graph","created_at":"2026-05-20T00:02:32Z","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":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T19:21:56.780155Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.590805Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.16616/integrity.json","findings":[],"snapshot_sha256":"67a0e49f81711bdfa644a87e14c4b8e77110527420f633a4aae03effe91542d3","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Autonomous research systems capable of generating complete scientific manuscripts have advanced rapidly, yet robust and realistic evaluation frameworks have failed to keep pace. To bridge this gap, we introduce MLReplicate, an end-to-end benchmark evaluating autonomous research systems on machine learning reproducibility. The benchmark was constructed from ICML 2025 outstanding papers reformulated into standardized input specifications and evaluated across 6 state-of-the-art research systems: AI SCIENTIST-V1, AI SCIENTIST-V2, AGENT LABORATORY, CYCLERESEARCHER, AI RESEARCHER, and TINY SCIENTIST","authors_text":"Diyana Muhammed, Farhana Keya, Gollam Rabby, Sasi Kiran Gaddipati, S\\\"oren Auer","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-15T20:35:32Z","title":"MLReplicate: Benchmarking Autonomous Research Systems for Machine Learning Reproducibility"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.16616","kind":"arxiv","version":1},"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:e326bc29c6eea3e2972245065cd184d56f270b0d876aca9d3e5b3c267febf2d0","target":"record","created_at":"2026-05-20T00:02:32Z","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":"11c9552c63d3a2d77f98537e85d15da516848cb5c6374b25c24fc00a79cd1e0b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-15T20:35:32Z","title_canon_sha256":"612be3bc31291556324c024c4c6ac5452c3dce269cc79881d377deb0bd733dec"},"schema_version":"1.0","source":{"id":"2605.16616","kind":"arxiv","version":1}},"canonical_sha256":"7e0d53db1b97dd599b1df3f2fcbb2fe60405ded792577748006c319425c2676b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7e0d53db1b97dd599b1df3f2fcbb2fe60405ded792577748006c319425c2676b","first_computed_at":"2026-05-20T00:02:32.620922Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:02:32.620922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0pErT6ceXvgje3dSaJiXKG8TaguZmUFnnKLXgNo80fJltXNNyygc1aXhTmlxwwFlEYyfbckpITGrGu+ll3n6BQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:02:32.621735Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.16616","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e326bc29c6eea3e2972245065cd184d56f270b0d876aca9d3e5b3c267febf2d0","sha256:3d0f6e906672fddf673325f34b226f3244dd00b8233e043d398f918660901a4a"],"state_sha256":"71b70e520eeb3e6ffd75b45aa0193111ae8ca41c8e0ffaf0dfb585a35eba97f3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Hf40iFMb20mNtlbz3Wmn4Y2gge0vdNfMeHht6D7eAWw6lRRizSJPMO02dPLNaHzHtN94RgCn531j23uLa+UeBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T06:16:35.757089Z","bundle_sha256":"e49a50af0b9c5ac6e77149e78a65bcc3c97e8f5d87e0c26727d034a8c0b17ee1"}}