{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:5NQBBU63AUJIV2EHECEPAMITCF","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":"83ec676de5bea3f20fd00e45fcb2895dc4ef81d6815eac992c6312f267bc1eb7","cross_cats_sorted":["cs.AI","cs.SE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MA","submitted_at":"2026-05-21T11:29:08Z","title_canon_sha256":"2dfd36b989134e7d61d89f819bd7b98966ae8486daeb9e0328058d40b3d315be"},"schema_version":"1.0","source":{"id":"2605.22343","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.22343","created_at":"2026-05-22T01:04:38Z"},{"alias_kind":"arxiv_version","alias_value":"2605.22343v1","created_at":"2026-05-22T01:04:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22343","created_at":"2026-05-22T01:04:38Z"},{"alias_kind":"pith_short_12","alias_value":"5NQBBU63AUJI","created_at":"2026-05-22T01:04:38Z"},{"alias_kind":"pith_short_16","alias_value":"5NQBBU63AUJIV2EH","created_at":"2026-05-22T01:04:38Z"},{"alias_kind":"pith_short_8","alias_value":"5NQBBU63","created_at":"2026-05-22T01:04:38Z"}],"graph_snapshots":[{"event_id":"sha256:b7a418df5ea8f5b8e9b4ada260a82568dd7832fb32fb52c33cb55bf79a544881","target":"graph","created_at":"2026-05-22T01:04:38Z","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/2605.22343/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Autonomous research systems increasingly make the scientific workflow executable: agents can propose ideas, run code, inspect results, and draft papers. But executable workflows do not by themselves produce research judgment. We analyze where current systems lose trial experience: weak evidence becomes prose, pilot signals become broad claims, memory remains textual, and recurring process failures do not change later behavior. We introduce Sibyl-AutoResearch, a self-evolving AutoResearch framework built around Scientific Trial-and-Error Harnesses. A harness lets agents run bounded trials, pres","authors_text":"Chang Xu, Chengcheng Wang, Jianyuan Guo, Qinhua Xie, Shiqi Wang, Wei He","cross_cats":["cs.AI","cs.SE"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MA","submitted_at":"2026-05-21T11:29:08Z","title":"Sibyl-AutoResearch: Autonomous Research Needs Self-Evolving Trial-and-Error Harnesses, Not Paper Generators"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22343","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":"sha256:9ae3281d0a274c2bfe63ad47098f889b960aae5f81e91f46036c31bd54dc9d5b","target":"record","created_at":"2026-05-22T01:04:38Z","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":"83ec676de5bea3f20fd00e45fcb2895dc4ef81d6815eac992c6312f267bc1eb7","cross_cats_sorted":["cs.AI","cs.SE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MA","submitted_at":"2026-05-21T11:29:08Z","title_canon_sha256":"2dfd36b989134e7d61d89f819bd7b98966ae8486daeb9e0328058d40b3d315be"},"schema_version":"1.0","source":{"id":"2605.22343","kind":"arxiv","version":1}},"canonical_sha256":"eb6010d3db05128ae8872088f031131169ad64fa6e15897f9a749f0305651137","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"eb6010d3db05128ae8872088f031131169ad64fa6e15897f9a749f0305651137","first_computed_at":"2026-05-22T01:04:38.755644Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:04:38.755644Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ln8MlHLhnmNKp1evikpz/9OmmUr+IHLR6/89gKaoyY8gQZRDI6e18OIVrkKUepiRQvJcux+zZ2EYgh97T0iWAA==","signature_status":"signed_v1","signed_at":"2026-05-22T01:04:38.756271Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.22343","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9ae3281d0a274c2bfe63ad47098f889b960aae5f81e91f46036c31bd54dc9d5b","sha256:b7a418df5ea8f5b8e9b4ada260a82568dd7832fb32fb52c33cb55bf79a544881"],"state_sha256":"3b57f685d8f2b4e51770c146231b8dcd70703e072f3ab526f59ac94141311508"}