{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TUCWOVSSDMMSPICSMI6TVCLOTD","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":"40ba6f2783394bd6fa4f36e91ad60f75f9c20ce1792688c67ce9e3780c86ba38","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-17T10:30:38Z","title_canon_sha256":"80c37f64a51eaba714afe02bb215a8f5a1726f1cee408a25154134636183e61f"},"schema_version":"1.0","source":{"id":"2605.17373","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.17373","created_at":"2026-05-20T00:03:55Z"},{"alias_kind":"arxiv_version","alias_value":"2605.17373v1","created_at":"2026-05-20T00:03:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17373","created_at":"2026-05-20T00:03:55Z"},{"alias_kind":"pith_short_12","alias_value":"TUCWOVSSDMMS","created_at":"2026-05-20T00:03:55Z"},{"alias_kind":"pith_short_16","alias_value":"TUCWOVSSDMMSPICS","created_at":"2026-05-20T00:03:55Z"},{"alias_kind":"pith_short_8","alias_value":"TUCWOVSS","created_at":"2026-05-20T00:03:55Z"}],"graph_snapshots":[{"event_id":"sha256:58e4cc432efcadbf414a2fdf8161b1bf1fd6d1c7de048c6f59a051ee0f871d41","target":"graph","created_at":"2026-05-20T00:03:55Z","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":"2026-05-19T21:41:57.776122Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T21:33:23.713019Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.17373/integrity.json","findings":[],"snapshot_sha256":"e87fbe9044cc55026dda214436c7ef0093c7bdf73ed632e67868cb191d5fa8bf","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"AI research agents accelerate ML research by automating hypothesis generation, experimentation, and empirical refinement. Existing agent strategies range from greedy hill-climbing to tree search and evolutionary optimization, yet which strategy choices drive performance remains unclear. Answering this question requires a benchmark that separates agent strategy (e.g., search topology) from execution infrastructure (e.g., code editor), so that performance differences are attributable to strategy rather than infrastructure, and that provides process-level metrics beyond final scores to analyze ex","authors_text":"Anirudh Goyal, Chang Liu, Dianbo Liu, Hou Hei Lam, Qiran Zou, Samson Yu, Srinivas Anumasa, Tianyi Zhang, Tingting Chen, Wenhao Zhao, Yiming Tang, Yingtao Zhu, Zhengyao Jiang, Zufeng Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-17T10:30:38Z","title":"FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17373","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:dac726f26e1242b81dd095212250708a902b91c389728a85b99552f14f989b81","target":"record","created_at":"2026-05-20T00:03:55Z","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":"40ba6f2783394bd6fa4f36e91ad60f75f9c20ce1792688c67ce9e3780c86ba38","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-17T10:30:38Z","title_canon_sha256":"80c37f64a51eaba714afe02bb215a8f5a1726f1cee408a25154134636183e61f"},"schema_version":"1.0","source":{"id":"2605.17373","kind":"arxiv","version":1}},"canonical_sha256":"9d056756521b1927a052623d3a896e98d620cdd016da8cd71fb68e2d29262d4c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9d056756521b1927a052623d3a896e98d620cdd016da8cd71fb68e2d29262d4c","first_computed_at":"2026-05-20T00:03:55.293034Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:03:55.293034Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"5cxoHK2yErjI3OxkpaUR8N++3C3EnMiAmph+hyUF5TjHm+/JUyjI7fxvqT/DmT5o/ZwTvMNTAcEwcYFyoZglDQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:03:55.293920Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.17373","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dac726f26e1242b81dd095212250708a902b91c389728a85b99552f14f989b81","sha256:58e4cc432efcadbf414a2fdf8161b1bf1fd6d1c7de048c6f59a051ee0f871d41"],"state_sha256":"1f8224915be17a17da1530e36148dfb25a5558296215644f0b766646b88bbeb3"}