{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LHNPWRM6NYBKPCUAV7QIND57XT","short_pith_number":"pith:LHNPWRM6","schema_version":"1.0","canonical_sha256":"59dafb459e6e02a78a80afe0868fbfbce2fa430e64e47514018980d347fdc3e6","source":{"kind":"arxiv","id":"2605.19156","version":1},"attestation_state":"computed","paper":{"title":"How Far Are We From True Auto-Research?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY","cs.LG","cs.MA"],"primary_cat":"cs.AI","authors_text":"Claire Cardie, Ning Wang, Sainyam Galhotra, Zhengxin Zhang","submitted_at":"2026-05-18T22:20:33Z","abstract_excerpt":"Recent auto-research systems can produce complete papers, but feasibility is not the same as quality, and the field still lacks a systematic study of how good agent-generated papers actually are. We introduce ResearchArena, a minimal scaffold that lets off-the-shelf agents (Claude Code using Opus 4.6, Codex using GPT-5.4, and Kimi Code using K2.5) carry out the full research loop themselves (ideation, experimentation, paper writing, self-refinement) under only lightweight guidance. Across 13 computer science seeds and 3 trials per agent-domain pair, ResearchArena yields 117 agent-generated pap"},"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":"2605.19156","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-18T22:20:33Z","cross_cats_sorted":["cs.CY","cs.LG","cs.MA"],"title_canon_sha256":"bc81834a9b491a44dc61fe52520c38928843c823fb9679b6d8d3de744394e1ee","abstract_canon_sha256":"bd84e08b66cba2475d2b5d2d21dfa5fa16eea5fbfdf11ca8034b292737d7fa99"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:30.687713Z","signature_b64":"E4Ybf+dBAzLqPEaNqQvSEZESGD42dwkmNyT/9d/5qF2USbQiKkheD9OkOoHlob+bQexh0dAAhnfU0b7WrFdSCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59dafb459e6e02a78a80afe0868fbfbce2fa430e64e47514018980d347fdc3e6","last_reissued_at":"2026-05-20T01:05:30.686923Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:30.686923Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Far Are We From True Auto-Research?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY","cs.LG","cs.MA"],"primary_cat":"cs.AI","authors_text":"Claire Cardie, Ning Wang, Sainyam Galhotra, Zhengxin Zhang","submitted_at":"2026-05-18T22:20:33Z","abstract_excerpt":"Recent auto-research systems can produce complete papers, but feasibility is not the same as quality, and the field still lacks a systematic study of how good agent-generated papers actually are. We introduce ResearchArena, a minimal scaffold that lets off-the-shelf agents (Claude Code using Opus 4.6, Codex using GPT-5.4, and Kimi Code using K2.5) carry out the full research loop themselves (ideation, experimentation, paper writing, self-refinement) under only lightweight guidance. Across 13 computer science seeds and 3 trials per agent-domain pair, ResearchArena yields 117 agent-generated pap"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19156","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/2605.19156/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.19156","created_at":"2026-05-20T01:05:30.687055+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.19156v1","created_at":"2026-05-20T01:05:30.687055+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19156","created_at":"2026-05-20T01:05:30.687055+00:00"},{"alias_kind":"pith_short_12","alias_value":"LHNPWRM6NYBK","created_at":"2026-05-20T01:05:30.687055+00:00"},{"alias_kind":"pith_short_16","alias_value":"LHNPWRM6NYBKPCUA","created_at":"2026-05-20T01:05:30.687055+00:00"},{"alias_kind":"pith_short_8","alias_value":"LHNPWRM6","created_at":"2026-05-20T01:05:30.687055+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LHNPWRM6NYBKPCUAV7QIND57XT","json":"https://pith.science/pith/LHNPWRM6NYBKPCUAV7QIND57XT.json","graph_json":"https://pith.science/api/pith-number/LHNPWRM6NYBKPCUAV7QIND57XT/graph.json","events_json":"https://pith.science/api/pith-number/LHNPWRM6NYBKPCUAV7QIND57XT/events.json","paper":"https://pith.science/paper/LHNPWRM6"},"agent_actions":{"view_html":"https://pith.science/pith/LHNPWRM6NYBKPCUAV7QIND57XT","download_json":"https://pith.science/pith/LHNPWRM6NYBKPCUAV7QIND57XT.json","view_paper":"https://pith.science/paper/LHNPWRM6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.19156&json=true","fetch_graph":"https://pith.science/api/pith-number/LHNPWRM6NYBKPCUAV7QIND57XT/graph.json","fetch_events":"https://pith.science/api/pith-number/LHNPWRM6NYBKPCUAV7QIND57XT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LHNPWRM6NYBKPCUAV7QIND57XT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LHNPWRM6NYBKPCUAV7QIND57XT/action/storage_attestation","attest_author":"https://pith.science/pith/LHNPWRM6NYBKPCUAV7QIND57XT/action/author_attestation","sign_citation":"https://pith.science/pith/LHNPWRM6NYBKPCUAV7QIND57XT/action/citation_signature","submit_replication":"https://pith.science/pith/LHNPWRM6NYBKPCUAV7QIND57XT/action/replication_record"}},"created_at":"2026-05-20T01:05:30.687055+00:00","updated_at":"2026-05-20T01:05:30.687055+00:00"}