{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2SCLS3WTQWEJQTO7YO6DBWSJYE","short_pith_number":"pith:2SCLS3WT","schema_version":"1.0","canonical_sha256":"d484b96ed38588984ddfc3bc30da49c13ac909f7a01af5512715b775d77f2a45","source":{"kind":"arxiv","id":"2607.20926","version":1},"attestation_state":"computed","paper":{"title":"SciExplore: Evaluating Autonomous Agents from Scientific Navigation to Information Integration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bin Liu, Kai Chen, Kuikun Liu, Weiming Zhang, Wenran Liu, Wenwei Zhang, Yanan Sun, Yinhao Tang, Youqing Fang","submitted_at":"2026-07-23T05:17:00Z","abstract_excerpt":"Scientific research involves complex information-seeking and reasoning workflows across heterogeneous sources. However, existing benchmarks primarily emphasize general-domain retrieval or static scientific question answering, and therefore fail to assess key capabilities required in realistic scientific research workflows. We introduce SciExplore, a benchmark designed to evaluate scientific information-seeking and reasoning capabilities of LLMs and agents. SciExplore comprises four task types covering 103 expert-curated tasks across more than ten scientific disciplines: scientific database nav"},"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":"2607.20926","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-23T05:17:00Z","cross_cats_sorted":[],"title_canon_sha256":"17754d6a6de1756405fe71947adff1b1d64b6f29cc1cd03934425375ef116d1a","abstract_canon_sha256":"de6212a21d99d1d57fe85d2c8fc1c805364885f1e78327b2984a6cd3e6ff1bfd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T01:23:41.163975Z","signature_b64":"tzFuDLouXoXnpPjtw46hB3hUJdIBpCJ71YwhV6vTCDENccCKRMwEYSf7f1aRbNUIfax97W0Fyc7qZ9nP9pJzBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d484b96ed38588984ddfc3bc30da49c13ac909f7a01af5512715b775d77f2a45","last_reissued_at":"2026-07-24T01:23:41.159663Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T01:23:41.159663Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SciExplore: Evaluating Autonomous Agents from Scientific Navigation to Information Integration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bin Liu, Kai Chen, Kuikun Liu, Weiming Zhang, Wenran Liu, Wenwei Zhang, Yanan Sun, Yinhao Tang, Youqing Fang","submitted_at":"2026-07-23T05:17:00Z","abstract_excerpt":"Scientific research involves complex information-seeking and reasoning workflows across heterogeneous sources. However, existing benchmarks primarily emphasize general-domain retrieval or static scientific question answering, and therefore fail to assess key capabilities required in realistic scientific research workflows. We introduce SciExplore, a benchmark designed to evaluate scientific information-seeking and reasoning capabilities of LLMs and agents. SciExplore comprises four task types covering 103 expert-curated tasks across more than ten scientific disciplines: scientific database nav"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.20926","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/2607.20926/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":"2607.20926","created_at":"2026-07-24T01:23:41.163523+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.20926v1","created_at":"2026-07-24T01:23:41.163523+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.20926","created_at":"2026-07-24T01:23:41.163523+00:00"},{"alias_kind":"pith_short_12","alias_value":"2SCLS3WTQWEJ","created_at":"2026-07-24T01:23:41.163523+00:00"},{"alias_kind":"pith_short_16","alias_value":"2SCLS3WTQWEJQTO7","created_at":"2026-07-24T01:23:41.163523+00:00"},{"alias_kind":"pith_short_8","alias_value":"2SCLS3WT","created_at":"2026-07-24T01:23:41.163523+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/2SCLS3WTQWEJQTO7YO6DBWSJYE","json":"https://pith.science/pith/2SCLS3WTQWEJQTO7YO6DBWSJYE.json","graph_json":"https://pith.science/api/pith-number/2SCLS3WTQWEJQTO7YO6DBWSJYE/graph.json","events_json":"https://pith.science/api/pith-number/2SCLS3WTQWEJQTO7YO6DBWSJYE/events.json","paper":"https://pith.science/paper/2SCLS3WT"},"agent_actions":{"view_html":"https://pith.science/pith/2SCLS3WTQWEJQTO7YO6DBWSJYE","download_json":"https://pith.science/pith/2SCLS3WTQWEJQTO7YO6DBWSJYE.json","view_paper":"https://pith.science/paper/2SCLS3WT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.20926&json=true","fetch_graph":"https://pith.science/api/pith-number/2SCLS3WTQWEJQTO7YO6DBWSJYE/graph.json","fetch_events":"https://pith.science/api/pith-number/2SCLS3WTQWEJQTO7YO6DBWSJYE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2SCLS3WTQWEJQTO7YO6DBWSJYE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2SCLS3WTQWEJQTO7YO6DBWSJYE/action/storage_attestation","attest_author":"https://pith.science/pith/2SCLS3WTQWEJQTO7YO6DBWSJYE/action/author_attestation","sign_citation":"https://pith.science/pith/2SCLS3WTQWEJQTO7YO6DBWSJYE/action/citation_signature","submit_replication":"https://pith.science/pith/2SCLS3WTQWEJQTO7YO6DBWSJYE/action/replication_record"}},"created_at":"2026-07-24T01:23:41.163523+00:00","updated_at":"2026-07-24T01:23:41.163523+00:00"}