{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:K4TXAHQZV35BL3VVAS5ZTP7Y3C","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":"690afd834db0c38d1c4cfb2c3e7e500f277ce89f382bdb6fe0cddb94d54dc497","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.IR","submitted_at":"2024-02-21T22:22:30Z","title_canon_sha256":"070b21d0ff9c002864334fccf68f97e33b2040bb3b6bca9f39f705035577dd24"},"schema_version":"1.0","source":{"id":"2402.14151","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.14151","created_at":"2026-07-05T08:04:18Z"},{"alias_kind":"arxiv_version","alias_value":"2402.14151v2","created_at":"2026-07-05T08:04:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.14151","created_at":"2026-07-05T08:04:18Z"},{"alias_kind":"pith_short_12","alias_value":"K4TXAHQZV35B","created_at":"2026-07-05T08:04:18Z"},{"alias_kind":"pith_short_16","alias_value":"K4TXAHQZV35BL3VV","created_at":"2026-07-05T08:04:18Z"},{"alias_kind":"pith_short_8","alias_value":"K4TXAHQZ","created_at":"2026-07-05T08:04:18Z"}],"graph_snapshots":[{"event_id":"sha256:8feca249743fcfc3b24b27429ce9c69913649928d4655ee6e2733e916b38815a","target":"graph","created_at":"2026-07-05T08:04:18Z","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/2402.14151/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present the Benchmark of Information Retrieval (IR) tasks with Complex Objectives (BIRCO). BIRCO evaluates the ability of IR systems to retrieve documents given multi-faceted user objectives. The benchmark's complexity and compact size make it suitable for evaluating large language model (LLM)-based information retrieval systems. We present a modular framework for investigating factors that may influence LLM performance on retrieval tasks, and identify a simple baseline model which matches or outperforms existing approaches and more complex alternatives. No approach achieves satisfactory pe","authors_text":"Jianyou Wang, Kaicheng Wang, Leon Bergen, Ramamohan Paturi, Weili Cao, Xiaoyue Wang","cross_cats":["cs.AI","cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.IR","submitted_at":"2024-02-21T22:22:30Z","title":"BIRCO: A Benchmark of Information Retrieval Tasks with Complex Objectives"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.14151","kind":"arxiv","version":2},"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":"sha256:7a56b48c0ae904fa247058539c22885cda954dea19bc0623067d59588cbbb59f","target":"record","created_at":"2026-07-05T08:04:18Z","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":"690afd834db0c38d1c4cfb2c3e7e500f277ce89f382bdb6fe0cddb94d54dc497","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.IR","submitted_at":"2024-02-21T22:22:30Z","title_canon_sha256":"070b21d0ff9c002864334fccf68f97e33b2040bb3b6bca9f39f705035577dd24"},"schema_version":"1.0","source":{"id":"2402.14151","kind":"arxiv","version":2}},"canonical_sha256":"5727701e19aefa15eeb504bb99bff8d892c4403af89ed417520ed30d5a7cbacf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5727701e19aefa15eeb504bb99bff8d892c4403af89ed417520ed30d5a7cbacf","first_computed_at":"2026-07-05T08:04:18.192738Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:04:18.192738Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tiB5SR+q7GOIi1o9i2W9jdgnPrw5oJfz/mBbRGbpxIaSqDHFXlrP87SQa3jQp83q2YcKkfkdWEyQQrQ3MvFaCA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:04:18.193213Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.14151","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7a56b48c0ae904fa247058539c22885cda954dea19bc0623067d59588cbbb59f","sha256:8feca249743fcfc3b24b27429ce9c69913649928d4655ee6e2733e916b38815a"],"state_sha256":"75b6a24f01174f90486e39239cdb35717509bfcd12be9741b41c5164451762de"}