{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:V4V44SXSEVTTRRBSC52FP5RMHZ","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":"d3edfb31dff7c2d3025b09fbd9e8227bdc58e6ad9fc76aac17a3b516dcd3f7ba","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-17T17:44:06Z","title_canon_sha256":"66303809cbf126e2833fdaef32d4296ee4ad4a7e7292a3b79ddbc8b9a12ec7b1"},"schema_version":"1.0","source":{"id":"2504.13128","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.13128","created_at":"2026-07-05T11:20:52Z"},{"alias_kind":"arxiv_version","alias_value":"2504.13128v2","created_at":"2026-07-05T11:20:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.13128","created_at":"2026-07-05T11:20:52Z"},{"alias_kind":"pith_short_12","alias_value":"V4V44SXSEVTT","created_at":"2026-07-05T11:20:52Z"},{"alias_kind":"pith_short_16","alias_value":"V4V44SXSEVTTRRBS","created_at":"2026-07-05T11:20:52Z"},{"alias_kind":"pith_short_8","alias_value":"V4V44SXS","created_at":"2026-07-05T11:20:52Z"}],"graph_snapshots":[{"event_id":"sha256:b69fb1090acac25910fa54f64bc0ad06f6091ba2cd83c9211b8a2b31534dd405","target":"graph","created_at":"2026-07-05T11:20:52Z","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/2504.13128/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce FreshStack, a holistic framework for automatically building information retrieval (IR) evaluation benchmarks by incorporating challenging questions and answers. FreshStack conducts the following steps: (1) automatic corpus collection from code and technical documentation, (2) nugget generation from community-asked questions and answers, and (3) nugget-level support, retrieving documents using a fusion of retrieval techniques and hybrid architectures. We use FreshStack to build five datasets on fast-growing, recent, and niche topics to ensure the tasks are sufficiently challenging.","authors_text":"Andrew Drozdov, Jimmy Lin, Michael Carbin, Nandan Thakur, Omar Khattab, Sam Havens","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-17T17:44:06Z","title":"FreshStack: Building Realistic Benchmarks for Evaluating Retrieval on Technical Documents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.13128","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:dfe3bf755655f4823ce37e5cc11511f88237d8d949a5a27dbcefd8649ae9bd33","target":"record","created_at":"2026-07-05T11:20:52Z","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":"d3edfb31dff7c2d3025b09fbd9e8227bdc58e6ad9fc76aac17a3b516dcd3f7ba","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.IR","submitted_at":"2025-04-17T17:44:06Z","title_canon_sha256":"66303809cbf126e2833fdaef32d4296ee4ad4a7e7292a3b79ddbc8b9a12ec7b1"},"schema_version":"1.0","source":{"id":"2504.13128","kind":"arxiv","version":2}},"canonical_sha256":"af2bce4af2256738c432177457f62c3e4ea3c7021ce2f9f61004d68fe4dae527","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"af2bce4af2256738c432177457f62c3e4ea3c7021ce2f9f61004d68fe4dae527","first_computed_at":"2026-07-05T11:20:52.822647Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:20:52.822647Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1rjICWBtp1zpL/i4JQ4P8Imr9b+S16I1bjDvZ+F25r2rruJljekUb+PiWR/ZiBElRVK6AAHA9rvCioTCOgyOCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:20:52.823143Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.13128","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dfe3bf755655f4823ce37e5cc11511f88237d8d949a5a27dbcefd8649ae9bd33","sha256:b69fb1090acac25910fa54f64bc0ad06f6091ba2cd83c9211b8a2b31534dd405"],"state_sha256":"2079811aed8cf50fa94061f07c5d8961d28300a15a7ea5aa7fcc0a51f370e3f9"}