{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OV7B2DLVLTGKWHJASJJ4PJOH6M","short_pith_number":"pith:OV7B2DLV","canonical_record":{"source":{"id":"2504.07385","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-10T02:08:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e316af07f995b4fa80dff6931069c111da0a1bbc45aaee0559df0f82a2602442","abstract_canon_sha256":"e07e1309ff48808a4b6c02fc57256c9c5733ca6b78b03aa69b6ea7e582082f68"},"schema_version":"1.0"},"canonical_sha256":"757e1d0d755cccab1d209253c7a5c7f325cbab9f24f4e9c5a507f535923b05a6","source":{"kind":"arxiv","id":"2504.07385","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.07385","created_at":"2026-07-01T01:17:39Z"},{"alias_kind":"arxiv_version","alias_value":"2504.07385v3","created_at":"2026-07-01T01:17:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.07385","created_at":"2026-07-01T01:17:39Z"},{"alias_kind":"pith_short_12","alias_value":"OV7B2DLVLTGK","created_at":"2026-07-01T01:17:39Z"},{"alias_kind":"pith_short_16","alias_value":"OV7B2DLVLTGKWHJA","created_at":"2026-07-01T01:17:39Z"},{"alias_kind":"pith_short_8","alias_value":"OV7B2DLV","created_at":"2026-07-01T01:17:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OV7B2DLVLTGKWHJASJJ4PJOH6M","target":"record","payload":{"canonical_record":{"source":{"id":"2504.07385","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-10T02:08:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e316af07f995b4fa80dff6931069c111da0a1bbc45aaee0559df0f82a2602442","abstract_canon_sha256":"e07e1309ff48808a4b6c02fc57256c9c5733ca6b78b03aa69b6ea7e582082f68"},"schema_version":"1.0"},"canonical_sha256":"757e1d0d755cccab1d209253c7a5c7f325cbab9f24f4e9c5a507f535923b05a6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T01:17:39.526493Z","signature_b64":"+dJRPa1ojExSdmjy9PG1SKkfqQ7/aoPXpKGzomIPreekErfKk2dBMmxoEp64/z6G/y5mh9N0lSXDm4VakVEwCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"757e1d0d755cccab1d209253c7a5c7f325cbab9f24f4e9c5a507f535923b05a6","last_reissued_at":"2026-07-01T01:17:39.525934Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T01:17:39.525934Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.07385","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-01T01:17:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xYIYX7eYfDL7gp7s1WfxgkszzuVvSQQ9qYx5isIO2yPgPYMYSaXJG6xWpCeow0tydM8iApxHkfzS8sEwTULfCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-03T02:00:33.857950Z"},"content_sha256":"682d85d33570026840d08683b792bb195e9838fb06dbd97dcff0dd23e08b5745","schema_version":"1.0","event_id":"sha256:682d85d33570026840d08683b792bb195e9838fb06dbd97dcff0dd23e08b5745"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OV7B2DLVLTGKWHJASJJ4PJOH6M","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SAGE: A Search-AuGmented Evaluation of Large Language Models on Free-Form QA","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ali Emami, Hassan Sajjad, Sher Badshah","submitted_at":"2025-04-10T02:08:41Z","abstract_excerpt":"As Large Language Models (LLMs) become increasingly used for question-answering (QA), relying on static, pre-annotated references for evaluation poses significant challenges in cost, scalability, and completeness. Meanwhile, using LLMs themselves as evaluators without external grounding remains unreliable for objective tasks, as they systematically over-accept incorrect answers, fabricate supporting rationales, and degrade sharply on questions that fall outside their training data. We propose Search-AuGmented Evaluation (SAGE), a framework to assess LLM outputs without fixed ground-truth answe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.07385","kind":"arxiv","version":3},"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/2504.07385/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-01T01:17:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6zOfTR5xpJA346Lu9L9iBF7DLdOten1l6SwUhf7jiREaPkAz3nP+fhjMMoaCzFVlwt0SFHH90Uwd2XdGFQ6SAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-03T02:00:33.858354Z"},"content_sha256":"0fe65a8c03ec448b7abf2d121e83156c058fb4a3427b24a5fcdfea19285f4ffa","schema_version":"1.0","event_id":"sha256:0fe65a8c03ec448b7abf2d121e83156c058fb4a3427b24a5fcdfea19285f4ffa"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OV7B2DLVLTGKWHJASJJ4PJOH6M/bundle.json","state_url":"https://pith.science/pith/OV7B2DLVLTGKWHJASJJ4PJOH6M/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OV7B2DLVLTGKWHJASJJ4PJOH6M/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-03T02:00:33Z","links":{"resolver":"https://pith.science/pith/OV7B2DLVLTGKWHJASJJ4PJOH6M","bundle":"https://pith.science/pith/OV7B2DLVLTGKWHJASJJ4PJOH6M/bundle.json","state":"https://pith.science/pith/OV7B2DLVLTGKWHJASJJ4PJOH6M/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OV7B2DLVLTGKWHJASJJ4PJOH6M/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OV7B2DLVLTGKWHJASJJ4PJOH6M","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":"e07e1309ff48808a4b6c02fc57256c9c5733ca6b78b03aa69b6ea7e582082f68","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-10T02:08:41Z","title_canon_sha256":"e316af07f995b4fa80dff6931069c111da0a1bbc45aaee0559df0f82a2602442"},"schema_version":"1.0","source":{"id":"2504.07385","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.07385","created_at":"2026-07-01T01:17:39Z"},{"alias_kind":"arxiv_version","alias_value":"2504.07385v3","created_at":"2026-07-01T01:17:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.07385","created_at":"2026-07-01T01:17:39Z"},{"alias_kind":"pith_short_12","alias_value":"OV7B2DLVLTGK","created_at":"2026-07-01T01:17:39Z"},{"alias_kind":"pith_short_16","alias_value":"OV7B2DLVLTGKWHJA","created_at":"2026-07-01T01:17:39Z"},{"alias_kind":"pith_short_8","alias_value":"OV7B2DLV","created_at":"2026-07-01T01:17:39Z"}],"graph_snapshots":[{"event_id":"sha256:0fe65a8c03ec448b7abf2d121e83156c058fb4a3427b24a5fcdfea19285f4ffa","target":"graph","created_at":"2026-07-01T01:17:39Z","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/2504.07385/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As Large Language Models (LLMs) become increasingly used for question-answering (QA), relying on static, pre-annotated references for evaluation poses significant challenges in cost, scalability, and completeness. Meanwhile, using LLMs themselves as evaluators without external grounding remains unreliable for objective tasks, as they systematically over-accept incorrect answers, fabricate supporting rationales, and degrade sharply on questions that fall outside their training data. We propose Search-AuGmented Evaluation (SAGE), a framework to assess LLM outputs without fixed ground-truth answe","authors_text":"Ali Emami, Hassan Sajjad, Sher Badshah","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-10T02:08:41Z","title":"SAGE: A Search-AuGmented Evaluation of Large Language Models on Free-Form QA"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.07385","kind":"arxiv","version":3},"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":"sha256:682d85d33570026840d08683b792bb195e9838fb06dbd97dcff0dd23e08b5745","target":"record","created_at":"2026-07-01T01:17:39Z","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":"e07e1309ff48808a4b6c02fc57256c9c5733ca6b78b03aa69b6ea7e582082f68","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-10T02:08:41Z","title_canon_sha256":"e316af07f995b4fa80dff6931069c111da0a1bbc45aaee0559df0f82a2602442"},"schema_version":"1.0","source":{"id":"2504.07385","kind":"arxiv","version":3}},"canonical_sha256":"757e1d0d755cccab1d209253c7a5c7f325cbab9f24f4e9c5a507f535923b05a6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"757e1d0d755cccab1d209253c7a5c7f325cbab9f24f4e9c5a507f535923b05a6","first_computed_at":"2026-07-01T01:17:39.525934Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-01T01:17:39.525934Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+dJRPa1ojExSdmjy9PG1SKkfqQ7/aoPXpKGzomIPreekErfKk2dBMmxoEp64/z6G/y5mh9N0lSXDm4VakVEwCA==","signature_status":"signed_v1","signed_at":"2026-07-01T01:17:39.526493Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.07385","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:682d85d33570026840d08683b792bb195e9838fb06dbd97dcff0dd23e08b5745","sha256:0fe65a8c03ec448b7abf2d121e83156c058fb4a3427b24a5fcdfea19285f4ffa"],"state_sha256":"60f46272dab3166a26ab31ea6b60b0b55c7d9dbd4f6bdc71e05480c11c088a5b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NNkr+JahnirQXzOIp2IhBFlW+nAZBOG156Ov5bhgXjk7yciViON+fPg7/6QbkLCABMwB1Clbm+KIb4yh5lKcCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-03T02:00:33.860418Z","bundle_sha256":"ab6b1b39960b8b65de2f36663be5f98f9660d1add8c837d7a98a27bd64378b73"}}