{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:FLRJUBD5ZPBZYBDODAKUXLKJ2B","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":"03d522acd2d51fa57f7ae7933c0625eb1c0b03f62f83bb9043ca280ca0e14b53","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-30T19:33:58Z","title_canon_sha256":"bb8753a0280773e64866bcbeb6fb89609d183c958fb9de9b7ce31e1dd7680f39"},"schema_version":"1.0","source":{"id":"2607.18245","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.18245","created_at":"2026-07-22T00:22:38Z"},{"alias_kind":"arxiv_version","alias_value":"2607.18245v1","created_at":"2026-07-22T00:22:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.18245","created_at":"2026-07-22T00:22:38Z"},{"alias_kind":"pith_short_12","alias_value":"FLRJUBD5ZPBZ","created_at":"2026-07-22T00:22:38Z"},{"alias_kind":"pith_short_16","alias_value":"FLRJUBD5ZPBZYBDO","created_at":"2026-07-22T00:22:38Z"},{"alias_kind":"pith_short_8","alias_value":"FLRJUBD5","created_at":"2026-07-22T00:22:38Z"}],"graph_snapshots":[{"event_id":"sha256:bcbcf61056eec1de31389d16d55dfa81f116b301aeaddec24c153d8ea5c0200a","target":"graph","created_at":"2026-07-22T00:22:38Z","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/2607.18245/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Exact-match evaluation of agent-calling obscures qualitatively different failure modes: a model may select the right function yet hallucinate argument values, or satisfy a schema while choosing a agent for the wrong reason. Existing benchmarks collapse these distinctions into a single binary score, leaving practitioners unable to diagnose where agent calls fail. We propose SAAG a cascaded diagnostic framework that decomposes agent-calling evaluation into three sequential stages: registry conformance, structural completeness, and argument grounding, each producing interpretable stage-specific d","authors_text":"Amit Sheth, Anvi Kohli, Ritvik Garimella, Vedant Khandelwal","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-30T19:33:58Z","title":"SAAG: Structured Agent Assessment and Grounding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.18245","kind":"arxiv","version":1},"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":"sha256:95513c58a61fb7fae463908b4f7c09f428ac48d5c10f49bb0df0a444511bb1bc","target":"record","created_at":"2026-07-22T00:22:38Z","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":"03d522acd2d51fa57f7ae7933c0625eb1c0b03f62f83bb9043ca280ca0e14b53","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-30T19:33:58Z","title_canon_sha256":"bb8753a0280773e64866bcbeb6fb89609d183c958fb9de9b7ce31e1dd7680f39"},"schema_version":"1.0","source":{"id":"2607.18245","kind":"arxiv","version":1}},"canonical_sha256":"2ae29a047dcbc39c046e18154bad49d051ec70b3dfe1d4d4fdc7520710072f08","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2ae29a047dcbc39c046e18154bad49d051ec70b3dfe1d4d4fdc7520710072f08","first_computed_at":"2026-07-22T00:22:38.316921Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-22T00:22:38.316921Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/O7oUi5auQNFuadgeZpUDcdd8OGJ4p7DcZII4pSGHsL18VxrZ21YIpcaIhjdQ71qHExh++UXuIh7NMRazQ37Dg==","signature_status":"signed_v1","signed_at":"2026-07-22T00:22:38.317766Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.18245","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:95513c58a61fb7fae463908b4f7c09f428ac48d5c10f49bb0df0a444511bb1bc","sha256:bcbcf61056eec1de31389d16d55dfa81f116b301aeaddec24c153d8ea5c0200a"],"state_sha256":"ab12c1cbbdd48f17e7bf1429177706f1c90623d0e929ab4208516db31c560eaf"}