{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:DFV5LZ4VDFVLF4EDMI44UID2O4","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":"b24263f0659e6e6a511d7f9fea421b9800e6b0e409695f097d382105b4114473","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T16:48:11Z","title_canon_sha256":"30f817b01cb3285453757e5dca36e9808a292dbdfd43fd9c7067b58233b5462d"},"schema_version":"1.0","source":{"id":"2506.01859","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01859","created_at":"2026-07-05T11:14:20Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01859v1","created_at":"2026-07-05T11:14:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01859","created_at":"2026-07-05T11:14:20Z"},{"alias_kind":"pith_short_12","alias_value":"DFV5LZ4VDFVL","created_at":"2026-07-05T11:14:20Z"},{"alias_kind":"pith_short_16","alias_value":"DFV5LZ4VDFVLF4ED","created_at":"2026-07-05T11:14:20Z"},{"alias_kind":"pith_short_8","alias_value":"DFV5LZ4V","created_at":"2026-07-05T11:14:20Z"}],"graph_snapshots":[{"event_id":"sha256:d86dced00716f2760c5384de66a12992beb7bc387a16006a9039e13de4180a46","target":"graph","created_at":"2026-07-05T11:14:20Z","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/2506.01859/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce Conversational Function-Calling Evaluation Through Turn-Level Interactions (CONFETTI), a conversational benchmark1 designed to evaluate the function-calling capabilities and response quality of large language models (LLMs). Current benchmarks lack comprehensive assessment of LLMs in complex conversational scenarios. CONFETTI addresses this gap through 109 human-simulated conversations, comprising 313 user turns and covering 86 APIs. These conversations explicitly target various conversational complexities, such as follow-ups, goal correction and switching, ambiguous and implicit g","authors_text":"Claudia Zaghi, James Gung, Katerina Margatina, Monica Sunkara, Raphael Shu, Tamer Alkhouli, Yi Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T16:48:11Z","title":"CONFETTI: Conversational Function-Calling Evaluation Through Turn-Level Interactions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01859","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:eedbb3e73fc1c33578a85c38601e6ecd32892474655d8a6e07c915139a972142","target":"record","created_at":"2026-07-05T11:14:20Z","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":"b24263f0659e6e6a511d7f9fea421b9800e6b0e409695f097d382105b4114473","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T16:48:11Z","title_canon_sha256":"30f817b01cb3285453757e5dca36e9808a292dbdfd43fd9c7067b58233b5462d"},"schema_version":"1.0","source":{"id":"2506.01859","kind":"arxiv","version":1}},"canonical_sha256":"196bd5e795196ab2f0836239ca207a770e2b82cf2ed0ddbfdd2285ec2e989d0f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"196bd5e795196ab2f0836239ca207a770e2b82cf2ed0ddbfdd2285ec2e989d0f","first_computed_at":"2026-07-05T11:14:20.510364Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:20.510364Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3Jrv8DI8WSh9sl/Drp8oUiK3OqzTXCXt2kOymuzvv4cbT3neVfVZtUBAHXiq5KWkr0w2EPAUFK0h/dYRKM5NCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:20.510815Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.01859","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:eedbb3e73fc1c33578a85c38601e6ecd32892474655d8a6e07c915139a972142","sha256:d86dced00716f2760c5384de66a12992beb7bc387a16006a9039e13de4180a46"],"state_sha256":"83d3c2904b7c956f98bba3783bc543e08ccc1752d7c3f5d4d1ac6015078cadae"}