{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:B6NTKMESZK3HQCLUUPWGK6OWS6","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":"a301ce441364810d1a60e3104cc2ed2626db445fe222badb978e7735eda9348f","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-28T11:28:04Z","title_canon_sha256":"d721f9f3b9136787c287e204b1e157d6146ca7d96d690039dfbedc933f1cbab9"},"schema_version":"1.0","source":{"id":"2506.22853","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.22853","created_at":"2026-07-05T11:30:45Z"},{"alias_kind":"arxiv_version","alias_value":"2506.22853v2","created_at":"2026-07-05T11:30:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.22853","created_at":"2026-07-05T11:30:45Z"},{"alias_kind":"pith_short_12","alias_value":"B6NTKMESZK3H","created_at":"2026-07-05T11:30:45Z"},{"alias_kind":"pith_short_16","alias_value":"B6NTKMESZK3HQCLU","created_at":"2026-07-05T11:30:45Z"},{"alias_kind":"pith_short_8","alias_value":"B6NTKMES","created_at":"2026-07-05T11:30:45Z"}],"graph_snapshots":[{"event_id":"sha256:a787e008b8d0a88ef61ad2ab605c26f1994b8955e0dda28f51f3e3755fa179db","target":"graph","created_at":"2026-07-05T11:30:45Z","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/2506.22853/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Existing function-calling benchmarks focus on single-turn interactions. However, they overlook the complexity of real-world scenarios. To quantify how existing benchmarks address practical applications, we introduce DICE-SCORE, a metric that evaluates the dispersion of tool-related information such as function name and parameter values throughout the dialogue. Analyzing existing benchmarks through DICE-SCORE reveals notably low scores, highlighting the need for more realistic scenarios. To address this gap, we present DICE-BENCH, a framework that constructs practical function-calling datasets ","authors_text":"Bongwon Suh, Donghyeon Lee, Dongseok Heo, Kyochul Jang, Kyusik Kim, Taewhoo Lee, Woojeong Kim","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-28T11:28:04Z","title":"DICE-BENCH: Evaluating the Tool-Use Capabilities of Large Language Models in Multi-Round, Multi-Party Dialogues"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.22853","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":"sha256:0d860b52143f0c99829f0bc45ad02835869aba3ffa4ffe6cd06ceebcae1b870f","target":"record","created_at":"2026-07-05T11:30:45Z","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":"a301ce441364810d1a60e3104cc2ed2626db445fe222badb978e7735eda9348f","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-28T11:28:04Z","title_canon_sha256":"d721f9f3b9136787c287e204b1e157d6146ca7d96d690039dfbedc933f1cbab9"},"schema_version":"1.0","source":{"id":"2506.22853","kind":"arxiv","version":2}},"canonical_sha256":"0f9b353092cab6780974a3ec6579d697a89708cbfa3d6496a5b9157e6260239a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0f9b353092cab6780974a3ec6579d697a89708cbfa3d6496a5b9157e6260239a","first_computed_at":"2026-07-05T11:30:45.662961Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:30:45.662961Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+yuJFYM65pYo4oTDD+/rwRgDxbDTpBrikKNOZShHvrpcnhD4Otr2X9+0d9vR0w8ZILROV3dipx8G4KhlKXsxDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:30:45.663479Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.22853","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0d860b52143f0c99829f0bc45ad02835869aba3ffa4ffe6cd06ceebcae1b870f","sha256:a787e008b8d0a88ef61ad2ab605c26f1994b8955e0dda28f51f3e3755fa179db"],"state_sha256":"67b65ccd7ef07e1345d5d69f415552e8fa681625e2f1b35435e76d0302a993b6"}