{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:N3MUIJFJ5U3GXWYMQD6AOONQKR","short_pith_number":"pith:N3MUIJFJ","canonical_record":{"source":{"id":"2411.14054","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-21T11:59:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f09a34a7f264f8ab6db49392f2c3e2ab10f9d03596fb36a72b34bfd2bfdc02b7","abstract_canon_sha256":"97fbec0b28ba945672983a41a38fb572308670cab79f40e9bd46b747e50dfd15"},"schema_version":"1.0"},"canonical_sha256":"6ed94424a9ed366bdb0c80fc0739b05450335ccd1502926551190052dcd9502f","source":{"kind":"arxiv","id":"2411.14054","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.14054","created_at":"2026-07-05T09:38:41Z"},{"alias_kind":"arxiv_version","alias_value":"2411.14054v1","created_at":"2026-07-05T09:38:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.14054","created_at":"2026-07-05T09:38:41Z"},{"alias_kind":"pith_short_12","alias_value":"N3MUIJFJ5U3G","created_at":"2026-07-05T09:38:41Z"},{"alias_kind":"pith_short_16","alias_value":"N3MUIJFJ5U3GXWYM","created_at":"2026-07-05T09:38:41Z"},{"alias_kind":"pith_short_8","alias_value":"N3MUIJFJ","created_at":"2026-07-05T09:38:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:N3MUIJFJ5U3GXWYMQD6AOONQKR","target":"record","payload":{"canonical_record":{"source":{"id":"2411.14054","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-21T11:59:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f09a34a7f264f8ab6db49392f2c3e2ab10f9d03596fb36a72b34bfd2bfdc02b7","abstract_canon_sha256":"97fbec0b28ba945672983a41a38fb572308670cab79f40e9bd46b747e50dfd15"},"schema_version":"1.0"},"canonical_sha256":"6ed94424a9ed366bdb0c80fc0739b05450335ccd1502926551190052dcd9502f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:38:41.035219Z","signature_b64":"DVgB3DDppvqZIek7s+oGSaKNluD+Vw8aojcqu1WZIOcMVf11K7ss/kPYQ7R7wyAYWRThvUIKw0UavY/9uS/PAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ed94424a9ed366bdb0c80fc0739b05450335ccd1502926551190052dcd9502f","last_reissued_at":"2026-07-05T09:38:41.034806Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:38:41.034806Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.14054","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:38:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"j80MKovDOt/bRr9F6RZ1Hq4mrFDVTXbvN8bGp/6wwFSd9Ur5aujGqbVVWS/76mTqsMmVcRPT3WidHnVKKHyVBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T02:35:16.258038Z"},"content_sha256":"3a4a71045377646991f11cd0a3608b1373611a7c6ec706d4e2e51e6c9bcb9658","schema_version":"1.0","event_id":"sha256:3a4a71045377646991f11cd0a3608b1373611a7c6ec706d4e2e51e6c9bcb9658"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:N3MUIJFJ5U3GXWYMQD6AOONQKR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FunctionChat-Bench: Comprehensive Evaluation of Language Models' Generative Capabilities in Korean Tool-use Dialogs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Byeongil Ko, Daniel Lee, Gaeun Seo, Myeongcheol Shin, Shinbok Lee, Sunghee Jung","submitted_at":"2024-11-21T11:59:13Z","abstract_excerpt":"This study investigates language models' generative capabilities in tool-use dialogs. We categorize the models' outputs in tool-use dialogs into four distinct types: Tool Call, Answer Completion, Slot Question, and Relevance Detection, which serve as aspects for evaluation. We introduce FunctionChat-Bench, comprising 700 evaluation items and automated assessment programs. Using this benchmark, we evaluate several language models that support function calling. Our findings indicate that while language models may exhibit high accuracy in single-turn Tool Call scenarios, this does not necessarily"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.14054","kind":"arxiv","version":1},"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/2411.14054/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":"2026-07-05T09:38:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qcxsRfL4Ybsn0p4Gtp8Lbuulxt42LCH1ai5nBS0oGT6NSp69/l5vW5MYUQUpWElyHuXPrTieBv89hiHLUGrqDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T02:35:16.258641Z"},"content_sha256":"02e0d04c53bcfbe31d27b1184a82c8a4c4d5ff2cd474b68ac289abd484b8b2e8","schema_version":"1.0","event_id":"sha256:02e0d04c53bcfbe31d27b1184a82c8a4c4d5ff2cd474b68ac289abd484b8b2e8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/N3MUIJFJ5U3GXWYMQD6AOONQKR/bundle.json","state_url":"https://pith.science/pith/N3MUIJFJ5U3GXWYMQD6AOONQKR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/N3MUIJFJ5U3GXWYMQD6AOONQKR/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":"2026-08-13T02:35:16Z","links":{"resolver":"https://pith.science/pith/N3MUIJFJ5U3GXWYMQD6AOONQKR","bundle":"https://pith.science/pith/N3MUIJFJ5U3GXWYMQD6AOONQKR/bundle.json","state":"https://pith.science/pith/N3MUIJFJ5U3GXWYMQD6AOONQKR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/N3MUIJFJ5U3GXWYMQD6AOONQKR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:N3MUIJFJ5U3GXWYMQD6AOONQKR","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":"97fbec0b28ba945672983a41a38fb572308670cab79f40e9bd46b747e50dfd15","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-21T11:59:13Z","title_canon_sha256":"f09a34a7f264f8ab6db49392f2c3e2ab10f9d03596fb36a72b34bfd2bfdc02b7"},"schema_version":"1.0","source":{"id":"2411.14054","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.14054","created_at":"2026-07-05T09:38:41Z"},{"alias_kind":"arxiv_version","alias_value":"2411.14054v1","created_at":"2026-07-05T09:38:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.14054","created_at":"2026-07-05T09:38:41Z"},{"alias_kind":"pith_short_12","alias_value":"N3MUIJFJ5U3G","created_at":"2026-07-05T09:38:41Z"},{"alias_kind":"pith_short_16","alias_value":"N3MUIJFJ5U3GXWYM","created_at":"2026-07-05T09:38:41Z"},{"alias_kind":"pith_short_8","alias_value":"N3MUIJFJ","created_at":"2026-07-05T09:38:41Z"}],"graph_snapshots":[{"event_id":"sha256:02e0d04c53bcfbe31d27b1184a82c8a4c4d5ff2cd474b68ac289abd484b8b2e8","target":"graph","created_at":"2026-07-05T09:38:41Z","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/2411.14054/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This study investigates language models' generative capabilities in tool-use dialogs. We categorize the models' outputs in tool-use dialogs into four distinct types: Tool Call, Answer Completion, Slot Question, and Relevance Detection, which serve as aspects for evaluation. We introduce FunctionChat-Bench, comprising 700 evaluation items and automated assessment programs. Using this benchmark, we evaluate several language models that support function calling. Our findings indicate that while language models may exhibit high accuracy in single-turn Tool Call scenarios, this does not necessarily","authors_text":"Byeongil Ko, Daniel Lee, Gaeun Seo, Myeongcheol Shin, Shinbok Lee, Sunghee Jung","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-21T11:59:13Z","title":"FunctionChat-Bench: Comprehensive Evaluation of Language Models' Generative Capabilities in Korean Tool-use Dialogs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.14054","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":"sha256:3a4a71045377646991f11cd0a3608b1373611a7c6ec706d4e2e51e6c9bcb9658","target":"record","created_at":"2026-07-05T09:38:41Z","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":"97fbec0b28ba945672983a41a38fb572308670cab79f40e9bd46b747e50dfd15","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-21T11:59:13Z","title_canon_sha256":"f09a34a7f264f8ab6db49392f2c3e2ab10f9d03596fb36a72b34bfd2bfdc02b7"},"schema_version":"1.0","source":{"id":"2411.14054","kind":"arxiv","version":1}},"canonical_sha256":"6ed94424a9ed366bdb0c80fc0739b05450335ccd1502926551190052dcd9502f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6ed94424a9ed366bdb0c80fc0739b05450335ccd1502926551190052dcd9502f","first_computed_at":"2026-07-05T09:38:41.034806Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:38:41.034806Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DVgB3DDppvqZIek7s+oGSaKNluD+Vw8aojcqu1WZIOcMVf11K7ss/kPYQ7R7wyAYWRThvUIKw0UavY/9uS/PAg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:38:41.035219Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.14054","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3a4a71045377646991f11cd0a3608b1373611a7c6ec706d4e2e51e6c9bcb9658","sha256:02e0d04c53bcfbe31d27b1184a82c8a4c4d5ff2cd474b68ac289abd484b8b2e8"],"state_sha256":"0962a9d077243fd4db05a9894803fadadf7c839a9608b06150201d73ef4683eb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"K0xDP1+4tGu4Qk4+7J70GFkziAxktKK5W51qqunaxjyJ9o1MZ9GjK5scLOLVU48dgD9btiMiu7m63pTGhE3TAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T02:35:16.262350Z","bundle_sha256":"cefb3584b843c2f244c65ac871395529fd475a117d3c622c478f33e26104a705"}}