{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:3KHTVI32RWK3GZR4YZJOTVRGVS","short_pith_number":"pith:3KHTVI32","canonical_record":{"source":{"id":"2407.08713","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-11T17:50:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"45e4dd412c4bce0bed06aaa57184a28206f28151340b58a5185941e24cc290fd","abstract_canon_sha256":"a2bd5b5447d8df2df6be8caec7e00cbc537c68a1fd2c5c2819fad8a1510ee92b"},"schema_version":"1.0"},"canonical_sha256":"da8f3aa37a8d95b3663cc652e9d626ac8f2cee03f8354865d81ff55d940488dc","source":{"kind":"arxiv","id":"2407.08713","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.08713","created_at":"2026-07-05T09:39:07Z"},{"alias_kind":"arxiv_version","alias_value":"2407.08713v2","created_at":"2026-07-05T09:39:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08713","created_at":"2026-07-05T09:39:07Z"},{"alias_kind":"pith_short_12","alias_value":"3KHTVI32RWK3","created_at":"2026-07-05T09:39:07Z"},{"alias_kind":"pith_short_16","alias_value":"3KHTVI32RWK3GZR4","created_at":"2026-07-05T09:39:07Z"},{"alias_kind":"pith_short_8","alias_value":"3KHTVI32","created_at":"2026-07-05T09:39:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:3KHTVI32RWK3GZR4YZJOTVRGVS","target":"record","payload":{"canonical_record":{"source":{"id":"2407.08713","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-11T17:50:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"45e4dd412c4bce0bed06aaa57184a28206f28151340b58a5185941e24cc290fd","abstract_canon_sha256":"a2bd5b5447d8df2df6be8caec7e00cbc537c68a1fd2c5c2819fad8a1510ee92b"},"schema_version":"1.0"},"canonical_sha256":"da8f3aa37a8d95b3663cc652e9d626ac8f2cee03f8354865d81ff55d940488dc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:39:07.839690Z","signature_b64":"f2fFuoCA3Ax3vP8iJmqmStxZWKuDnzTvdNhNORX3rTvMs1jOlrwQv0+zTvDtXbn1SDBy/lzS02xq4QeEHQEPAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"da8f3aa37a8d95b3663cc652e9d626ac8f2cee03f8354865d81ff55d940488dc","last_reissued_at":"2026-07-05T09:39:07.839198Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:39:07.839198Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.08713","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:39:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EGQsc3ZmkOG5uOKmJzmewQwotQyldWPOv6+/671tjH1CfeYI0w07QBpo1KJvv25yELhX3pGlc+Fb4ngEkzt3BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T23:54:45.779555Z"},"content_sha256":"dbaf867fd5c96741b8d3d61f2dec0c72cefde126c8002c5ae292a52bb49b59cb","schema_version":"1.0","event_id":"sha256:dbaf867fd5c96741b8d3d61f2dec0c72cefde126c8002c5ae292a52bb49b59cb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:3KHTVI32RWK3GZR4YZJOTVRGVS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GTA: A Benchmark for General Tool Agents","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Cailian Chen, Jize Wang, Kai Chen, Songyang Zhang, Xinyi Le, Yining Li, Zerun Ma","submitted_at":"2024-07-11T17:50:09Z","abstract_excerpt":"Significant focus has been placed on integrating large language models (LLMs) with various tools in developing general-purpose agents. This poses a challenge to LLMs' tool-use capabilities. However, there are evident gaps between existing tool-use evaluations and real-world scenarios. Current evaluations often use AI-generated queries, single-step tasks, dummy tools, and text-only interactions, failing to reveal the agents' real-world problem-solving abilities effectively. To address this, we propose GTA, a benchmark for General Tool Agents, featuring three main aspects: (i) Real user queries:"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08713","kind":"arxiv","version":2},"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/2407.08713/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":"2026-07-05T09:39:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QqeW35KSCAe8rcNOuca0x6DITRqh2HQOL22u+JdyRgb4kPGcE3S52/vBExET0+2XMPIoXXRitecQ2fSE7Xd5Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T23:54:45.780035Z"},"content_sha256":"e4c206f8ecb1184d8ca41122146a9e7ffb462f26897144c96969954cf94d4e66","schema_version":"1.0","event_id":"sha256:e4c206f8ecb1184d8ca41122146a9e7ffb462f26897144c96969954cf94d4e66"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3KHTVI32RWK3GZR4YZJOTVRGVS/bundle.json","state_url":"https://pith.science/pith/3KHTVI32RWK3GZR4YZJOTVRGVS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3KHTVI32RWK3GZR4YZJOTVRGVS/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":"2026-08-07T23:54:45Z","links":{"resolver":"https://pith.science/pith/3KHTVI32RWK3GZR4YZJOTVRGVS","bundle":"https://pith.science/pith/3KHTVI32RWK3GZR4YZJOTVRGVS/bundle.json","state":"https://pith.science/pith/3KHTVI32RWK3GZR4YZJOTVRGVS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3KHTVI32RWK3GZR4YZJOTVRGVS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:3KHTVI32RWK3GZR4YZJOTVRGVS","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":"a2bd5b5447d8df2df6be8caec7e00cbc537c68a1fd2c5c2819fad8a1510ee92b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-11T17:50:09Z","title_canon_sha256":"45e4dd412c4bce0bed06aaa57184a28206f28151340b58a5185941e24cc290fd"},"schema_version":"1.0","source":{"id":"2407.08713","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.08713","created_at":"2026-07-05T09:39:07Z"},{"alias_kind":"arxiv_version","alias_value":"2407.08713v2","created_at":"2026-07-05T09:39:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08713","created_at":"2026-07-05T09:39:07Z"},{"alias_kind":"pith_short_12","alias_value":"3KHTVI32RWK3","created_at":"2026-07-05T09:39:07Z"},{"alias_kind":"pith_short_16","alias_value":"3KHTVI32RWK3GZR4","created_at":"2026-07-05T09:39:07Z"},{"alias_kind":"pith_short_8","alias_value":"3KHTVI32","created_at":"2026-07-05T09:39:07Z"}],"graph_snapshots":[{"event_id":"sha256:e4c206f8ecb1184d8ca41122146a9e7ffb462f26897144c96969954cf94d4e66","target":"graph","created_at":"2026-07-05T09:39:07Z","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/2407.08713/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Significant focus has been placed on integrating large language models (LLMs) with various tools in developing general-purpose agents. This poses a challenge to LLMs' tool-use capabilities. However, there are evident gaps between existing tool-use evaluations and real-world scenarios. Current evaluations often use AI-generated queries, single-step tasks, dummy tools, and text-only interactions, failing to reveal the agents' real-world problem-solving abilities effectively. To address this, we propose GTA, a benchmark for General Tool Agents, featuring three main aspects: (i) Real user queries:","authors_text":"Cailian Chen, Jize Wang, Kai Chen, Songyang Zhang, Xinyi Le, Yining Li, Zerun Ma","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-11T17:50:09Z","title":"GTA: A Benchmark for General Tool Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08713","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:dbaf867fd5c96741b8d3d61f2dec0c72cefde126c8002c5ae292a52bb49b59cb","target":"record","created_at":"2026-07-05T09:39:07Z","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":"a2bd5b5447d8df2df6be8caec7e00cbc537c68a1fd2c5c2819fad8a1510ee92b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-11T17:50:09Z","title_canon_sha256":"45e4dd412c4bce0bed06aaa57184a28206f28151340b58a5185941e24cc290fd"},"schema_version":"1.0","source":{"id":"2407.08713","kind":"arxiv","version":2}},"canonical_sha256":"da8f3aa37a8d95b3663cc652e9d626ac8f2cee03f8354865d81ff55d940488dc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"da8f3aa37a8d95b3663cc652e9d626ac8f2cee03f8354865d81ff55d940488dc","first_computed_at":"2026-07-05T09:39:07.839198Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:39:07.839198Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"f2fFuoCA3Ax3vP8iJmqmStxZWKuDnzTvdNhNORX3rTvMs1jOlrwQv0+zTvDtXbn1SDBy/lzS02xq4QeEHQEPAw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:39:07.839690Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.08713","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dbaf867fd5c96741b8d3d61f2dec0c72cefde126c8002c5ae292a52bb49b59cb","sha256:e4c206f8ecb1184d8ca41122146a9e7ffb462f26897144c96969954cf94d4e66"],"state_sha256":"cc9e685cdce81c2bf18908684a4d460d175938cb45d52934a1cf4a889e1d619e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AzXKxgs/WnGiCceTrlPts+LRQGCwwjyPEcxD2DLMyfAwdiQlcpCmnJP9uyvB1AAyURiIS8M3mpdDqY3QozdbCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T23:54:45.784443Z","bundle_sha256":"f139c8cf689d253e693d107a17046428ee92a0724a12dd36b30aed1400771712"}}