{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:QFDEQQ3Y5ROTVOZJJIVC4VXUEZ","short_pith_number":"pith:QFDEQQ3Y","canonical_record":{"source":{"id":"2506.13977","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-06-11T17:59:18Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"b04a48b19d25da6973e0eb7a486647a9817b77d632af2d63ee5f1ef5522e73e3","abstract_canon_sha256":"d3127440d71df07687f7645066d95887ae00130903f1cbfea160d5d0d512760a"},"schema_version":"1.0"},"canonical_sha256":"8146484378ec5d3abb294a2a2e56f42667c8ea9c41255e5506bcfa8a7dde93e2","source":{"kind":"arxiv","id":"2506.13977","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.13977","created_at":"2026-07-05T11:22:46Z"},{"alias_kind":"arxiv_version","alias_value":"2506.13977v1","created_at":"2026-07-05T11:22:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.13977","created_at":"2026-07-05T11:22:46Z"},{"alias_kind":"pith_short_12","alias_value":"QFDEQQ3Y5ROT","created_at":"2026-07-05T11:22:46Z"},{"alias_kind":"pith_short_16","alias_value":"QFDEQQ3Y5ROTVOZJ","created_at":"2026-07-05T11:22:46Z"},{"alias_kind":"pith_short_8","alias_value":"QFDEQQ3Y","created_at":"2026-07-05T11:22:46Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:QFDEQQ3Y5ROTVOZJJIVC4VXUEZ","target":"record","payload":{"canonical_record":{"source":{"id":"2506.13977","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-06-11T17:59:18Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"b04a48b19d25da6973e0eb7a486647a9817b77d632af2d63ee5f1ef5522e73e3","abstract_canon_sha256":"d3127440d71df07687f7645066d95887ae00130903f1cbfea160d5d0d512760a"},"schema_version":"1.0"},"canonical_sha256":"8146484378ec5d3abb294a2a2e56f42667c8ea9c41255e5506bcfa8a7dde93e2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:22:46.550562Z","signature_b64":"N/gQHVtx582zFmVrYYAkgUnz3TvwkN3uqMVZcAlkerx8pbHcpkyemKbeJSKJkEHJoSANEhnnBaK1NxWG36jFCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8146484378ec5d3abb294a2a2e56f42667c8ea9c41255e5506bcfa8a7dde93e2","last_reissued_at":"2026-07-05T11:22:46.550078Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:22:46.550078Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.13977","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":"2026-07-05T11:22:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fq+gAal8DxiflFn7C4Ktq0/GZBt/LAz4KONutmC05M2aMBxClpi4sDLP6OoLmhIpovjtWlfjap0/8Dz37FhfBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T03:11:44.963544Z"},"content_sha256":"eb086928614dc4f7ca57e207aa346449421bb6b071bf6ab1bf8dd3fadd3c6777","schema_version":"1.0","event_id":"sha256:eb086928614dc4f7ca57e207aa346449421bb6b071bf6ab1bf8dd3fadd3c6777"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:QFDEQQ3Y5ROTVOZJJIVC4VXUEZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CRITICTOOL: Evaluating Self-Critique Capabilities of Large Language Models in Tool-Calling Error Scenarios","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Feng Zhao, Junjie Ye, Lin Chen, Qi Mao, Shiting Huang, Siyu Yuan, Yu Zeng, Zehui Chen, Zhen Fang","submitted_at":"2025-06-11T17:59:18Z","abstract_excerpt":"The ability of large language models (LLMs) to utilize external tools has enabled them to tackle an increasingly diverse range of tasks. However, as the tasks become more complex and long-horizon, the intricate tool utilization process may trigger various unexpected errors. Therefore, how to effectively handle such errors, including identifying, diagnosing, and recovering from them, has emerged as a key research direction for advancing tool learning. In this work, we first extensively analyze the types of errors encountered during the function-calling process on several competitive tool evalua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.13977","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/2506.13977/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":"2026-07-05T11:22:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y4V33ZsO7Pm55gKC9vfNzo77mtfSFGMNX4vCVDvZt2LA/+hrgXqs5FSF3ZJv98S8Dy7WzZ9ZHizv2XZ3ks5ABg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T03:11:44.964472Z"},"content_sha256":"b39ad441bf73550c8eae08023224d4f7e8d12e327e4a6f51bb4c3efa6eee7330","schema_version":"1.0","event_id":"sha256:b39ad441bf73550c8eae08023224d4f7e8d12e327e4a6f51bb4c3efa6eee7330"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QFDEQQ3Y5ROTVOZJJIVC4VXUEZ/bundle.json","state_url":"https://pith.science/pith/QFDEQQ3Y5ROTVOZJJIVC4VXUEZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QFDEQQ3Y5ROTVOZJJIVC4VXUEZ/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":"2026-08-06T03:11:44Z","links":{"resolver":"https://pith.science/pith/QFDEQQ3Y5ROTVOZJJIVC4VXUEZ","bundle":"https://pith.science/pith/QFDEQQ3Y5ROTVOZJJIVC4VXUEZ/bundle.json","state":"https://pith.science/pith/QFDEQQ3Y5ROTVOZJJIVC4VXUEZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QFDEQQ3Y5ROTVOZJJIVC4VXUEZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:QFDEQQ3Y5ROTVOZJJIVC4VXUEZ","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":"d3127440d71df07687f7645066d95887ae00130903f1cbfea160d5d0d512760a","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-06-11T17:59:18Z","title_canon_sha256":"b04a48b19d25da6973e0eb7a486647a9817b77d632af2d63ee5f1ef5522e73e3"},"schema_version":"1.0","source":{"id":"2506.13977","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.13977","created_at":"2026-07-05T11:22:46Z"},{"alias_kind":"arxiv_version","alias_value":"2506.13977v1","created_at":"2026-07-05T11:22:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.13977","created_at":"2026-07-05T11:22:46Z"},{"alias_kind":"pith_short_12","alias_value":"QFDEQQ3Y5ROT","created_at":"2026-07-05T11:22:46Z"},{"alias_kind":"pith_short_16","alias_value":"QFDEQQ3Y5ROTVOZJ","created_at":"2026-07-05T11:22:46Z"},{"alias_kind":"pith_short_8","alias_value":"QFDEQQ3Y","created_at":"2026-07-05T11:22:46Z"}],"graph_snapshots":[{"event_id":"sha256:b39ad441bf73550c8eae08023224d4f7e8d12e327e4a6f51bb4c3efa6eee7330","target":"graph","created_at":"2026-07-05T11:22:46Z","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/2506.13977/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The ability of large language models (LLMs) to utilize external tools has enabled them to tackle an increasingly diverse range of tasks. However, as the tasks become more complex and long-horizon, the intricate tool utilization process may trigger various unexpected errors. Therefore, how to effectively handle such errors, including identifying, diagnosing, and recovering from them, has emerged as a key research direction for advancing tool learning. In this work, we first extensively analyze the types of errors encountered during the function-calling process on several competitive tool evalua","authors_text":"Feng Zhao, Junjie Ye, Lin Chen, Qi Mao, Shiting Huang, Siyu Yuan, Yu Zeng, Zehui Chen, Zhen Fang","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-06-11T17:59:18Z","title":"CRITICTOOL: Evaluating Self-Critique Capabilities of Large Language Models in Tool-Calling Error Scenarios"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.13977","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:eb086928614dc4f7ca57e207aa346449421bb6b071bf6ab1bf8dd3fadd3c6777","target":"record","created_at":"2026-07-05T11:22:46Z","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":"d3127440d71df07687f7645066d95887ae00130903f1cbfea160d5d0d512760a","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-06-11T17:59:18Z","title_canon_sha256":"b04a48b19d25da6973e0eb7a486647a9817b77d632af2d63ee5f1ef5522e73e3"},"schema_version":"1.0","source":{"id":"2506.13977","kind":"arxiv","version":1}},"canonical_sha256":"8146484378ec5d3abb294a2a2e56f42667c8ea9c41255e5506bcfa8a7dde93e2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8146484378ec5d3abb294a2a2e56f42667c8ea9c41255e5506bcfa8a7dde93e2","first_computed_at":"2026-07-05T11:22:46.550078Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:22:46.550078Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"N/gQHVtx582zFmVrYYAkgUnz3TvwkN3uqMVZcAlkerx8pbHcpkyemKbeJSKJkEHJoSANEhnnBaK1NxWG36jFCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:22:46.550562Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.13977","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:eb086928614dc4f7ca57e207aa346449421bb6b071bf6ab1bf8dd3fadd3c6777","sha256:b39ad441bf73550c8eae08023224d4f7e8d12e327e4a6f51bb4c3efa6eee7330"],"state_sha256":"5e5502ced79c2b2bf080d9bccaf515ffbd7a4d2d22c146c01a9195dd13d713d6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hSOunfJYTI68vz8DFygt631hRiha4AXNlVrrb5DFpkURzO83F4AT0bHJaWIKmQ/sGkGzJQSowjZIS/QxxjiBAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T03:11:44.971020Z","bundle_sha256":"5ac99dd64fff9c893b99a51fb36a4c8054c855faa95ca6b04901d27ee2e30c94"}}