{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:RSJG4OBM4CJDYEWM6X6ROGIRKG","short_pith_number":"pith:RSJG4OBM","canonical_record":{"source":{"id":"2403.07714","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-12T14:57:40Z","cross_cats_sorted":[],"title_canon_sha256":"27dfabeaefae83fa308d6d7570096d84aeaae9f2ddaaf3ea1364a64dceecee0b","abstract_canon_sha256":"3fc69f0a02d1407fd2462ce29e69f05415f556777a84f5bff598f6e2d39d2500"},"schema_version":"1.0"},"canonical_sha256":"8c926e382ce0923c12ccf5fd171911519ed5209168b9abd43c666b6fbb2be9f4","source":{"kind":"arxiv","id":"2403.07714","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.07714","created_at":"2026-07-05T10:24:20Z"},{"alias_kind":"arxiv_version","alias_value":"2403.07714v5","created_at":"2026-07-05T10:24:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.07714","created_at":"2026-07-05T10:24:20Z"},{"alias_kind":"pith_short_12","alias_value":"RSJG4OBM4CJD","created_at":"2026-07-05T10:24:20Z"},{"alias_kind":"pith_short_16","alias_value":"RSJG4OBM4CJDYEWM","created_at":"2026-07-05T10:24:20Z"},{"alias_kind":"pith_short_8","alias_value":"RSJG4OBM","created_at":"2026-07-05T10:24:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:RSJG4OBM4CJDYEWM6X6ROGIRKG","target":"record","payload":{"canonical_record":{"source":{"id":"2403.07714","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-12T14:57:40Z","cross_cats_sorted":[],"title_canon_sha256":"27dfabeaefae83fa308d6d7570096d84aeaae9f2ddaaf3ea1364a64dceecee0b","abstract_canon_sha256":"3fc69f0a02d1407fd2462ce29e69f05415f556777a84f5bff598f6e2d39d2500"},"schema_version":"1.0"},"canonical_sha256":"8c926e382ce0923c12ccf5fd171911519ed5209168b9abd43c666b6fbb2be9f4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:24:20.577951Z","signature_b64":"ERXpY3QB32EM+MdjV9R+aItO4N/X+ZvhG97Pq/U1r7mDXOsEWBPVtO2o82HI9QkADc1Tu2RCwoo6Cxp09AIZCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8c926e382ce0923c12ccf5fd171911519ed5209168b9abd43c666b6fbb2be9f4","last_reissued_at":"2026-07-05T10:24:20.577074Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:24:20.577074Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.07714","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:24:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mMzCVamsW7RI88EiiktvLI+3QIJy/sy49Qw1+oZjxhYlOcL6iKWZkDilSOR2YdcvHknZrDcJC5JROJAEnreQAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T06:06:30.449579Z"},"content_sha256":"833b3d909eb094655f0a7793ea07e8da89f0cceb122b20ade2fa192a61c759ef","schema_version":"1.0","event_id":"sha256:833b3d909eb094655f0a7793ea07e8da89f0cceb122b20ade2fa192a61c759ef"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:RSJG4OBM4CJDYEWM6X6ROGIRKG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hao Wang, Maosong Sun, Peng Li, Shihao Liang, Sijie Cheng, Yang Liu, Yujia Qin, Zhicheng Guo, Zhiyuan Liu","submitted_at":"2024-03-12T14:57:40Z","abstract_excerpt":"Large Language Models (LLMs) have witnessed remarkable advancements in recent years, prompting the exploration of tool learning, which integrates LLMs with external tools to address diverse real-world challenges. Assessing the capability of LLMs to utilise tools necessitates large-scale and stable benchmarks. However, previous works relied on either hand-crafted online tools with limited scale, or large-scale real online APIs suffering from instability of API status. To address this problem, we introduce StableToolBench, a benchmark evolving from ToolBench, proposing a virtual API server and s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.07714","kind":"arxiv","version":5},"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/2403.07714/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":"2026-07-05T10:24:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9R4VMjNKTf5b8YIp8WQEIBTFfcLL0THarcOdXGJeJY0Tdirw6zzN8tT6dekAgihfeFjypDORwLKB3BIT/jkuBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T06:06:30.450132Z"},"content_sha256":"f34cf3028080a36f4ecc3a921326480583f7e59eed4ba1bfcb8b24cbc0f128e9","schema_version":"1.0","event_id":"sha256:f34cf3028080a36f4ecc3a921326480583f7e59eed4ba1bfcb8b24cbc0f128e9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RSJG4OBM4CJDYEWM6X6ROGIRKG/bundle.json","state_url":"https://pith.science/pith/RSJG4OBM4CJDYEWM6X6ROGIRKG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RSJG4OBM4CJDYEWM6X6ROGIRKG/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":"2026-08-04T06:06:30Z","links":{"resolver":"https://pith.science/pith/RSJG4OBM4CJDYEWM6X6ROGIRKG","bundle":"https://pith.science/pith/RSJG4OBM4CJDYEWM6X6ROGIRKG/bundle.json","state":"https://pith.science/pith/RSJG4OBM4CJDYEWM6X6ROGIRKG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RSJG4OBM4CJDYEWM6X6ROGIRKG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:RSJG4OBM4CJDYEWM6X6ROGIRKG","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":"3fc69f0a02d1407fd2462ce29e69f05415f556777a84f5bff598f6e2d39d2500","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-12T14:57:40Z","title_canon_sha256":"27dfabeaefae83fa308d6d7570096d84aeaae9f2ddaaf3ea1364a64dceecee0b"},"schema_version":"1.0","source":{"id":"2403.07714","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.07714","created_at":"2026-07-05T10:24:20Z"},{"alias_kind":"arxiv_version","alias_value":"2403.07714v5","created_at":"2026-07-05T10:24:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.07714","created_at":"2026-07-05T10:24:20Z"},{"alias_kind":"pith_short_12","alias_value":"RSJG4OBM4CJD","created_at":"2026-07-05T10:24:20Z"},{"alias_kind":"pith_short_16","alias_value":"RSJG4OBM4CJDYEWM","created_at":"2026-07-05T10:24:20Z"},{"alias_kind":"pith_short_8","alias_value":"RSJG4OBM","created_at":"2026-07-05T10:24:20Z"}],"graph_snapshots":[{"event_id":"sha256:f34cf3028080a36f4ecc3a921326480583f7e59eed4ba1bfcb8b24cbc0f128e9","target":"graph","created_at":"2026-07-05T10:24:20Z","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/2403.07714/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have witnessed remarkable advancements in recent years, prompting the exploration of tool learning, which integrates LLMs with external tools to address diverse real-world challenges. Assessing the capability of LLMs to utilise tools necessitates large-scale and stable benchmarks. However, previous works relied on either hand-crafted online tools with limited scale, or large-scale real online APIs suffering from instability of API status. To address this problem, we introduce StableToolBench, a benchmark evolving from ToolBench, proposing a virtual API server and s","authors_text":"Hao Wang, Maosong Sun, Peng Li, Shihao Liang, Sijie Cheng, Yang Liu, Yujia Qin, Zhicheng Guo, Zhiyuan Liu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-12T14:57:40Z","title":"StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.07714","kind":"arxiv","version":5},"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":"sha256:833b3d909eb094655f0a7793ea07e8da89f0cceb122b20ade2fa192a61c759ef","target":"record","created_at":"2026-07-05T10:24:20Z","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":"3fc69f0a02d1407fd2462ce29e69f05415f556777a84f5bff598f6e2d39d2500","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-12T14:57:40Z","title_canon_sha256":"27dfabeaefae83fa308d6d7570096d84aeaae9f2ddaaf3ea1364a64dceecee0b"},"schema_version":"1.0","source":{"id":"2403.07714","kind":"arxiv","version":5}},"canonical_sha256":"8c926e382ce0923c12ccf5fd171911519ed5209168b9abd43c666b6fbb2be9f4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8c926e382ce0923c12ccf5fd171911519ed5209168b9abd43c666b6fbb2be9f4","first_computed_at":"2026-07-05T10:24:20.577074Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:24:20.577074Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ERXpY3QB32EM+MdjV9R+aItO4N/X+ZvhG97Pq/U1r7mDXOsEWBPVtO2o82HI9QkADc1Tu2RCwoo6Cxp09AIZCg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:24:20.577951Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.07714","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:833b3d909eb094655f0a7793ea07e8da89f0cceb122b20ade2fa192a61c759ef","sha256:f34cf3028080a36f4ecc3a921326480583f7e59eed4ba1bfcb8b24cbc0f128e9"],"state_sha256":"6e318aced5033e8fda1421b212306bd4bbae3cfe2f489d7ca7b9dca3c8c31655"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FwZHzsLgsg5WQOYBrPbrAmBnYB4CpMy663POCLGYqcJVFdtWInF+I2IcZCv6hIybE1nLZIt+Qgui27gLiS4wBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T06:06:30.454685Z","bundle_sha256":"aff7ee72927bb57836fdc9aec410d523b54297996fcf3c74602fc57ea9199e3c"}}