{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:BCBKPMDXQZVPMBT3IWTBRLTS4J","short_pith_number":"pith:BCBKPMDX","canonical_record":{"source":{"id":"2504.18851","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-26T08:30:02Z","cross_cats_sorted":[],"title_canon_sha256":"27cf471d9e17d99e920fd332c030701d4cc4fdc32f6130edd3b496e13654f712","abstract_canon_sha256":"74eff34debf39b25160b52ce29f00f8b129446e3e38fb565100d4cb08b43f581"},"schema_version":"1.0"},"canonical_sha256":"0882a7b077866af6067b45a618ae72e2692b06515a61f63af685c9d849490595","source":{"kind":"arxiv","id":"2504.18851","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.18851","created_at":"2026-07-05T10:54:22Z"},{"alias_kind":"arxiv_version","alias_value":"2504.18851v1","created_at":"2026-07-05T10:54:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.18851","created_at":"2026-07-05T10:54:22Z"},{"alias_kind":"pith_short_12","alias_value":"BCBKPMDXQZVP","created_at":"2026-07-05T10:54:22Z"},{"alias_kind":"pith_short_16","alias_value":"BCBKPMDXQZVPMBT3","created_at":"2026-07-05T10:54:22Z"},{"alias_kind":"pith_short_8","alias_value":"BCBKPMDX","created_at":"2026-07-05T10:54:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:BCBKPMDXQZVPMBT3IWTBRLTS4J","target":"record","payload":{"canonical_record":{"source":{"id":"2504.18851","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-26T08:30:02Z","cross_cats_sorted":[],"title_canon_sha256":"27cf471d9e17d99e920fd332c030701d4cc4fdc32f6130edd3b496e13654f712","abstract_canon_sha256":"74eff34debf39b25160b52ce29f00f8b129446e3e38fb565100d4cb08b43f581"},"schema_version":"1.0"},"canonical_sha256":"0882a7b077866af6067b45a618ae72e2692b06515a61f63af685c9d849490595","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:54:22.445439Z","signature_b64":"v7xNBUBSVoQjV+DjdzE/s5MLXIDXA9zj9GeKRLbKbYgrfYG67khgvlLOJhVhZJbBOwa3ppD6DLOeGxEnEHJ7AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0882a7b077866af6067b45a618ae72e2692b06515a61f63af685c9d849490595","last_reissued_at":"2026-07-05T10:54:22.444924Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:54:22.444924Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.18851","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":"2026-07-05T10:54:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZV6n93z0MZJmnis+VfcilCR+Xo9LzoKyYtzDFP59FR0sRfTYoUCW4gzkVv7UmL53SLDIzImbbNXjq8bPf4tEAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T20:55:32.211610Z"},"content_sha256":"44b5cc6136379cb5317d19e36b47d32eb89d2dfc9f842523c5a6cfe72a961590","schema_version":"1.0","event_id":"sha256:44b5cc6136379cb5317d19e36b47d32eb89d2dfc9f842523c5a6cfe72a961590"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:BCBKPMDXQZVPMBT3IWTBRLTS4J","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"When2Call: When (not) to Call Tools","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ameya Sunil Mahabaleshwarkar, Hayley Ross, Yoshi Suhara","submitted_at":"2025-04-26T08:30:02Z","abstract_excerpt":"Leveraging external tools is a key feature for modern Language Models (LMs) to expand their capabilities and integrate them into existing systems. However, existing benchmarks primarily focus on the accuracy of tool calling -- whether the correct tool is called with the correct parameters -- and less on evaluating when LMs should (not) call tools. We develop a new benchmark, When2Call, which evaluates tool-calling decision-making: when to generate a tool call, when to ask follow-up questions and when to admit the question can't be answered with the tools provided. We find that state-of-the-art"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.18851","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/2504.18851/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":"2026-07-05T10:54:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JscX4LPT98TM5O0VkGZBhJVuxe/YhTx9p/UmOMiZhN+VxQ4MYGNK4Sp0yv01QjmGaIwzeq7QiGQlcOSBMsArCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T20:55:32.212133Z"},"content_sha256":"c1afb691187b0313cdd2733f671c9caec92e2dada1eff50c51d75f70c240c831","schema_version":"1.0","event_id":"sha256:c1afb691187b0313cdd2733f671c9caec92e2dada1eff50c51d75f70c240c831"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BCBKPMDXQZVPMBT3IWTBRLTS4J/bundle.json","state_url":"https://pith.science/pith/BCBKPMDXQZVPMBT3IWTBRLTS4J/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BCBKPMDXQZVPMBT3IWTBRLTS4J/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":"2026-08-08T20:55:32Z","links":{"resolver":"https://pith.science/pith/BCBKPMDXQZVPMBT3IWTBRLTS4J","bundle":"https://pith.science/pith/BCBKPMDXQZVPMBT3IWTBRLTS4J/bundle.json","state":"https://pith.science/pith/BCBKPMDXQZVPMBT3IWTBRLTS4J/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BCBKPMDXQZVPMBT3IWTBRLTS4J/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BCBKPMDXQZVPMBT3IWTBRLTS4J","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":"74eff34debf39b25160b52ce29f00f8b129446e3e38fb565100d4cb08b43f581","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-26T08:30:02Z","title_canon_sha256":"27cf471d9e17d99e920fd332c030701d4cc4fdc32f6130edd3b496e13654f712"},"schema_version":"1.0","source":{"id":"2504.18851","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.18851","created_at":"2026-07-05T10:54:22Z"},{"alias_kind":"arxiv_version","alias_value":"2504.18851v1","created_at":"2026-07-05T10:54:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.18851","created_at":"2026-07-05T10:54:22Z"},{"alias_kind":"pith_short_12","alias_value":"BCBKPMDXQZVP","created_at":"2026-07-05T10:54:22Z"},{"alias_kind":"pith_short_16","alias_value":"BCBKPMDXQZVPMBT3","created_at":"2026-07-05T10:54:22Z"},{"alias_kind":"pith_short_8","alias_value":"BCBKPMDX","created_at":"2026-07-05T10:54:22Z"}],"graph_snapshots":[{"event_id":"sha256:c1afb691187b0313cdd2733f671c9caec92e2dada1eff50c51d75f70c240c831","target":"graph","created_at":"2026-07-05T10:54:22Z","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/2504.18851/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Leveraging external tools is a key feature for modern Language Models (LMs) to expand their capabilities and integrate them into existing systems. However, existing benchmarks primarily focus on the accuracy of tool calling -- whether the correct tool is called with the correct parameters -- and less on evaluating when LMs should (not) call tools. We develop a new benchmark, When2Call, which evaluates tool-calling decision-making: when to generate a tool call, when to ask follow-up questions and when to admit the question can't be answered with the tools provided. We find that state-of-the-art","authors_text":"Ameya Sunil Mahabaleshwarkar, Hayley Ross, Yoshi Suhara","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-26T08:30:02Z","title":"When2Call: When (not) to Call Tools"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.18851","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":"sha256:44b5cc6136379cb5317d19e36b47d32eb89d2dfc9f842523c5a6cfe72a961590","target":"record","created_at":"2026-07-05T10:54:22Z","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":"74eff34debf39b25160b52ce29f00f8b129446e3e38fb565100d4cb08b43f581","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-26T08:30:02Z","title_canon_sha256":"27cf471d9e17d99e920fd332c030701d4cc4fdc32f6130edd3b496e13654f712"},"schema_version":"1.0","source":{"id":"2504.18851","kind":"arxiv","version":1}},"canonical_sha256":"0882a7b077866af6067b45a618ae72e2692b06515a61f63af685c9d849490595","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0882a7b077866af6067b45a618ae72e2692b06515a61f63af685c9d849490595","first_computed_at":"2026-07-05T10:54:22.444924Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:54:22.444924Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"v7xNBUBSVoQjV+DjdzE/s5MLXIDXA9zj9GeKRLbKbYgrfYG67khgvlLOJhVhZJbBOwa3ppD6DLOeGxEnEHJ7AA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:54:22.445439Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.18851","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:44b5cc6136379cb5317d19e36b47d32eb89d2dfc9f842523c5a6cfe72a961590","sha256:c1afb691187b0313cdd2733f671c9caec92e2dada1eff50c51d75f70c240c831"],"state_sha256":"d5c9b4fb8b9523fb38f45a5b0a7bedbee2165e0a83c6e6a7415647ca2709ab3c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YYkGjXxI6Ou92iKrGEPdadlnwwVOoP5Dd51wuXE6LzJUCxCtUimvIdUffHdA/QtCZiNJQ/4K2Yw7P5tbTk71Aw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T20:55:32.216648Z","bundle_sha256":"afcf145d0eb7204f56798e48f15f0b62b779a59da0fd3ccfb3f5250cb650f758"}}