{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:SDOMHYLKDXMD3TPJVCP4H3ZUEZ","short_pith_number":"pith:SDOMHYLK","canonical_record":{"source":{"id":"2507.18973","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-25T05:57:47Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"e95514ed269cafc879a1528f1a4766f9bf2fef48fd15232983654665542abf90","abstract_canon_sha256":"da4df962d640349c4be6fe0f3f1a985a88af700ecf8a28d7b6f0d087ad242856"},"schema_version":"1.0"},"canonical_sha256":"90dcc3e16a1dd83dcde9a89fc3ef34264a7edf1fa3b971344f816751b003e9d4","source":{"kind":"arxiv","id":"2507.18973","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.18973","created_at":"2026-07-05T11:57:38Z"},{"alias_kind":"arxiv_version","alias_value":"2507.18973v2","created_at":"2026-07-05T11:57:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.18973","created_at":"2026-07-05T11:57:38Z"},{"alias_kind":"pith_short_12","alias_value":"SDOMHYLKDXMD","created_at":"2026-07-05T11:57:38Z"},{"alias_kind":"pith_short_16","alias_value":"SDOMHYLKDXMD3TPJ","created_at":"2026-07-05T11:57:38Z"},{"alias_kind":"pith_short_8","alias_value":"SDOMHYLK","created_at":"2026-07-05T11:57:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:SDOMHYLKDXMD3TPJVCP4H3ZUEZ","target":"record","payload":{"canonical_record":{"source":{"id":"2507.18973","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-25T05:57:47Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"e95514ed269cafc879a1528f1a4766f9bf2fef48fd15232983654665542abf90","abstract_canon_sha256":"da4df962d640349c4be6fe0f3f1a985a88af700ecf8a28d7b6f0d087ad242856"},"schema_version":"1.0"},"canonical_sha256":"90dcc3e16a1dd83dcde9a89fc3ef34264a7edf1fa3b971344f816751b003e9d4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:57:38.139853Z","signature_b64":"uQLGs8XPPp0wXZ+a/wEAVo1g7PPR7nFlewai4XpSFsQ7Rev9FF/Jxv9ReC1XfjNAneM60srBQ8EcyZUvf4UhCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"90dcc3e16a1dd83dcde9a89fc3ef34264a7edf1fa3b971344f816751b003e9d4","last_reissued_at":"2026-07-05T11:57:38.139163Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:57:38.139163Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.18973","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":"2026-07-05T11:57:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"966LiZCYIBzQoRWFK7ZDlRggQNvOrWqRhTrx4TNzABs3m8dixJ2TVwLM1Csab9KTosKLwYWQLkxrqiqeiii8AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T23:53:19.364093Z"},"content_sha256":"d4f8e51cc973a9ec55fe5fb0e01d5352120e9c89a3540bf39e3ea10fd78daffd","schema_version":"1.0","event_id":"sha256:d4f8e51cc973a9ec55fe5fb0e01d5352120e9c89a3540bf39e3ea10fd78daffd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:SDOMHYLKDXMD3TPJVCP4H3ZUEZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Toolbox, Not a Hammer -- Multi-TAG: Scaling Math Reasoning with Multi-Tool Aggregation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Bohan Yao, Vikas Yadav","submitted_at":"2025-07-25T05:57:47Z","abstract_excerpt":"Augmenting large language models (LLMs) with external tools is a promising avenue for developing high-performance mathematical reasoning systems. Prior tool-augmented approaches typically finetune an LLM to select and invoke a single tool at each reasoning step and show promising results on simpler math reasoning benchmarks such as GSM8K. However, these approaches struggle with more complex math problems that require precise reasoning over multiple steps. To address this limitation, in this work, we propose Multi-TAG, a Multi-Tool AGgregation-based framework. Instead of relying on a single too"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.18973","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/2507.18973/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":"2026-07-05T11:57:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8Z+UcqhMMEt5S4JT0a4bZ3OISgMMPnvrkIRgktvv/8vPnyX6b0Buivw9ywzDYpN8F7hQr6klrPseU20vXenMDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T23:53:19.366671Z"},"content_sha256":"f0654c71afaa202b51a0ccb0b544c27536900604767fff74e9e57fe85859f521","schema_version":"1.0","event_id":"sha256:f0654c71afaa202b51a0ccb0b544c27536900604767fff74e9e57fe85859f521"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SDOMHYLKDXMD3TPJVCP4H3ZUEZ/bundle.json","state_url":"https://pith.science/pith/SDOMHYLKDXMD3TPJVCP4H3ZUEZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SDOMHYLKDXMD3TPJVCP4H3ZUEZ/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":"2026-08-14T23:53:19Z","links":{"resolver":"https://pith.science/pith/SDOMHYLKDXMD3TPJVCP4H3ZUEZ","bundle":"https://pith.science/pith/SDOMHYLKDXMD3TPJVCP4H3ZUEZ/bundle.json","state":"https://pith.science/pith/SDOMHYLKDXMD3TPJVCP4H3ZUEZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SDOMHYLKDXMD3TPJVCP4H3ZUEZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SDOMHYLKDXMD3TPJVCP4H3ZUEZ","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":"da4df962d640349c4be6fe0f3f1a985a88af700ecf8a28d7b6f0d087ad242856","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-25T05:57:47Z","title_canon_sha256":"e95514ed269cafc879a1528f1a4766f9bf2fef48fd15232983654665542abf90"},"schema_version":"1.0","source":{"id":"2507.18973","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.18973","created_at":"2026-07-05T11:57:38Z"},{"alias_kind":"arxiv_version","alias_value":"2507.18973v2","created_at":"2026-07-05T11:57:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.18973","created_at":"2026-07-05T11:57:38Z"},{"alias_kind":"pith_short_12","alias_value":"SDOMHYLKDXMD","created_at":"2026-07-05T11:57:38Z"},{"alias_kind":"pith_short_16","alias_value":"SDOMHYLKDXMD3TPJ","created_at":"2026-07-05T11:57:38Z"},{"alias_kind":"pith_short_8","alias_value":"SDOMHYLK","created_at":"2026-07-05T11:57:38Z"}],"graph_snapshots":[{"event_id":"sha256:f0654c71afaa202b51a0ccb0b544c27536900604767fff74e9e57fe85859f521","target":"graph","created_at":"2026-07-05T11:57:38Z","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/2507.18973/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Augmenting large language models (LLMs) with external tools is a promising avenue for developing high-performance mathematical reasoning systems. Prior tool-augmented approaches typically finetune an LLM to select and invoke a single tool at each reasoning step and show promising results on simpler math reasoning benchmarks such as GSM8K. However, these approaches struggle with more complex math problems that require precise reasoning over multiple steps. To address this limitation, in this work, we propose Multi-TAG, a Multi-Tool AGgregation-based framework. Instead of relying on a single too","authors_text":"Bohan Yao, Vikas Yadav","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-25T05:57:47Z","title":"A Toolbox, Not a Hammer -- Multi-TAG: Scaling Math Reasoning with Multi-Tool Aggregation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.18973","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:d4f8e51cc973a9ec55fe5fb0e01d5352120e9c89a3540bf39e3ea10fd78daffd","target":"record","created_at":"2026-07-05T11:57:38Z","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":"da4df962d640349c4be6fe0f3f1a985a88af700ecf8a28d7b6f0d087ad242856","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-25T05:57:47Z","title_canon_sha256":"e95514ed269cafc879a1528f1a4766f9bf2fef48fd15232983654665542abf90"},"schema_version":"1.0","source":{"id":"2507.18973","kind":"arxiv","version":2}},"canonical_sha256":"90dcc3e16a1dd83dcde9a89fc3ef34264a7edf1fa3b971344f816751b003e9d4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"90dcc3e16a1dd83dcde9a89fc3ef34264a7edf1fa3b971344f816751b003e9d4","first_computed_at":"2026-07-05T11:57:38.139163Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:57:38.139163Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uQLGs8XPPp0wXZ+a/wEAVo1g7PPR7nFlewai4XpSFsQ7Rev9FF/Jxv9ReC1XfjNAneM60srBQ8EcyZUvf4UhCg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:57:38.139853Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.18973","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d4f8e51cc973a9ec55fe5fb0e01d5352120e9c89a3540bf39e3ea10fd78daffd","sha256:f0654c71afaa202b51a0ccb0b544c27536900604767fff74e9e57fe85859f521"],"state_sha256":"753ef4747d9f1a44477ca5e3644d95b5c7db6a5240555eabb9bab637e9287746"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ymv8YlJcD3SAuur+Eap8EA82cD1MwvKlQvtDT0O0Y5PiJGCNffHK4LkJvwiOEVrnDXzVzQyQRUnoeYy6FWe1Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T23:53:19.378020Z","bundle_sha256":"867845d4564b78395cae7fc6a029b77f645d2aa6faab6ce8685223cd4bdd5c41"}}