{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:4EQCCNQA4K5VG7SUXTGNMYQNJD","short_pith_number":"pith:4EQCCNQA","canonical_record":{"source":{"id":"2405.19444","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-05-29T18:45:55Z","cross_cats_sorted":[],"title_canon_sha256":"a9d63145b133aeb3a827e3aa283cb6a36a3143206c016f1ecc0dfed9c7142694","abstract_canon_sha256":"fa4c3df8d4cb8c9bb63cc6004bed0e0352b5bb5aac2369fd54a01560041f95f0"},"schema_version":"1.0"},"canonical_sha256":"e120213600e2bb537e54bcccd6620d48de3511ec7e3f7195659a18637db0ca8a","source":{"kind":"arxiv","id":"2405.19444","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.19444","created_at":"2026-07-05T08:25:07Z"},{"alias_kind":"arxiv_version","alias_value":"2405.19444v1","created_at":"2026-07-05T08:25:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.19444","created_at":"2026-07-05T08:25:07Z"},{"alias_kind":"pith_short_12","alias_value":"4EQCCNQA4K5V","created_at":"2026-07-05T08:25:07Z"},{"alias_kind":"pith_short_16","alias_value":"4EQCCNQA4K5VG7SU","created_at":"2026-07-05T08:25:07Z"},{"alias_kind":"pith_short_8","alias_value":"4EQCCNQA","created_at":"2026-07-05T08:25:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:4EQCCNQA4K5VG7SUXTGNMYQNJD","target":"record","payload":{"canonical_record":{"source":{"id":"2405.19444","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-05-29T18:45:55Z","cross_cats_sorted":[],"title_canon_sha256":"a9d63145b133aeb3a827e3aa283cb6a36a3143206c016f1ecc0dfed9c7142694","abstract_canon_sha256":"fa4c3df8d4cb8c9bb63cc6004bed0e0352b5bb5aac2369fd54a01560041f95f0"},"schema_version":"1.0"},"canonical_sha256":"e120213600e2bb537e54bcccd6620d48de3511ec7e3f7195659a18637db0ca8a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:25:07.612377Z","signature_b64":"KH4sTO9RHn3sWmkZ8J7rFSdi+gPLLJysCOvIqTNVMvcVY869C1XBa/y3akKoXOivL2H+pnVodGC02i0jE3ctCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e120213600e2bb537e54bcccd6620d48de3511ec7e3f7195659a18637db0ca8a","last_reissued_at":"2026-07-05T08:25:07.611833Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:25:07.611833Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.19444","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":"2026-07-05T08:25:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NZT6tyh4t4jEgxtjC56lKAkQ2nA/BoCE5D/+VtjfAuiF7JeSXIJAd94ikTBywSWYDiaVE0Aa+noBrV8Os+B1Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T11:27:01.578857Z"},"content_sha256":"c63a7cb48d64fd088fee155d202fa10193602e5e36edfffd7154d4d8d008a3d5","schema_version":"1.0","event_id":"sha256:c63a7cb48d64fd088fee155d202fa10193602e5e36edfffd7154d4d8d008a3d5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:4EQCCNQA4K5VG7SUXTGNMYQNJD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MathChat: Benchmarking Mathematical Reasoning and Instruction Following in Multi-Turn Interactions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Dian Yu, Dong Yu, Wenhao Yu, Wenlin Yao, Xiangliang Zhang, Zhenwen Liang, Zhihan Zhang","submitted_at":"2024-05-29T18:45:55Z","abstract_excerpt":"Large language models (LLMs) have demonstrated impressive capabilities in mathematical problem solving, particularly in single turn question answering formats. However, real world scenarios often involve mathematical question answering that requires multi turn or interactive information exchanges, and the performance of LLMs on these tasks is still underexplored. This paper introduces MathChat, a comprehensive benchmark specifically designed to evaluate LLMs across a broader spectrum of mathematical tasks. These tasks are structured to assess the models' abilities in multiturn interactions and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.19444","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/2405.19444/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":"2026-07-05T08:25:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HaQsjV2xAJlETN6LNmPBBhvIRHHKTN2dEZHH5mDePdoH6CZcZYnobH6+3dKZ6HEFqMHt7r5vIRrWbVtDkXFyCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T11:27:01.579379Z"},"content_sha256":"7fcedf0ffd0ab617d4c6b854173209e0edd62ad56bc1927de895536ae6fe3e3a","schema_version":"1.0","event_id":"sha256:7fcedf0ffd0ab617d4c6b854173209e0edd62ad56bc1927de895536ae6fe3e3a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4EQCCNQA4K5VG7SUXTGNMYQNJD/bundle.json","state_url":"https://pith.science/pith/4EQCCNQA4K5VG7SUXTGNMYQNJD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4EQCCNQA4K5VG7SUXTGNMYQNJD/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":"2026-08-09T11:27:01Z","links":{"resolver":"https://pith.science/pith/4EQCCNQA4K5VG7SUXTGNMYQNJD","bundle":"https://pith.science/pith/4EQCCNQA4K5VG7SUXTGNMYQNJD/bundle.json","state":"https://pith.science/pith/4EQCCNQA4K5VG7SUXTGNMYQNJD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4EQCCNQA4K5VG7SUXTGNMYQNJD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:4EQCCNQA4K5VG7SUXTGNMYQNJD","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":"fa4c3df8d4cb8c9bb63cc6004bed0e0352b5bb5aac2369fd54a01560041f95f0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-05-29T18:45:55Z","title_canon_sha256":"a9d63145b133aeb3a827e3aa283cb6a36a3143206c016f1ecc0dfed9c7142694"},"schema_version":"1.0","source":{"id":"2405.19444","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.19444","created_at":"2026-07-05T08:25:07Z"},{"alias_kind":"arxiv_version","alias_value":"2405.19444v1","created_at":"2026-07-05T08:25:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.19444","created_at":"2026-07-05T08:25:07Z"},{"alias_kind":"pith_short_12","alias_value":"4EQCCNQA4K5V","created_at":"2026-07-05T08:25:07Z"},{"alias_kind":"pith_short_16","alias_value":"4EQCCNQA4K5VG7SU","created_at":"2026-07-05T08:25:07Z"},{"alias_kind":"pith_short_8","alias_value":"4EQCCNQA","created_at":"2026-07-05T08:25:07Z"}],"graph_snapshots":[{"event_id":"sha256:7fcedf0ffd0ab617d4c6b854173209e0edd62ad56bc1927de895536ae6fe3e3a","target":"graph","created_at":"2026-07-05T08:25:07Z","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/2405.19444/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have demonstrated impressive capabilities in mathematical problem solving, particularly in single turn question answering formats. However, real world scenarios often involve mathematical question answering that requires multi turn or interactive information exchanges, and the performance of LLMs on these tasks is still underexplored. This paper introduces MathChat, a comprehensive benchmark specifically designed to evaluate LLMs across a broader spectrum of mathematical tasks. These tasks are structured to assess the models' abilities in multiturn interactions and","authors_text":"Dian Yu, Dong Yu, Wenhao Yu, Wenlin Yao, Xiangliang Zhang, Zhenwen Liang, Zhihan Zhang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-05-29T18:45:55Z","title":"MathChat: Benchmarking Mathematical Reasoning and Instruction Following in Multi-Turn Interactions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.19444","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:c63a7cb48d64fd088fee155d202fa10193602e5e36edfffd7154d4d8d008a3d5","target":"record","created_at":"2026-07-05T08:25:07Z","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":"fa4c3df8d4cb8c9bb63cc6004bed0e0352b5bb5aac2369fd54a01560041f95f0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-05-29T18:45:55Z","title_canon_sha256":"a9d63145b133aeb3a827e3aa283cb6a36a3143206c016f1ecc0dfed9c7142694"},"schema_version":"1.0","source":{"id":"2405.19444","kind":"arxiv","version":1}},"canonical_sha256":"e120213600e2bb537e54bcccd6620d48de3511ec7e3f7195659a18637db0ca8a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e120213600e2bb537e54bcccd6620d48de3511ec7e3f7195659a18637db0ca8a","first_computed_at":"2026-07-05T08:25:07.611833Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:25:07.611833Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KH4sTO9RHn3sWmkZ8J7rFSdi+gPLLJysCOvIqTNVMvcVY869C1XBa/y3akKoXOivL2H+pnVodGC02i0jE3ctCg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:25:07.612377Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.19444","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c63a7cb48d64fd088fee155d202fa10193602e5e36edfffd7154d4d8d008a3d5","sha256:7fcedf0ffd0ab617d4c6b854173209e0edd62ad56bc1927de895536ae6fe3e3a"],"state_sha256":"a52498b8c2991de372aa89336733512ec449440fac415ac2aa083b55cdef6026"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fn+vwesheF+N12DTlIiwD8S3IMOKXBNQSNDy+BUAJRLtrNjnNU6pwXtHDtZ2jyf+PWta0dxQeLBZo73ZBbArCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T11:27:01.583717Z","bundle_sha256":"bcd3da4b04c99594e785fa5a31adfb07b90405e7ce352a570c618b1070bb2287"}}