{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:2TUQ55VU5H64N3LEITLD7WTD3X","short_pith_number":"pith:2TUQ55VU","canonical_record":{"source":{"id":"2405.14804","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-23T17:13:50Z","cross_cats_sorted":[],"title_canon_sha256":"ec6cc4561005529ed2621adbe862e902bdfcc7d6a6a1d2816ff8ec84f107ab25","abstract_canon_sha256":"522da650a735ff5e1a546dc082975a44cecef6aeb369f14af66b7ca1ffc917d4"},"schema_version":"1.0"},"canonical_sha256":"d4e90ef6b4e9fdc6ed6444d63fda63ddc8e63a72bf2d76a2a50b7661d67abaca","source":{"kind":"arxiv","id":"2405.14804","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.14804","created_at":"2026-07-05T10:19:55Z"},{"alias_kind":"arxiv_version","alias_value":"2405.14804v4","created_at":"2026-07-05T10:19:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14804","created_at":"2026-07-05T10:19:55Z"},{"alias_kind":"pith_short_12","alias_value":"2TUQ55VU5H64","created_at":"2026-07-05T10:19:55Z"},{"alias_kind":"pith_short_16","alias_value":"2TUQ55VU5H64N3LE","created_at":"2026-07-05T10:19:55Z"},{"alias_kind":"pith_short_8","alias_value":"2TUQ55VU","created_at":"2026-07-05T10:19:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:2TUQ55VU5H64N3LEITLD7WTD3X","target":"record","payload":{"canonical_record":{"source":{"id":"2405.14804","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-23T17:13:50Z","cross_cats_sorted":[],"title_canon_sha256":"ec6cc4561005529ed2621adbe862e902bdfcc7d6a6a1d2816ff8ec84f107ab25","abstract_canon_sha256":"522da650a735ff5e1a546dc082975a44cecef6aeb369f14af66b7ca1ffc917d4"},"schema_version":"1.0"},"canonical_sha256":"d4e90ef6b4e9fdc6ed6444d63fda63ddc8e63a72bf2d76a2a50b7661d67abaca","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:19:55.239711Z","signature_b64":"+/obBvRLHEEvlP1PFwOls2Wq8ZbQpE8MkhbYLzYfsBxK7lqdQO9vTXynaZC56has0DiLJvzJbRSWhQCBXAJ+CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d4e90ef6b4e9fdc6ed6444d63fda63ddc8e63a72bf2d76a2a50b7661d67abaca","last_reissued_at":"2026-07-05T10:19:55.239350Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:19:55.239350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.14804","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:19:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DVGmm8IAjMI4g2hR4ei3RTpzNnGuuj1a1aTHQsL3EogNAyewD1ZD26zOFWzyeBkmVMrCob2aSfp4GJLB8qbEBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T21:02:45.209496Z"},"content_sha256":"bd618efb2d9ef59587c1ab9b5edd5bc40098cbc47fadb9ee3fe0e747b1a5b03c","schema_version":"1.0","event_id":"sha256:bd618efb2d9ef59587c1ab9b5edd5bc40098cbc47fadb9ee3fe0e747b1a5b03c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:2TUQ55VU5H64N3LEITLD7WTD3X","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Can LLMs Solve longer Math Word Problems Better?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Can Yang, Tong Xiao, Xin Xu, Yang Wang, Zhenya Huang, Zitong Chao","submitted_at":"2024-05-23T17:13:50Z","abstract_excerpt":"Math Word Problems (MWPs) play a vital role in assessing the capabilities of Large Language Models (LLMs), yet current research primarily focuses on questions with concise contexts. The impact of longer contexts on mathematical reasoning remains under-explored. This study pioneers the investigation of Context Length Generalizability (CoLeG), which refers to the ability of LLMs to solve MWPs with extended narratives. We introduce Extended Grade-School Math (E-GSM), a collection of MWPs featuring lengthy narratives, and propose two novel metrics to evaluate the efficacy and resilience of LLMs in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14804","kind":"arxiv","version":4},"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/2405.14804/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":"2026-07-05T10:19:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"k6KbnzAdyX9zkN4XL3TQNiM9SvqqvA6KYuePV/tip18Cvs5jkVW52m7Gi35yfsiw6sRSj36yBnIbYRJHPr8/Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T21:02:45.210151Z"},"content_sha256":"b3f262faa9afdba8e1dc081f341d3039bac105f1a7c10a6d28d89bb3d18462b2","schema_version":"1.0","event_id":"sha256:b3f262faa9afdba8e1dc081f341d3039bac105f1a7c10a6d28d89bb3d18462b2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2TUQ55VU5H64N3LEITLD7WTD3X/bundle.json","state_url":"https://pith.science/pith/2TUQ55VU5H64N3LEITLD7WTD3X/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2TUQ55VU5H64N3LEITLD7WTD3X/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":"2026-08-06T21:02:45Z","links":{"resolver":"https://pith.science/pith/2TUQ55VU5H64N3LEITLD7WTD3X","bundle":"https://pith.science/pith/2TUQ55VU5H64N3LEITLD7WTD3X/bundle.json","state":"https://pith.science/pith/2TUQ55VU5H64N3LEITLD7WTD3X/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2TUQ55VU5H64N3LEITLD7WTD3X/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:2TUQ55VU5H64N3LEITLD7WTD3X","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":"522da650a735ff5e1a546dc082975a44cecef6aeb369f14af66b7ca1ffc917d4","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-23T17:13:50Z","title_canon_sha256":"ec6cc4561005529ed2621adbe862e902bdfcc7d6a6a1d2816ff8ec84f107ab25"},"schema_version":"1.0","source":{"id":"2405.14804","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.14804","created_at":"2026-07-05T10:19:55Z"},{"alias_kind":"arxiv_version","alias_value":"2405.14804v4","created_at":"2026-07-05T10:19:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14804","created_at":"2026-07-05T10:19:55Z"},{"alias_kind":"pith_short_12","alias_value":"2TUQ55VU5H64","created_at":"2026-07-05T10:19:55Z"},{"alias_kind":"pith_short_16","alias_value":"2TUQ55VU5H64N3LE","created_at":"2026-07-05T10:19:55Z"},{"alias_kind":"pith_short_8","alias_value":"2TUQ55VU","created_at":"2026-07-05T10:19:55Z"}],"graph_snapshots":[{"event_id":"sha256:b3f262faa9afdba8e1dc081f341d3039bac105f1a7c10a6d28d89bb3d18462b2","target":"graph","created_at":"2026-07-05T10:19:55Z","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/2405.14804/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Math Word Problems (MWPs) play a vital role in assessing the capabilities of Large Language Models (LLMs), yet current research primarily focuses on questions with concise contexts. The impact of longer contexts on mathematical reasoning remains under-explored. This study pioneers the investigation of Context Length Generalizability (CoLeG), which refers to the ability of LLMs to solve MWPs with extended narratives. We introduce Extended Grade-School Math (E-GSM), a collection of MWPs featuring lengthy narratives, and propose two novel metrics to evaluate the efficacy and resilience of LLMs in","authors_text":"Can Yang, Tong Xiao, Xin Xu, Yang Wang, Zhenya Huang, Zitong Chao","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-23T17:13:50Z","title":"Can LLMs Solve longer Math Word Problems Better?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14804","kind":"arxiv","version":4},"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:bd618efb2d9ef59587c1ab9b5edd5bc40098cbc47fadb9ee3fe0e747b1a5b03c","target":"record","created_at":"2026-07-05T10:19:55Z","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":"522da650a735ff5e1a546dc082975a44cecef6aeb369f14af66b7ca1ffc917d4","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-23T17:13:50Z","title_canon_sha256":"ec6cc4561005529ed2621adbe862e902bdfcc7d6a6a1d2816ff8ec84f107ab25"},"schema_version":"1.0","source":{"id":"2405.14804","kind":"arxiv","version":4}},"canonical_sha256":"d4e90ef6b4e9fdc6ed6444d63fda63ddc8e63a72bf2d76a2a50b7661d67abaca","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d4e90ef6b4e9fdc6ed6444d63fda63ddc8e63a72bf2d76a2a50b7661d67abaca","first_computed_at":"2026-07-05T10:19:55.239350Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:19:55.239350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+/obBvRLHEEvlP1PFwOls2Wq8ZbQpE8MkhbYLzYfsBxK7lqdQO9vTXynaZC56has0DiLJvzJbRSWhQCBXAJ+CA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:19:55.239711Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.14804","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bd618efb2d9ef59587c1ab9b5edd5bc40098cbc47fadb9ee3fe0e747b1a5b03c","sha256:b3f262faa9afdba8e1dc081f341d3039bac105f1a7c10a6d28d89bb3d18462b2"],"state_sha256":"f084cf8f5c4dde96184173bf21c42e4c8e9ee5f9dcc240f121a7c45281a2b65d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hNhCP7S/krL6aOHilYgFDSq8mub9aFep0fFnSjpM4fegLQ/ySS4JEQ4xkPUTE6v9w6YvoyR6LMAbdWrvDHlBBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T21:02:45.215681Z","bundle_sha256":"61a4eff5f1ce152663db34e6ea71ae4464dcb805141ac7fc98122c9607fd7bdc"}}