{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:KEHSTKGYKTS24FULIGK7SZ7CPC","short_pith_number":"pith:KEHSTKGY","canonical_record":{"source":{"id":"2509.01395","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-01T11:41:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d194d4f646d71f204c7a56a0bd9b4df8baf0b99cee5eafb690b687f6a60a2d8a","abstract_canon_sha256":"8438abd1f1d164474a3eb76c61ba7a9cc8e5e05f74611b38b9a5f465be483b98"},"schema_version":"1.0"},"canonical_sha256":"510f29a8d854e5ae168b4195f967e27882769b6b11004fc0c220b30a2af8e435","source":{"kind":"arxiv","id":"2509.01395","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.01395","created_at":"2026-07-05T12:02:59Z"},{"alias_kind":"arxiv_version","alias_value":"2509.01395v1","created_at":"2026-07-05T12:02:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.01395","created_at":"2026-07-05T12:02:59Z"},{"alias_kind":"pith_short_12","alias_value":"KEHSTKGYKTS2","created_at":"2026-07-05T12:02:59Z"},{"alias_kind":"pith_short_16","alias_value":"KEHSTKGYKTS24FUL","created_at":"2026-07-05T12:02:59Z"},{"alias_kind":"pith_short_8","alias_value":"KEHSTKGY","created_at":"2026-07-05T12:02:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:KEHSTKGYKTS24FULIGK7SZ7CPC","target":"record","payload":{"canonical_record":{"source":{"id":"2509.01395","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-01T11:41:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d194d4f646d71f204c7a56a0bd9b4df8baf0b99cee5eafb690b687f6a60a2d8a","abstract_canon_sha256":"8438abd1f1d164474a3eb76c61ba7a9cc8e5e05f74611b38b9a5f465be483b98"},"schema_version":"1.0"},"canonical_sha256":"510f29a8d854e5ae168b4195f967e27882769b6b11004fc0c220b30a2af8e435","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:02:59.762858Z","signature_b64":"iHueLs32V4hgqVkJA/pkjAp3/MTn0zUJd3MCHIHeDFf0/cXhlYdBdPL4Hir2SM5pd8iZBVcjRe2Z/F+o6Q44DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"510f29a8d854e5ae168b4195f967e27882769b6b11004fc0c220b30a2af8e435","last_reissued_at":"2026-07-05T12:02:59.762292Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:02:59.762292Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.01395","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":"2026-07-05T12:02:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DTu0kINnbISXPUvM3w+H2cxzl1Kjql4kUr0v97SYS7wDc9EVfxxSrkgR6Dx1nOAZclms6TMks8SNlUn0gnCRDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T07:11:27.994570Z"},"content_sha256":"aa0d96807d6cf2310c6960aaa9d31d5f9f33d4031a25fa65f81cee747456ee24","schema_version":"1.0","event_id":"sha256:aa0d96807d6cf2310c6960aaa9d31d5f9f33d4031a25fa65f81cee747456ee24"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:KEHSTKGYKTS24FULIGK7SZ7CPC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LLMs cannot spot math errors, even when allowed to peek into the solution","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ekaterina Kochmar, Kaushal Kumar Maurya, KV Aditya Srivatsa","submitted_at":"2025-09-01T11:41:10Z","abstract_excerpt":"Large language models (LLMs) demonstrate remarkable performance on math word problems, yet they have been shown to struggle with meta-reasoning tasks such as identifying errors in student solutions. In this work, we investigate the challenge of locating the first error step in stepwise solutions using two error reasoning datasets: VtG and PRM800K. Our experiments show that state-of-the-art LLMs struggle to locate the first error step in student solutions even when given access to the reference solution. To that end, we propose an approach that generates an intermediate corrected student soluti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.01395","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/2509.01395/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":"2026-07-05T12:02:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sRLFZylnnUZAyXSgnUtU26re5RpYHNUasILDgK732eeO5YuwnEdlvCwM/pVtKFslo7Y/g5bdB8M3rubPkV4zDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T07:11:27.994975Z"},"content_sha256":"06b94f5595d4a057eb6c985f360d45f4c7d325c60a2a5ed3b4608cc47fbd38c3","schema_version":"1.0","event_id":"sha256:06b94f5595d4a057eb6c985f360d45f4c7d325c60a2a5ed3b4608cc47fbd38c3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KEHSTKGYKTS24FULIGK7SZ7CPC/bundle.json","state_url":"https://pith.science/pith/KEHSTKGYKTS24FULIGK7SZ7CPC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KEHSTKGYKTS24FULIGK7SZ7CPC/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":"2026-08-06T07:11:27Z","links":{"resolver":"https://pith.science/pith/KEHSTKGYKTS24FULIGK7SZ7CPC","bundle":"https://pith.science/pith/KEHSTKGYKTS24FULIGK7SZ7CPC/bundle.json","state":"https://pith.science/pith/KEHSTKGYKTS24FULIGK7SZ7CPC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KEHSTKGYKTS24FULIGK7SZ7CPC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:KEHSTKGYKTS24FULIGK7SZ7CPC","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":"8438abd1f1d164474a3eb76c61ba7a9cc8e5e05f74611b38b9a5f465be483b98","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-01T11:41:10Z","title_canon_sha256":"d194d4f646d71f204c7a56a0bd9b4df8baf0b99cee5eafb690b687f6a60a2d8a"},"schema_version":"1.0","source":{"id":"2509.01395","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.01395","created_at":"2026-07-05T12:02:59Z"},{"alias_kind":"arxiv_version","alias_value":"2509.01395v1","created_at":"2026-07-05T12:02:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.01395","created_at":"2026-07-05T12:02:59Z"},{"alias_kind":"pith_short_12","alias_value":"KEHSTKGYKTS2","created_at":"2026-07-05T12:02:59Z"},{"alias_kind":"pith_short_16","alias_value":"KEHSTKGYKTS24FUL","created_at":"2026-07-05T12:02:59Z"},{"alias_kind":"pith_short_8","alias_value":"KEHSTKGY","created_at":"2026-07-05T12:02:59Z"}],"graph_snapshots":[{"event_id":"sha256:06b94f5595d4a057eb6c985f360d45f4c7d325c60a2a5ed3b4608cc47fbd38c3","target":"graph","created_at":"2026-07-05T12:02:59Z","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/2509.01395/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) demonstrate remarkable performance on math word problems, yet they have been shown to struggle with meta-reasoning tasks such as identifying errors in student solutions. In this work, we investigate the challenge of locating the first error step in stepwise solutions using two error reasoning datasets: VtG and PRM800K. Our experiments show that state-of-the-art LLMs struggle to locate the first error step in student solutions even when given access to the reference solution. To that end, we propose an approach that generates an intermediate corrected student soluti","authors_text":"Ekaterina Kochmar, Kaushal Kumar Maurya, KV Aditya Srivatsa","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-01T11:41:10Z","title":"LLMs cannot spot math errors, even when allowed to peek into the solution"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.01395","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:aa0d96807d6cf2310c6960aaa9d31d5f9f33d4031a25fa65f81cee747456ee24","target":"record","created_at":"2026-07-05T12:02:59Z","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":"8438abd1f1d164474a3eb76c61ba7a9cc8e5e05f74611b38b9a5f465be483b98","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-01T11:41:10Z","title_canon_sha256":"d194d4f646d71f204c7a56a0bd9b4df8baf0b99cee5eafb690b687f6a60a2d8a"},"schema_version":"1.0","source":{"id":"2509.01395","kind":"arxiv","version":1}},"canonical_sha256":"510f29a8d854e5ae168b4195f967e27882769b6b11004fc0c220b30a2af8e435","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"510f29a8d854e5ae168b4195f967e27882769b6b11004fc0c220b30a2af8e435","first_computed_at":"2026-07-05T12:02:59.762292Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:02:59.762292Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iHueLs32V4hgqVkJA/pkjAp3/MTn0zUJd3MCHIHeDFf0/cXhlYdBdPL4Hir2SM5pd8iZBVcjRe2Z/F+o6Q44DA==","signature_status":"signed_v1","signed_at":"2026-07-05T12:02:59.762858Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.01395","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:aa0d96807d6cf2310c6960aaa9d31d5f9f33d4031a25fa65f81cee747456ee24","sha256:06b94f5595d4a057eb6c985f360d45f4c7d325c60a2a5ed3b4608cc47fbd38c3"],"state_sha256":"c14287aed8df13f614091c9165c4a5e7c44e8943b8a5a10bf4e23478eb66e294"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yb7AOIvO74LAup6G884lE40bD7yO4rrQHsJVstfUHyH4FlU6lJuqzzCE9fTuIze14Zp8ZVK36QZ87Ly7SouaAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T07:11:27.999399Z","bundle_sha256":"97a8b57a31e58175448ae4cdfbe9918b7090e315a85a3883b15083b85286c5c7"}}