{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:4ZZE2XKSSORDHON62Y4IR5D7DV","short_pith_number":"pith:4ZZE2XKS","canonical_record":{"source":{"id":"2509.06332","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-08T04:31:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"419922923695f11f64e2b3fcf2ea2c9c1f72a39a0dcfe8b5e41aa9061ebc8502","abstract_canon_sha256":"d6632e965f18bfca3fc21fce7d44f5f33ed97420ba95cbc78bb4dd90a0c22a86"},"schema_version":"1.0"},"canonical_sha256":"e6724d5d5293a233b9bed63888f47f1d6893494286311b877062574f9344ebd2","source":{"kind":"arxiv","id":"2509.06332","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.06332","created_at":"2026-07-05T12:06:33Z"},{"alias_kind":"arxiv_version","alias_value":"2509.06332v1","created_at":"2026-07-05T12:06:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.06332","created_at":"2026-07-05T12:06:33Z"},{"alias_kind":"pith_short_12","alias_value":"4ZZE2XKSSORD","created_at":"2026-07-05T12:06:33Z"},{"alias_kind":"pith_short_16","alias_value":"4ZZE2XKSSORDHON6","created_at":"2026-07-05T12:06:33Z"},{"alias_kind":"pith_short_8","alias_value":"4ZZE2XKS","created_at":"2026-07-05T12:06:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:4ZZE2XKSSORDHON62Y4IR5D7DV","target":"record","payload":{"canonical_record":{"source":{"id":"2509.06332","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-08T04:31:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"419922923695f11f64e2b3fcf2ea2c9c1f72a39a0dcfe8b5e41aa9061ebc8502","abstract_canon_sha256":"d6632e965f18bfca3fc21fce7d44f5f33ed97420ba95cbc78bb4dd90a0c22a86"},"schema_version":"1.0"},"canonical_sha256":"e6724d5d5293a233b9bed63888f47f1d6893494286311b877062574f9344ebd2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:06:33.928061Z","signature_b64":"bNTo7BPh5XajhooGOpxB0G9DYB4EP6W+jINcUCp1WNhoffCKTNqcXxmnkT9si/JB7/A9WB5P7aNd9rSc1iJ6BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e6724d5d5293a233b9bed63888f47f1d6893494286311b877062574f9344ebd2","last_reissued_at":"2026-07-05T12:06:33.927491Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:06:33.927491Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.06332","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":"2026-07-05T12:06:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Q38xw0bq0s3AVsarxVGSzJowaytbXNIi9+ys9Z3FSCu5h5c+NuQxr8gexerrScRFsKnthe6gAiNYKRcJVrbiAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T19:35:44.541098Z"},"content_sha256":"1de01fc844638968c94aaa93a5ad9d2885afa8a1926d0405e07985de7070b509","schema_version":"1.0","event_id":"sha256:1de01fc844638968c94aaa93a5ad9d2885afa8a1926d0405e07985de7070b509"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:4ZZE2XKSSORDHON62Y4IR5D7DV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Fragile Number Sense: Probing the Elemental Limits of Numerical Reasoning in LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Aashwin Ananda Mishra, Roussel Rahman","submitted_at":"2025-09-08T04:31:12Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated remarkable emergent capabilities, yet the robustness of their numerical reasoning remains an open question. While standard benchmarks evaluate LLM reasoning on complex problem sets using aggregated metrics, they often obscure foundational weaknesses. In this work, we probe LLM mathematical numeracy by evaluating performance on problems of escalating complexity, from constituent operations to combinatorial puzzles. We test several state-of-the-art LLM-based agents on a 100-problem challenge comprising four categories: (1) basic arithmetic, (2) adva"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.06332","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/2509.06332/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":"2026-07-05T12:06:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AjVSefBYDtHYbAXZARNn7qmkMDR2NKJE5GO5vd2KQ9WvXJuzGaq4gSkK02OKiVrXtrPxJgFecePkltVa+mgIDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T19:35:44.541483Z"},"content_sha256":"88a98d9e703553e6350707483d2c30a1bcf6f807252c13ccee2929c1e8f6f664","schema_version":"1.0","event_id":"sha256:88a98d9e703553e6350707483d2c30a1bcf6f807252c13ccee2929c1e8f6f664"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4ZZE2XKSSORDHON62Y4IR5D7DV/bundle.json","state_url":"https://pith.science/pith/4ZZE2XKSSORDHON62Y4IR5D7DV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4ZZE2XKSSORDHON62Y4IR5D7DV/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":"2026-08-03T19:35:44Z","links":{"resolver":"https://pith.science/pith/4ZZE2XKSSORDHON62Y4IR5D7DV","bundle":"https://pith.science/pith/4ZZE2XKSSORDHON62Y4IR5D7DV/bundle.json","state":"https://pith.science/pith/4ZZE2XKSSORDHON62Y4IR5D7DV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4ZZE2XKSSORDHON62Y4IR5D7DV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4ZZE2XKSSORDHON62Y4IR5D7DV","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":"d6632e965f18bfca3fc21fce7d44f5f33ed97420ba95cbc78bb4dd90a0c22a86","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-08T04:31:12Z","title_canon_sha256":"419922923695f11f64e2b3fcf2ea2c9c1f72a39a0dcfe8b5e41aa9061ebc8502"},"schema_version":"1.0","source":{"id":"2509.06332","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.06332","created_at":"2026-07-05T12:06:33Z"},{"alias_kind":"arxiv_version","alias_value":"2509.06332v1","created_at":"2026-07-05T12:06:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.06332","created_at":"2026-07-05T12:06:33Z"},{"alias_kind":"pith_short_12","alias_value":"4ZZE2XKSSORD","created_at":"2026-07-05T12:06:33Z"},{"alias_kind":"pith_short_16","alias_value":"4ZZE2XKSSORDHON6","created_at":"2026-07-05T12:06:33Z"},{"alias_kind":"pith_short_8","alias_value":"4ZZE2XKS","created_at":"2026-07-05T12:06:33Z"}],"graph_snapshots":[{"event_id":"sha256:88a98d9e703553e6350707483d2c30a1bcf6f807252c13ccee2929c1e8f6f664","target":"graph","created_at":"2026-07-05T12:06:33Z","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/2509.06332/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have demonstrated remarkable emergent capabilities, yet the robustness of their numerical reasoning remains an open question. While standard benchmarks evaluate LLM reasoning on complex problem sets using aggregated metrics, they often obscure foundational weaknesses. In this work, we probe LLM mathematical numeracy by evaluating performance on problems of escalating complexity, from constituent operations to combinatorial puzzles. We test several state-of-the-art LLM-based agents on a 100-problem challenge comprising four categories: (1) basic arithmetic, (2) adva","authors_text":"Aashwin Ananda Mishra, Roussel Rahman","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-08T04:31:12Z","title":"A Fragile Number Sense: Probing the Elemental Limits of Numerical Reasoning in LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.06332","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":"sha256:1de01fc844638968c94aaa93a5ad9d2885afa8a1926d0405e07985de7070b509","target":"record","created_at":"2026-07-05T12:06:33Z","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":"d6632e965f18bfca3fc21fce7d44f5f33ed97420ba95cbc78bb4dd90a0c22a86","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-08T04:31:12Z","title_canon_sha256":"419922923695f11f64e2b3fcf2ea2c9c1f72a39a0dcfe8b5e41aa9061ebc8502"},"schema_version":"1.0","source":{"id":"2509.06332","kind":"arxiv","version":1}},"canonical_sha256":"e6724d5d5293a233b9bed63888f47f1d6893494286311b877062574f9344ebd2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e6724d5d5293a233b9bed63888f47f1d6893494286311b877062574f9344ebd2","first_computed_at":"2026-07-05T12:06:33.927491Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:06:33.927491Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bNTo7BPh5XajhooGOpxB0G9DYB4EP6W+jINcUCp1WNhoffCKTNqcXxmnkT9si/JB7/A9WB5P7aNd9rSc1iJ6BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T12:06:33.928061Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.06332","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1de01fc844638968c94aaa93a5ad9d2885afa8a1926d0405e07985de7070b509","sha256:88a98d9e703553e6350707483d2c30a1bcf6f807252c13ccee2929c1e8f6f664"],"state_sha256":"4e63085ff96dff0571aa28c8aba24834f137fc8266a3b2f1655ff9695dadaff9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zbUgRLNVJPbHI83j6Bd4BSVlPaOkZmEZx1WYrXGz7C/mGNMLAocXt04+OA+YysU/W2nt9T7/x5QyoCaD2bLtCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T19:35:44.544093Z","bundle_sha256":"13ab84b1f9f57948d8f91c85f96a82ff6455425da15d4008808656956f480438"}}