{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:KNULEB4U7CAYPWYCV6FBFRU3JT","short_pith_number":"pith:KNULEB4U","canonical_record":{"source":{"id":"2608.08447","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-09T03:30:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"801622d8e0545d17d1b96d7e836e52d2c9628dfd25449b19ab6bbba99890fc84","abstract_canon_sha256":"d2d55f8ed5c6ae792cb0a6004facf859fc549cacd96e9619bbb907e76e9e2526"},"schema_version":"1.0"},"canonical_sha256":"5368b20794f88187db02af8a12c69b4cf1eba86f96e8e69e5a4054d7150792c2","source":{"kind":"arxiv","id":"2608.08447","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.08447","created_at":"2026-08-11T01:22:39Z"},{"alias_kind":"arxiv_version","alias_value":"2608.08447v1","created_at":"2026-08-11T01:22:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.08447","created_at":"2026-08-11T01:22:39Z"},{"alias_kind":"pith_short_12","alias_value":"KNULEB4U7CAY","created_at":"2026-08-11T01:22:39Z"},{"alias_kind":"pith_short_16","alias_value":"KNULEB4U7CAYPWYC","created_at":"2026-08-11T01:22:39Z"},{"alias_kind":"pith_short_8","alias_value":"KNULEB4U","created_at":"2026-08-11T01:22:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:KNULEB4U7CAYPWYCV6FBFRU3JT","target":"record","payload":{"canonical_record":{"source":{"id":"2608.08447","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-09T03:30:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"801622d8e0545d17d1b96d7e836e52d2c9628dfd25449b19ab6bbba99890fc84","abstract_canon_sha256":"d2d55f8ed5c6ae792cb0a6004facf859fc549cacd96e9619bbb907e76e9e2526"},"schema_version":"1.0"},"canonical_sha256":"5368b20794f88187db02af8a12c69b4cf1eba86f96e8e69e5a4054d7150792c2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T01:22:39.456619Z","signature_b64":"1wVnuL52pbZzgwXxBksUu8Jjvae+0N3BuJtEMz4keCRv5D0GNozdoLTbqSMeB1BSkrhJoczpsnsm85HWI7IkDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5368b20794f88187db02af8a12c69b4cf1eba86f96e8e69e5a4054d7150792c2","last_reissued_at":"2026-08-11T01:22:39.454238Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T01:22:39.454238Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.08447","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":"2026-08-11T01:22:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lpNCoeYGdkFDuRKkeO2AbPfmgm+Lu8bmSAmCvSc6nRDyfXqJd6orn7aG63imeD/+umy82yjLzd/D3UTwDftxDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T17:19:00.005849Z"},"content_sha256":"7b49028a14879fe5ce69d390275f19cad44174733d728828972680ccd7dd45d1","schema_version":"1.0","event_id":"sha256:7b49028a14879fe5ce69d390275f19cad44174733d728828972680ccd7dd45d1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:KNULEB4U7CAYPWYCV6FBFRU3JT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Hidden Language Consistency Phenomena in Reasoning LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Kelly Marchisio, Muhammad Ali Shafique","submitted_at":"2026-08-09T03:30:14Z","abstract_excerpt":"Multilingual reasoning models are commonly evaluated by whether they arrive at the correct answer, but not by whether they preserve the intended language while reasoning and responding. This omission conceals important multilingual behaviors that emerge as tasks become harder. In this paper, we study task difficulty, task accuracy, thinking-language consistency (TC), and answer-language consistency (AC) across reasoning models using PolyMath benchmark in eight languages and four difficulty levels. We uncover four findings: (1) language consistency exhibits four difficulty-dependent behaviors: "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.08447","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/2608.08447/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":"2026-08-11T01:22:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+GIlRTXnD62RxLeSe3i2q+2yp9Xma2+sYHXFL+GL2Hnoqih8veknrUtyr1jlTdxVaiYUN8uHVU0EkwZczbUAAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T17:19:00.006357Z"},"content_sha256":"4fa3084675fd59a6d045dbf34c9d95337d68d0a569ea4d587ab168aa3d2d6653","schema_version":"1.0","event_id":"sha256:4fa3084675fd59a6d045dbf34c9d95337d68d0a569ea4d587ab168aa3d2d6653"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:KNULEB4U7CAYPWYCV6FBFRU3JT","target":"integrity","payload":{"note":"DOI is split by whitespace or line breaks in the printed bibliography. Reconstructed DOI 10.1162/tacl_a_00276 resolves to 'Natural Questions: A Benchmark for Question Answering Research'. A reader following the printed text alone cannot reach it.","snippet":"Kwiatkowski, Tom and Palomaki, Jennimaria and Redfield, Olivia and Collins, Michael and Parikh, Ankur and Alberti, Chris and Epstein, Danielle and Polosukhin, Illia and Devlin, Jacob and Lee, Kenton and Toutanova, Kristina and Jones, Llion ","arxiv_id":"2608.08447","detector":"doi_compliance","evidence":{"ref_index":125,"verdict_class":"incontrovertible","resolved_title":"Natural Questions: A Benchmark for Question Answering Research","printed_excerpt":"10.1162/tacl","reconstructed_doi":"10.1162/tacl_a_00276"},"severity":"advisory","ref_index":125,"audited_at":"2026-08-14T05:29:06.640615Z","event_type":"pith.integrity.v1","detected_doi":"10.1162/tacl_a_00276","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"8864c4a0154e2d07cc9de569658c56231f7a556fd027131ac03a97d2d8346a4c","paper_version":1,"verdict_class":"incontrovertible","resolved_title":"Natural Questions: A Benchmark for Question Answering Research","detector_version":"1.1.0","detected_arxiv_id":null,"integrity_event_id":19665,"payload_sha256":"9830ffbf591c52ad6a3c1f743985048722261b5eeb9e213678f7a5c8c7d33830","signature_b64":"NCW0FMljkIoEVU+UexALsc9F4S/uLNuELl0tLJJul0fpVcx2ILI1Y4MiikCQzZzlsFUoCyw/GJLwr+i/aX6lDQ==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-14T05:33:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y4brSiUdSH35nc9uMMIhjz0m4hQuaT7aWzT3G3E7rR9RYcpdSzw4DzlVwRLiNnGKPF5GrLkNZv1QxUwEkyDsDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T17:19:00.031527Z"},"content_sha256":"b3efe83a37821b52910948d23afbbaf4a34bff39e64cb1ece717868fa45358ed","schema_version":"1.0","event_id":"sha256:b3efe83a37821b52910948d23afbbaf4a34bff39e64cb1ece717868fa45358ed"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KNULEB4U7CAYPWYCV6FBFRU3JT/bundle.json","state_url":"https://pith.science/pith/KNULEB4U7CAYPWYCV6FBFRU3JT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KNULEB4U7CAYPWYCV6FBFRU3JT/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":"2026-08-14T17:19:00Z","links":{"resolver":"https://pith.science/pith/KNULEB4U7CAYPWYCV6FBFRU3JT","bundle":"https://pith.science/pith/KNULEB4U7CAYPWYCV6FBFRU3JT/bundle.json","state":"https://pith.science/pith/KNULEB4U7CAYPWYCV6FBFRU3JT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KNULEB4U7CAYPWYCV6FBFRU3JT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:KNULEB4U7CAYPWYCV6FBFRU3JT","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"d2d55f8ed5c6ae792cb0a6004facf859fc549cacd96e9619bbb907e76e9e2526","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-09T03:30:14Z","title_canon_sha256":"801622d8e0545d17d1b96d7e836e52d2c9628dfd25449b19ab6bbba99890fc84"},"schema_version":"1.0","source":{"id":"2608.08447","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.08447","created_at":"2026-08-11T01:22:39Z"},{"alias_kind":"arxiv_version","alias_value":"2608.08447v1","created_at":"2026-08-11T01:22:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.08447","created_at":"2026-08-11T01:22:39Z"},{"alias_kind":"pith_short_12","alias_value":"KNULEB4U7CAY","created_at":"2026-08-11T01:22:39Z"},{"alias_kind":"pith_short_16","alias_value":"KNULEB4U7CAYPWYC","created_at":"2026-08-11T01:22:39Z"},{"alias_kind":"pith_short_8","alias_value":"KNULEB4U","created_at":"2026-08-11T01:22:39Z"}],"graph_snapshots":[{"event_id":"sha256:4fa3084675fd59a6d045dbf34c9d95337d68d0a569ea4d587ab168aa3d2d6653","target":"graph","created_at":"2026-08-11T01:22:39Z","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/2608.08447/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multilingual reasoning models are commonly evaluated by whether they arrive at the correct answer, but not by whether they preserve the intended language while reasoning and responding. This omission conceals important multilingual behaviors that emerge as tasks become harder. In this paper, we study task difficulty, task accuracy, thinking-language consistency (TC), and answer-language consistency (AC) across reasoning models using PolyMath benchmark in eight languages and four difficulty levels. We uncover four findings: (1) language consistency exhibits four difficulty-dependent behaviors: ","authors_text":"Kelly Marchisio, Muhammad Ali Shafique","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-09T03:30:14Z","title":"Hidden Language Consistency Phenomena in Reasoning LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.08447","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":"sha256:7b49028a14879fe5ce69d390275f19cad44174733d728828972680ccd7dd45d1","target":"record","created_at":"2026-08-11T01:22:39Z","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":"d2d55f8ed5c6ae792cb0a6004facf859fc549cacd96e9619bbb907e76e9e2526","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-09T03:30:14Z","title_canon_sha256":"801622d8e0545d17d1b96d7e836e52d2c9628dfd25449b19ab6bbba99890fc84"},"schema_version":"1.0","source":{"id":"2608.08447","kind":"arxiv","version":1}},"canonical_sha256":"5368b20794f88187db02af8a12c69b4cf1eba86f96e8e69e5a4054d7150792c2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5368b20794f88187db02af8a12c69b4cf1eba86f96e8e69e5a4054d7150792c2","first_computed_at":"2026-08-11T01:22:39.454238Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-11T01:22:39.454238Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1wVnuL52pbZzgwXxBksUu8Jjvae+0N3BuJtEMz4keCRv5D0GNozdoLTbqSMeB1BSkrhJoczpsnsm85HWI7IkDw==","signature_status":"signed_v1","signed_at":"2026-08-11T01:22:39.456619Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.08447","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7b49028a14879fe5ce69d390275f19cad44174733d728828972680ccd7dd45d1","sha256:4fa3084675fd59a6d045dbf34c9d95337d68d0a569ea4d587ab168aa3d2d6653","sha256:b3efe83a37821b52910948d23afbbaf4a34bff39e64cb1ece717868fa45358ed"],"state_sha256":"df52e1959d98681ff4365d413b90f7cbbb1536870c6e4ebfe7e3395198cf53d2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7jRlINvZPzxF8urcn99VksCX4tMUSUcBZL0djZiTe7BuNJJmLDCrkp2/yXLc0iiiJakPd5KbLoyWDw7y+HoVBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T17:19:00.034450Z","bundle_sha256":"6c80a4d7536ac55973ffa3d8b404bf5909beb398eba0561553bb4afce6bd1ca2"}}