{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:3ZLJYRNOO3KLTYPIPE5X3GJIWE","short_pith_number":"pith:3ZLJYRNO","canonical_record":{"source":{"id":"2505.19690","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-26T08:49:19Z","cross_cats_sorted":[],"title_canon_sha256":"0652a6c7303e55d156dad1410a192611d259c4e4619560f8947180e22302ee93","abstract_canon_sha256":"387e9bb15c43d369ff2ae025e31ac6ef5f1f07e7d15679caa22de2c81b3a9c9d"},"schema_version":"1.0"},"canonical_sha256":"de569c45ae76d4b9e1e8793b7d9928b10ddf077c0f4e13c224e06d6332abcc41","source":{"kind":"arxiv","id":"2505.19690","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19690","created_at":"2026-07-05T11:09:34Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19690v1","created_at":"2026-07-05T11:09:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19690","created_at":"2026-07-05T11:09:34Z"},{"alias_kind":"pith_short_12","alias_value":"3ZLJYRNOO3KL","created_at":"2026-07-05T11:09:34Z"},{"alias_kind":"pith_short_16","alias_value":"3ZLJYRNOO3KLTYPI","created_at":"2026-07-05T11:09:34Z"},{"alias_kind":"pith_short_8","alias_value":"3ZLJYRNO","created_at":"2026-07-05T11:09:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:3ZLJYRNOO3KLTYPIPE5X3GJIWE","target":"record","payload":{"canonical_record":{"source":{"id":"2505.19690","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-26T08:49:19Z","cross_cats_sorted":[],"title_canon_sha256":"0652a6c7303e55d156dad1410a192611d259c4e4619560f8947180e22302ee93","abstract_canon_sha256":"387e9bb15c43d369ff2ae025e31ac6ef5f1f07e7d15679caa22de2c81b3a9c9d"},"schema_version":"1.0"},"canonical_sha256":"de569c45ae76d4b9e1e8793b7d9928b10ddf077c0f4e13c224e06d6332abcc41","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:34.524481Z","signature_b64":"V3tH0BnyQUE3ObuQzsi3oTGvhLJbt4QzOvqqxrFnzp5cAuiItjZ1d8eGIJr0lPFYDwYwubzbSwtTmUjp5OBDCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de569c45ae76d4b9e1e8793b7d9928b10ddf077c0f4e13c224e06d6332abcc41","last_reissued_at":"2026-07-05T11:09:34.523953Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:34.523953Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.19690","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":"2026-07-05T11:09:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u1eaZhtftw0M+8lhigG3r6UGZ4CCJaJMTAOPiLl6NdEbxnypkX4+d2YcrMZfgZxEEDAvM4SgxUbMRUzll9QRDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-18T17:14:45.960499Z"},"content_sha256":"54f2e5d0b14e598e5673cf0d835de18f379e5545efcffb6a9f0135b3c5ef948c","schema_version":"1.0","event_id":"sha256:54f2e5d0b14e598e5673cf0d835de18f379e5545efcffb6a9f0135b3c5ef948c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:3ZLJYRNOO3KLTYPIPE5X3GJIWE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Beyond Safe Answers: A Benchmark for Evaluating True Risk Awareness in Large Reasoning Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Baihui Zheng, Boren Zheng, Bo Zheng, Jiaheng Liu, Jian Yang, Kaifu Zhang, Kerui Cao, Weixun Wang, Wenbo Su, Xiaoyong Zhu, Yingshui Tan, Zhendong Liu","submitted_at":"2025-05-26T08:49:19Z","abstract_excerpt":"Despite the remarkable proficiency of \\textit{Large Reasoning Models} (LRMs) in handling complex reasoning tasks, their reliability in safety-critical scenarios remains uncertain. Existing evaluations primarily assess response-level safety, neglecting a critical issue we identify as \\textbf{\\textit{Superficial Safety Alignment} (SSA)} -- a phenomenon where models produce superficially safe outputs while internal reasoning processes fail to genuinely detect and mitigate underlying risks, resulting in inconsistent safety behaviors across multiple sampling attempts. To systematically investigate "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19690","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/2505.19690/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":"2026-07-05T11:09:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kxkgekWQTuuq0fGUEuyGPUukMA2yce3HBDWCM0SMr8AexPgypSTOiiqOu3Tq4pTnO82J5/s40HnG76zybp6NBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-18T17:14:45.960878Z"},"content_sha256":"deff8cb907f7120f98b0ea443e9e54a9c3c5d4ba6ab62b8f84e52946c7048324","schema_version":"1.0","event_id":"sha256:deff8cb907f7120f98b0ea443e9e54a9c3c5d4ba6ab62b8f84e52946c7048324"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3ZLJYRNOO3KLTYPIPE5X3GJIWE/bundle.json","state_url":"https://pith.science/pith/3ZLJYRNOO3KLTYPIPE5X3GJIWE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3ZLJYRNOO3KLTYPIPE5X3GJIWE/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":"2026-07-18T17:14:45Z","links":{"resolver":"https://pith.science/pith/3ZLJYRNOO3KLTYPIPE5X3GJIWE","bundle":"https://pith.science/pith/3ZLJYRNOO3KLTYPIPE5X3GJIWE/bundle.json","state":"https://pith.science/pith/3ZLJYRNOO3KLTYPIPE5X3GJIWE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3ZLJYRNOO3KLTYPIPE5X3GJIWE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:3ZLJYRNOO3KLTYPIPE5X3GJIWE","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":"387e9bb15c43d369ff2ae025e31ac6ef5f1f07e7d15679caa22de2c81b3a9c9d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-26T08:49:19Z","title_canon_sha256":"0652a6c7303e55d156dad1410a192611d259c4e4619560f8947180e22302ee93"},"schema_version":"1.0","source":{"id":"2505.19690","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19690","created_at":"2026-07-05T11:09:34Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19690v1","created_at":"2026-07-05T11:09:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19690","created_at":"2026-07-05T11:09:34Z"},{"alias_kind":"pith_short_12","alias_value":"3ZLJYRNOO3KL","created_at":"2026-07-05T11:09:34Z"},{"alias_kind":"pith_short_16","alias_value":"3ZLJYRNOO3KLTYPI","created_at":"2026-07-05T11:09:34Z"},{"alias_kind":"pith_short_8","alias_value":"3ZLJYRNO","created_at":"2026-07-05T11:09:34Z"}],"graph_snapshots":[{"event_id":"sha256:deff8cb907f7120f98b0ea443e9e54a9c3c5d4ba6ab62b8f84e52946c7048324","target":"graph","created_at":"2026-07-05T11:09:34Z","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/2505.19690/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Despite the remarkable proficiency of \\textit{Large Reasoning Models} (LRMs) in handling complex reasoning tasks, their reliability in safety-critical scenarios remains uncertain. Existing evaluations primarily assess response-level safety, neglecting a critical issue we identify as \\textbf{\\textit{Superficial Safety Alignment} (SSA)} -- a phenomenon where models produce superficially safe outputs while internal reasoning processes fail to genuinely detect and mitigate underlying risks, resulting in inconsistent safety behaviors across multiple sampling attempts. To systematically investigate ","authors_text":"Baihui Zheng, Boren Zheng, Bo Zheng, Jiaheng Liu, Jian Yang, Kaifu Zhang, Kerui Cao, Weixun Wang, Wenbo Su, Xiaoyong Zhu, Yingshui Tan, Zhendong Liu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-26T08:49:19Z","title":"Beyond Safe Answers: A Benchmark for Evaluating True Risk Awareness in Large Reasoning Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19690","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":"sha256:54f2e5d0b14e598e5673cf0d835de18f379e5545efcffb6a9f0135b3c5ef948c","target":"record","created_at":"2026-07-05T11:09:34Z","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":"387e9bb15c43d369ff2ae025e31ac6ef5f1f07e7d15679caa22de2c81b3a9c9d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-26T08:49:19Z","title_canon_sha256":"0652a6c7303e55d156dad1410a192611d259c4e4619560f8947180e22302ee93"},"schema_version":"1.0","source":{"id":"2505.19690","kind":"arxiv","version":1}},"canonical_sha256":"de569c45ae76d4b9e1e8793b7d9928b10ddf077c0f4e13c224e06d6332abcc41","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"de569c45ae76d4b9e1e8793b7d9928b10ddf077c0f4e13c224e06d6332abcc41","first_computed_at":"2026-07-05T11:09:34.523953Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:09:34.523953Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"V3tH0BnyQUE3ObuQzsi3oTGvhLJbt4QzOvqqxrFnzp5cAuiItjZ1d8eGIJr0lPFYDwYwubzbSwtTmUjp5OBDCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:09:34.524481Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.19690","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:54f2e5d0b14e598e5673cf0d835de18f379e5545efcffb6a9f0135b3c5ef948c","sha256:deff8cb907f7120f98b0ea443e9e54a9c3c5d4ba6ab62b8f84e52946c7048324"],"state_sha256":"f0a1f2e4a4efa139e353b08cea928941ef0c9a08c1814b1283a38ab0127d6e8d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mLWGpdCDykgEre/xUw5Kt+WbMihrEDxo8cNgp5EHi8hHmdXbfF86a+oQeQ7x8PhiaokIt/cmoPQG8CdVS0CmAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-18T17:14:45.963285Z","bundle_sha256":"67a84983cecf89da96c4037d461ba08576b7197f4f09544bccffba8e49231fa8"}}