{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:EKMU4QMREAPGY2LAUMNMBAVOJ5","short_pith_number":"pith:EKMU4QMR","canonical_record":{"source":{"id":"2508.15648","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T15:26:09Z","cross_cats_sorted":[],"title_canon_sha256":"57685d427a26aba20c13746e01a8b7cf39f3017575f20de220dc2c73688b3766","abstract_canon_sha256":"943a549122944f60e0d7510c21bdc3ba101dbab3935498ea3a168fcf21845c2a"},"schema_version":"1.0"},"canonical_sha256":"22994e4191201e6c6960a31ac082ae4f687c22781d1ad0425fd2db879996bf10","source":{"kind":"arxiv","id":"2508.15648","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.15648","created_at":"2026-07-05T11:59:20Z"},{"alias_kind":"arxiv_version","alias_value":"2508.15648v2","created_at":"2026-07-05T11:59:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.15648","created_at":"2026-07-05T11:59:20Z"},{"alias_kind":"pith_short_12","alias_value":"EKMU4QMREAPG","created_at":"2026-07-05T11:59:20Z"},{"alias_kind":"pith_short_16","alias_value":"EKMU4QMREAPGY2LA","created_at":"2026-07-05T11:59:20Z"},{"alias_kind":"pith_short_8","alias_value":"EKMU4QMR","created_at":"2026-07-05T11:59:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:EKMU4QMREAPGY2LAUMNMBAVOJ5","target":"record","payload":{"canonical_record":{"source":{"id":"2508.15648","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T15:26:09Z","cross_cats_sorted":[],"title_canon_sha256":"57685d427a26aba20c13746e01a8b7cf39f3017575f20de220dc2c73688b3766","abstract_canon_sha256":"943a549122944f60e0d7510c21bdc3ba101dbab3935498ea3a168fcf21845c2a"},"schema_version":"1.0"},"canonical_sha256":"22994e4191201e6c6960a31ac082ae4f687c22781d1ad0425fd2db879996bf10","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:59:20.465021Z","signature_b64":"TduDUB0c0Vd+PwriAjq8hg1Ol1op/wjOixqZp7M6MBZ8DAAW8DRTvI4m2P60wk8Mn8yXi+ANUi5apd/Medx0AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"22994e4191201e6c6960a31ac082ae4f687c22781d1ad0425fd2db879996bf10","last_reissued_at":"2026-07-05T11:59:20.464564Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:59:20.464564Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.15648","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:59:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"X23z6o9RGZwnSXvbQm2KVhXxB3kUO00sKmSgPNdsqJKjYBXJASbmsoZbC0rpa/CW85L0mZ80DAaUq29oJt5YAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T06:07:38.865873Z"},"content_sha256":"5318eebe8d3a5b9aa58ab43d4924d3f958ced30785220056ac622bfa4fdc33f8","schema_version":"1.0","event_id":"sha256:5318eebe8d3a5b9aa58ab43d4924d3f958ced30785220056ac622bfa4fdc33f8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:EKMU4QMREAPGY2LAUMNMBAVOJ5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SDGO: Self-Discrimination-Guided Optimization for Consistent Safety in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dailin Li, Jiajun Chen, Jiaming Wang, Peng Ding, Shujian Huang, Wei Zou, Wen Sun","submitted_at":"2025-08-21T15:26:09Z","abstract_excerpt":"Large Language Models (LLMs) excel at various natural language processing tasks but remain vulnerable to jailbreaking attacks that induce harmful content generation. In this paper, we reveal a critical safety inconsistency: LLMs can more effectively identify harmful requests as discriminators than defend against them as generators. This insight inspires us to explore aligning the model's inherent discrimination and generation capabilities. To this end, we propose SDGO (Self-Discrimination-Guided Optimization), a reinforcement learning framework that leverages the model's own discrimination cap"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.15648","kind":"arxiv","version":2},"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/2508.15648/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":"2026-07-05T11:59:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zjy+tn5v+eLbgh1W29NUTa8ihWS8oZZGrleBH4biunRNRk2EpT8paidAaBGKHp2rTqEXSzmb6P2c7AiTBBYKCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T06:07:38.866196Z"},"content_sha256":"4e317a3614b4a3b845915efef9c0bd666c2b138a1412e98699dcddbba4de3eb8","schema_version":"1.0","event_id":"sha256:4e317a3614b4a3b845915efef9c0bd666c2b138a1412e98699dcddbba4de3eb8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EKMU4QMREAPGY2LAUMNMBAVOJ5/bundle.json","state_url":"https://pith.science/pith/EKMU4QMREAPGY2LAUMNMBAVOJ5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EKMU4QMREAPGY2LAUMNMBAVOJ5/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":"2026-08-05T06:07:38Z","links":{"resolver":"https://pith.science/pith/EKMU4QMREAPGY2LAUMNMBAVOJ5","bundle":"https://pith.science/pith/EKMU4QMREAPGY2LAUMNMBAVOJ5/bundle.json","state":"https://pith.science/pith/EKMU4QMREAPGY2LAUMNMBAVOJ5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EKMU4QMREAPGY2LAUMNMBAVOJ5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:EKMU4QMREAPGY2LAUMNMBAVOJ5","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":"943a549122944f60e0d7510c21bdc3ba101dbab3935498ea3a168fcf21845c2a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T15:26:09Z","title_canon_sha256":"57685d427a26aba20c13746e01a8b7cf39f3017575f20de220dc2c73688b3766"},"schema_version":"1.0","source":{"id":"2508.15648","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.15648","created_at":"2026-07-05T11:59:20Z"},{"alias_kind":"arxiv_version","alias_value":"2508.15648v2","created_at":"2026-07-05T11:59:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.15648","created_at":"2026-07-05T11:59:20Z"},{"alias_kind":"pith_short_12","alias_value":"EKMU4QMREAPG","created_at":"2026-07-05T11:59:20Z"},{"alias_kind":"pith_short_16","alias_value":"EKMU4QMREAPGY2LA","created_at":"2026-07-05T11:59:20Z"},{"alias_kind":"pith_short_8","alias_value":"EKMU4QMR","created_at":"2026-07-05T11:59:20Z"}],"graph_snapshots":[{"event_id":"sha256:4e317a3614b4a3b845915efef9c0bd666c2b138a1412e98699dcddbba4de3eb8","target":"graph","created_at":"2026-07-05T11:59:20Z","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/2508.15648/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) excel at various natural language processing tasks but remain vulnerable to jailbreaking attacks that induce harmful content generation. In this paper, we reveal a critical safety inconsistency: LLMs can more effectively identify harmful requests as discriminators than defend against them as generators. This insight inspires us to explore aligning the model's inherent discrimination and generation capabilities. To this end, we propose SDGO (Self-Discrimination-Guided Optimization), a reinforcement learning framework that leverages the model's own discrimination cap","authors_text":"Dailin Li, Jiajun Chen, Jiaming Wang, Peng Ding, Shujian Huang, Wei Zou, Wen Sun","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T15:26:09Z","title":"SDGO: Self-Discrimination-Guided Optimization for Consistent Safety in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.15648","kind":"arxiv","version":2},"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":"sha256:5318eebe8d3a5b9aa58ab43d4924d3f958ced30785220056ac622bfa4fdc33f8","target":"record","created_at":"2026-07-05T11:59:20Z","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":"943a549122944f60e0d7510c21bdc3ba101dbab3935498ea3a168fcf21845c2a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-21T15:26:09Z","title_canon_sha256":"57685d427a26aba20c13746e01a8b7cf39f3017575f20de220dc2c73688b3766"},"schema_version":"1.0","source":{"id":"2508.15648","kind":"arxiv","version":2}},"canonical_sha256":"22994e4191201e6c6960a31ac082ae4f687c22781d1ad0425fd2db879996bf10","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"22994e4191201e6c6960a31ac082ae4f687c22781d1ad0425fd2db879996bf10","first_computed_at":"2026-07-05T11:59:20.464564Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:59:20.464564Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TduDUB0c0Vd+PwriAjq8hg1Ol1op/wjOixqZp7M6MBZ8DAAW8DRTvI4m2P60wk8Mn8yXi+ANUi5apd/Medx0AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:59:20.465021Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.15648","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5318eebe8d3a5b9aa58ab43d4924d3f958ced30785220056ac622bfa4fdc33f8","sha256:4e317a3614b4a3b845915efef9c0bd666c2b138a1412e98699dcddbba4de3eb8"],"state_sha256":"b2f097b76145012407bf68641c3f4ac49da9073a8763b6cef2cb1801dc37e47e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xit/Gh4TOn7q+llWdL/XHuodFoWl37KM1UptgG2Wu/do4q6PmKVK3NOmu8a/K8JDzZADn6IYdQlSX9xKdw+5Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T06:07:38.869414Z","bundle_sha256":"932014908762650d5a9c4637cee1b0d519da72e51cfb0bbc547e3b99674570a6"}}