{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:SPCRGGLLOQHRVH3QVFBQGHDKBV","short_pith_number":"pith:SPCRGGLL","schema_version":"1.0","canonical_sha256":"93c513196b740f1a9f70a943031c6a0d651970fce602e39ee3f710675a0e6b8d","source":{"kind":"arxiv","id":"2302.09270","version":3},"attestation_state":"computed","paper":{"title":"Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Hao Sun, Jiale Cheng, Jiawen Deng, Minlie Huang, Zhexin Zhang","submitted_at":"2023-02-18T09:32:55Z","abstract_excerpt":"As generative large model capabilities advance, safety concerns become more pronounced in their outputs. To ensure the sustainable growth of the AI ecosystem, it's imperative to undertake a holistic evaluation and refinement of associated safety risks. This survey presents a framework for safety research pertaining to large models, delineating the landscape of safety risks as well as safety evaluation and improvement methods. We begin by introducing safety issues of wide concern, then delve into safety evaluation methods for large models, encompassing preference-based testing, adversarial atta"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2302.09270","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-02-18T09:32:55Z","cross_cats_sorted":[],"title_canon_sha256":"3875cc69efef6c01c06aa24199a1e05c8ead0d316ee0dbc3a1ff4c1a46600fbf","abstract_canon_sha256":"565e6e3e94c1c4417043d46079da158e26c9bbfa0430b48071c4159dd323ff3d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:18:24.903085Z","signature_b64":"RhANPQXdsfqkTgwzqg9ggsAtr7LQug86o76Hs5xlG+rs8FfRQN2WJ2O5q+Gvuf+wC4umFOZxdwMSEbiDkDVIDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"93c513196b740f1a9f70a943031c6a0d651970fce602e39ee3f710675a0e6b8d","last_reissued_at":"2026-07-05T07:18:24.902342Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:18:24.902342Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Hao Sun, Jiale Cheng, Jiawen Deng, Minlie Huang, Zhexin Zhang","submitted_at":"2023-02-18T09:32:55Z","abstract_excerpt":"As generative large model capabilities advance, safety concerns become more pronounced in their outputs. To ensure the sustainable growth of the AI ecosystem, it's imperative to undertake a holistic evaluation and refinement of associated safety risks. This survey presents a framework for safety research pertaining to large models, delineating the landscape of safety risks as well as safety evaluation and improvement methods. We begin by introducing safety issues of wide concern, then delve into safety evaluation methods for large models, encompassing preference-based testing, adversarial atta"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.09270","kind":"arxiv","version":3},"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/2302.09270/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2302.09270","created_at":"2026-07-05T07:18:24.902471+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.09270v3","created_at":"2026-07-05T07:18:24.902471+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.09270","created_at":"2026-07-05T07:18:24.902471+00:00"},{"alias_kind":"pith_short_12","alias_value":"SPCRGGLLOQHR","created_at":"2026-07-05T07:18:24.902471+00:00"},{"alias_kind":"pith_short_16","alias_value":"SPCRGGLLOQHRVH3Q","created_at":"2026-07-05T07:18:24.902471+00:00"},{"alias_kind":"pith_short_8","alias_value":"SPCRGGLL","created_at":"2026-07-05T07:18:24.902471+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.19292","citing_title":"The safety failures we are not instrumenting: a perspective on hidden safety-critical challenges in modern AI systems","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SPCRGGLLOQHRVH3QVFBQGHDKBV","json":"https://pith.science/pith/SPCRGGLLOQHRVH3QVFBQGHDKBV.json","graph_json":"https://pith.science/api/pith-number/SPCRGGLLOQHRVH3QVFBQGHDKBV/graph.json","events_json":"https://pith.science/api/pith-number/SPCRGGLLOQHRVH3QVFBQGHDKBV/events.json","paper":"https://pith.science/paper/SPCRGGLL"},"agent_actions":{"view_html":"https://pith.science/pith/SPCRGGLLOQHRVH3QVFBQGHDKBV","download_json":"https://pith.science/pith/SPCRGGLLOQHRVH3QVFBQGHDKBV.json","view_paper":"https://pith.science/paper/SPCRGGLL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.09270&json=true","fetch_graph":"https://pith.science/api/pith-number/SPCRGGLLOQHRVH3QVFBQGHDKBV/graph.json","fetch_events":"https://pith.science/api/pith-number/SPCRGGLLOQHRVH3QVFBQGHDKBV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SPCRGGLLOQHRVH3QVFBQGHDKBV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SPCRGGLLOQHRVH3QVFBQGHDKBV/action/storage_attestation","attest_author":"https://pith.science/pith/SPCRGGLLOQHRVH3QVFBQGHDKBV/action/author_attestation","sign_citation":"https://pith.science/pith/SPCRGGLLOQHRVH3QVFBQGHDKBV/action/citation_signature","submit_replication":"https://pith.science/pith/SPCRGGLLOQHRVH3QVFBQGHDKBV/action/replication_record"}},"created_at":"2026-07-05T07:18:24.902471+00:00","updated_at":"2026-07-05T07:18:24.902471+00:00"}