{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:45EOK6ETEQ47H5KRUWD6CM37HE","short_pith_number":"pith:45EOK6ET","canonical_record":{"source":{"id":"2411.12762","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-16T13:07:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"af641aad60811b0dbf7dd37938e7d4ca6931a3994daa19fb82c330e161770985","abstract_canon_sha256":"7d65d44812b732bd8a537faec14b132ef0a3c05106735c770899b9146aab5a6e"},"schema_version":"1.0"},"canonical_sha256":"e748e578932439f3f551a587e1337f390c6d92a9933de09007bd5a7d06557cff","source":{"kind":"arxiv","id":"2411.12762","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.12762","created_at":"2026-07-05T09:41:03Z"},{"alias_kind":"arxiv_version","alias_value":"2411.12762v2","created_at":"2026-07-05T09:41:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.12762","created_at":"2026-07-05T09:41:03Z"},{"alias_kind":"pith_short_12","alias_value":"45EOK6ETEQ47","created_at":"2026-07-05T09:41:03Z"},{"alias_kind":"pith_short_16","alias_value":"45EOK6ETEQ47H5KR","created_at":"2026-07-05T09:41:03Z"},{"alias_kind":"pith_short_8","alias_value":"45EOK6ET","created_at":"2026-07-05T09:41:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:45EOK6ETEQ47H5KRUWD6CM37HE","target":"record","payload":{"canonical_record":{"source":{"id":"2411.12762","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-16T13:07:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"af641aad60811b0dbf7dd37938e7d4ca6931a3994daa19fb82c330e161770985","abstract_canon_sha256":"7d65d44812b732bd8a537faec14b132ef0a3c05106735c770899b9146aab5a6e"},"schema_version":"1.0"},"canonical_sha256":"e748e578932439f3f551a587e1337f390c6d92a9933de09007bd5a7d06557cff","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:41:03.844387Z","signature_b64":"fNy1TlaBnW2mROjvaA6oq/Wi0NUmB3P2vra+jMpRNiYvr2179rhhYrsJOS+KPGNjD5SZcomVdB6tZgeznCxpBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e748e578932439f3f551a587e1337f390c6d92a9933de09007bd5a7d06557cff","last_reissued_at":"2026-07-05T09:41:03.843795Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:41:03.843795Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.12762","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":"2026-07-05T09:41:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"y1GMipeA8tpFCSYLiZCr6mdrr14LsbgmRkzJmcpTZbXwMpASiN2GNcBaOSSXmAyoIpOeZAp+RVxYEB2jhHdjCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T14:15:21.447657Z"},"content_sha256":"adf6e8aa40b016b45f126369e8e1637162dca7eb05c4e5a5e57cb6c6e36757fc","schema_version":"1.0","event_id":"sha256:adf6e8aa40b016b45f126369e8e1637162dca7eb05c4e5a5e57cb6c6e36757fc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:45EOK6ETEQ47H5KRUWD6CM37HE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Playing Language Game with LLMs Leads to Jailbreaking","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Congyi Li, Fangming Dong, Kai Chen, Shu Wu, Yu Peng, Zewen Long","submitted_at":"2024-11-16T13:07:13Z","abstract_excerpt":"The advent of large language models (LLMs) has spurred the development of numerous jailbreak techniques aimed at circumventing their security defenses against malicious attacks. An effective jailbreak approach is to identify a domain where safety generalization fails, a phenomenon known as mismatched generalization. In this paper, we introduce two novel jailbreak methods based on mismatched generalization: natural language games and custom language games, both of which effectively bypass the safety mechanisms of LLMs, with various kinds and different variants, making them hard to defend and le"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.12762","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/2411.12762/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":"2026-07-05T09:41:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7Ljj2vnj7Ft72QHo+W9wMf2hmKRBw0YlbeiXjnLHOCP/6+dpjau/YtH6dgI422V5C0AxsldT8C8XeHewUVECCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T14:15:21.448209Z"},"content_sha256":"97bbdefac16fcf3086c5079d6baf15415def032968ecba9767caa2b359fd6e41","schema_version":"1.0","event_id":"sha256:97bbdefac16fcf3086c5079d6baf15415def032968ecba9767caa2b359fd6e41"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/45EOK6ETEQ47H5KRUWD6CM37HE/bundle.json","state_url":"https://pith.science/pith/45EOK6ETEQ47H5KRUWD6CM37HE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/45EOK6ETEQ47H5KRUWD6CM37HE/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":"2026-08-09T14:15:21Z","links":{"resolver":"https://pith.science/pith/45EOK6ETEQ47H5KRUWD6CM37HE","bundle":"https://pith.science/pith/45EOK6ETEQ47H5KRUWD6CM37HE/bundle.json","state":"https://pith.science/pith/45EOK6ETEQ47H5KRUWD6CM37HE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/45EOK6ETEQ47H5KRUWD6CM37HE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:45EOK6ETEQ47H5KRUWD6CM37HE","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":"7d65d44812b732bd8a537faec14b132ef0a3c05106735c770899b9146aab5a6e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-16T13:07:13Z","title_canon_sha256":"af641aad60811b0dbf7dd37938e7d4ca6931a3994daa19fb82c330e161770985"},"schema_version":"1.0","source":{"id":"2411.12762","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.12762","created_at":"2026-07-05T09:41:03Z"},{"alias_kind":"arxiv_version","alias_value":"2411.12762v2","created_at":"2026-07-05T09:41:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.12762","created_at":"2026-07-05T09:41:03Z"},{"alias_kind":"pith_short_12","alias_value":"45EOK6ETEQ47","created_at":"2026-07-05T09:41:03Z"},{"alias_kind":"pith_short_16","alias_value":"45EOK6ETEQ47H5KR","created_at":"2026-07-05T09:41:03Z"},{"alias_kind":"pith_short_8","alias_value":"45EOK6ET","created_at":"2026-07-05T09:41:03Z"}],"graph_snapshots":[{"event_id":"sha256:97bbdefac16fcf3086c5079d6baf15415def032968ecba9767caa2b359fd6e41","target":"graph","created_at":"2026-07-05T09:41:03Z","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/2411.12762/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The advent of large language models (LLMs) has spurred the development of numerous jailbreak techniques aimed at circumventing their security defenses against malicious attacks. An effective jailbreak approach is to identify a domain where safety generalization fails, a phenomenon known as mismatched generalization. In this paper, we introduce two novel jailbreak methods based on mismatched generalization: natural language games and custom language games, both of which effectively bypass the safety mechanisms of LLMs, with various kinds and different variants, making them hard to defend and le","authors_text":"Congyi Li, Fangming Dong, Kai Chen, Shu Wu, Yu Peng, Zewen Long","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-16T13:07:13Z","title":"Playing Language Game with LLMs Leads to Jailbreaking"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.12762","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:adf6e8aa40b016b45f126369e8e1637162dca7eb05c4e5a5e57cb6c6e36757fc","target":"record","created_at":"2026-07-05T09:41:03Z","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":"7d65d44812b732bd8a537faec14b132ef0a3c05106735c770899b9146aab5a6e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-16T13:07:13Z","title_canon_sha256":"af641aad60811b0dbf7dd37938e7d4ca6931a3994daa19fb82c330e161770985"},"schema_version":"1.0","source":{"id":"2411.12762","kind":"arxiv","version":2}},"canonical_sha256":"e748e578932439f3f551a587e1337f390c6d92a9933de09007bd5a7d06557cff","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e748e578932439f3f551a587e1337f390c6d92a9933de09007bd5a7d06557cff","first_computed_at":"2026-07-05T09:41:03.843795Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:41:03.843795Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fNy1TlaBnW2mROjvaA6oq/Wi0NUmB3P2vra+jMpRNiYvr2179rhhYrsJOS+KPGNjD5SZcomVdB6tZgeznCxpBw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:41:03.844387Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.12762","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:adf6e8aa40b016b45f126369e8e1637162dca7eb05c4e5a5e57cb6c6e36757fc","sha256:97bbdefac16fcf3086c5079d6baf15415def032968ecba9767caa2b359fd6e41"],"state_sha256":"6171ed4ad65b9e440df88ceecda9541cdad55e8f814dc4f481ecfdf8d2eeae39"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZsqskRMFXUSeqAdxel1/JeRsA3Tv4frGF+ir3i1EEBfAiL193/To/fcTqTtHYqyQ6FLeKr5u/mc6OBPqg8LBAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T14:15:21.451963Z","bundle_sha256":"12d1b19c1023d7195aed78a33958812964e5b5b49ad6bbe3790bece2611b5caa"}}