{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:7HMVMCR2QRY6RP54GLD3NNNKIF","short_pith_number":"pith:7HMVMCR2","canonical_record":{"source":{"id":"2402.01706","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T02:57:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1f3ceba301d13bafa62c5827cb72f69bc202ff4b320c912da90224e6a4d15a34","abstract_canon_sha256":"658591e456a69865057e26f3f090d9ddb10ecd2d1f59251ef46af67c58267585"},"schema_version":"1.0"},"canonical_sha256":"f9d9560a3a8471e8bfbc32c7b6b5aa41467aca42287e3794869dc6e98e9c623d","source":{"kind":"arxiv","id":"2402.01706","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.01706","created_at":"2026-07-05T07:41:05Z"},{"alias_kind":"arxiv_version","alias_value":"2402.01706v1","created_at":"2026-07-05T07:41:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.01706","created_at":"2026-07-05T07:41:05Z"},{"alias_kind":"pith_short_12","alias_value":"7HMVMCR2QRY6","created_at":"2026-07-05T07:41:05Z"},{"alias_kind":"pith_short_16","alias_value":"7HMVMCR2QRY6RP54","created_at":"2026-07-05T07:41:05Z"},{"alias_kind":"pith_short_8","alias_value":"7HMVMCR2","created_at":"2026-07-05T07:41:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:7HMVMCR2QRY6RP54GLD3NNNKIF","target":"record","payload":{"canonical_record":{"source":{"id":"2402.01706","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T02:57:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1f3ceba301d13bafa62c5827cb72f69bc202ff4b320c912da90224e6a4d15a34","abstract_canon_sha256":"658591e456a69865057e26f3f090d9ddb10ecd2d1f59251ef46af67c58267585"},"schema_version":"1.0"},"canonical_sha256":"f9d9560a3a8471e8bfbc32c7b6b5aa41467aca42287e3794869dc6e98e9c623d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:41:05.380281Z","signature_b64":"LYRMScspUJ3ENWapEiieeARlOJ2pE33Ng1fMa10OKGSkTSVmR3QwxG1CtBgoWVU0XDZXyxMyxNSFT+f8C3MQBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f9d9560a3a8471e8bfbc32c7b6b5aa41467aca42287e3794869dc6e98e9c623d","last_reissued_at":"2026-07-05T07:41:05.379871Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:41:05.379871Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.01706","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":"2026-07-05T07:41:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Kj6p+AfKwyR7Kmr8AHIz2EQAmyX/3eNpeDrCgzQ671t2OgMz7Rd5WAOupTUal5mvYTOcsRIiO1NSq/HI1k3VDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T14:33:52.020774Z"},"content_sha256":"d7c1093d283f904c98d45e07a20289e9d7ee944acfd0a007bdf127592d5e4095","schema_version":"1.0","event_id":"sha256:d7c1093d283f904c98d45e07a20289e9d7ee944acfd0a007bdf127592d5e4095"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:7HMVMCR2QRY6RP54GLD3NNNKIF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MULTIVERSE: Exposing Large Language Model Alignment Problems in Diverse Worlds","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Xiangyu Zhang, Xiaolong Jin, Zhuo Zhang","submitted_at":"2024-01-25T02:57:40Z","abstract_excerpt":"Large Language Model (LLM) alignment aims to ensure that LLM outputs match with human values. Researchers have demonstrated the severity of alignment problems with a large spectrum of jailbreak techniques that can induce LLMs to produce malicious content during conversations. Finding the corresponding jailbreaking prompts usually requires substantial human intelligence or computation resources. In this paper, we report that LLMs have different levels of alignment in various contexts. As such, by systematically constructing many contexts, called worlds, leveraging a Domain Specific Language des"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.01706","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/2402.01706/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":"2026-07-05T07:41:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XcsmIjFcBa0YRDA7DMp6go+ZucERK5zj+AEnGR5WysJau7jhjqHfFL/KWYQKJDTJJ+xtbQhXk0STEv13JfJzDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T14:33:52.021315Z"},"content_sha256":"ca07c0ad879972d05ce4764de61373ba0057a0b8a1dfb57376c6d455649d5732","schema_version":"1.0","event_id":"sha256:ca07c0ad879972d05ce4764de61373ba0057a0b8a1dfb57376c6d455649d5732"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7HMVMCR2QRY6RP54GLD3NNNKIF/bundle.json","state_url":"https://pith.science/pith/7HMVMCR2QRY6RP54GLD3NNNKIF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7HMVMCR2QRY6RP54GLD3NNNKIF/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":"2026-08-10T14:33:52Z","links":{"resolver":"https://pith.science/pith/7HMVMCR2QRY6RP54GLD3NNNKIF","bundle":"https://pith.science/pith/7HMVMCR2QRY6RP54GLD3NNNKIF/bundle.json","state":"https://pith.science/pith/7HMVMCR2QRY6RP54GLD3NNNKIF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7HMVMCR2QRY6RP54GLD3NNNKIF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:7HMVMCR2QRY6RP54GLD3NNNKIF","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":"658591e456a69865057e26f3f090d9ddb10ecd2d1f59251ef46af67c58267585","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T02:57:40Z","title_canon_sha256":"1f3ceba301d13bafa62c5827cb72f69bc202ff4b320c912da90224e6a4d15a34"},"schema_version":"1.0","source":{"id":"2402.01706","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.01706","created_at":"2026-07-05T07:41:05Z"},{"alias_kind":"arxiv_version","alias_value":"2402.01706v1","created_at":"2026-07-05T07:41:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.01706","created_at":"2026-07-05T07:41:05Z"},{"alias_kind":"pith_short_12","alias_value":"7HMVMCR2QRY6","created_at":"2026-07-05T07:41:05Z"},{"alias_kind":"pith_short_16","alias_value":"7HMVMCR2QRY6RP54","created_at":"2026-07-05T07:41:05Z"},{"alias_kind":"pith_short_8","alias_value":"7HMVMCR2","created_at":"2026-07-05T07:41:05Z"}],"graph_snapshots":[{"event_id":"sha256:ca07c0ad879972d05ce4764de61373ba0057a0b8a1dfb57376c6d455649d5732","target":"graph","created_at":"2026-07-05T07:41:05Z","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/2402.01706/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Model (LLM) alignment aims to ensure that LLM outputs match with human values. Researchers have demonstrated the severity of alignment problems with a large spectrum of jailbreak techniques that can induce LLMs to produce malicious content during conversations. Finding the corresponding jailbreaking prompts usually requires substantial human intelligence or computation resources. In this paper, we report that LLMs have different levels of alignment in various contexts. As such, by systematically constructing many contexts, called worlds, leveraging a Domain Specific Language des","authors_text":"Xiangyu Zhang, Xiaolong Jin, Zhuo Zhang","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T02:57:40Z","title":"MULTIVERSE: Exposing Large Language Model Alignment Problems in Diverse Worlds"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.01706","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:d7c1093d283f904c98d45e07a20289e9d7ee944acfd0a007bdf127592d5e4095","target":"record","created_at":"2026-07-05T07:41:05Z","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":"658591e456a69865057e26f3f090d9ddb10ecd2d1f59251ef46af67c58267585","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-25T02:57:40Z","title_canon_sha256":"1f3ceba301d13bafa62c5827cb72f69bc202ff4b320c912da90224e6a4d15a34"},"schema_version":"1.0","source":{"id":"2402.01706","kind":"arxiv","version":1}},"canonical_sha256":"f9d9560a3a8471e8bfbc32c7b6b5aa41467aca42287e3794869dc6e98e9c623d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f9d9560a3a8471e8bfbc32c7b6b5aa41467aca42287e3794869dc6e98e9c623d","first_computed_at":"2026-07-05T07:41:05.379871Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:41:05.379871Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LYRMScspUJ3ENWapEiieeARlOJ2pE33Ng1fMa10OKGSkTSVmR3QwxG1CtBgoWVU0XDZXyxMyxNSFT+f8C3MQBA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:41:05.380281Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.01706","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d7c1093d283f904c98d45e07a20289e9d7ee944acfd0a007bdf127592d5e4095","sha256:ca07c0ad879972d05ce4764de61373ba0057a0b8a1dfb57376c6d455649d5732"],"state_sha256":"b2b68f4c52795a43a73262b416771f333cfe50a9a14e478b21b8b2dbf4667601"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"y8jpJgDpsvO7u9ZpSXcOz4Gz8Khlyq7D7z8LjXdsXLqRZDH/U3QXcNxVx4VxYFGm9D5+rapoQU+5XIEbe+yEAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T14:33:52.025426Z","bundle_sha256":"1bf40f8ed94314ed19a8a05b2186b94bfee61694847ef88efd6c327d5f049dfd"}}