{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:6UYKVYRV7SO5U2ZME45THZ6IZD","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":"78c8efb3a9e17b8ec09cd419da571fd81e5de75e261a7d635c9f5b64df3e8f88","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-06-17T03:03:34Z","title_canon_sha256":"fdd563f6d5eec5be61eb3550809768eb306542d95a279bdee074c8b5ad0f0d64"},"schema_version":"1.0","source":{"id":"2406.12935","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.12935","created_at":"2026-07-05T09:57:43Z"},{"alias_kind":"arxiv_version","alias_value":"2406.12935v2","created_at":"2026-07-05T09:57:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12935","created_at":"2026-07-05T09:57:43Z"},{"alias_kind":"pith_short_12","alias_value":"6UYKVYRV7SO5","created_at":"2026-07-05T09:57:43Z"},{"alias_kind":"pith_short_16","alias_value":"6UYKVYRV7SO5U2ZM","created_at":"2026-07-05T09:57:43Z"},{"alias_kind":"pith_short_8","alias_value":"6UYKVYRV","created_at":"2026-07-05T09:57:43Z"}],"graph_snapshots":[{"event_id":"sha256:574050d88a7a5c35d58f43218252c1c318700e00c1e370b461eb156bd5d4da9c","target":"graph","created_at":"2026-07-05T09:57:43Z","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/2406.12935/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are expected to follow instructions from users and engage in conversations. Techniques to enhance LLMs' instruction-following capabilities typically fine-tune them using data structured according to a predefined chat template. Although chat templates are shown to be effective in optimizing LLM performance, their impact on safety alignment of LLMs has been less understood, which is crucial for deploying LLMs safely at scale.\n  In this paper, we investigate how chat templates affect safety alignment of LLMs. We identify a common vulnerability, named ChatBug, that is ","authors_text":"Bill Yuchen Lin, Fengqing Jiang, Luyao Niu, Radha Poovendran, Zhangchen Xu","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-06-17T03:03:34Z","title":"ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12935","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:a11961646227cb37caa8998e621953df8a5050d4b5c71c2cd921141406341bd0","target":"record","created_at":"2026-07-05T09:57:43Z","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":"78c8efb3a9e17b8ec09cd419da571fd81e5de75e261a7d635c9f5b64df3e8f88","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-06-17T03:03:34Z","title_canon_sha256":"fdd563f6d5eec5be61eb3550809768eb306542d95a279bdee074c8b5ad0f0d64"},"schema_version":"1.0","source":{"id":"2406.12935","kind":"arxiv","version":2}},"canonical_sha256":"f530aae235fc9dda6b2c273b33e7c8c8e582ece769aa5ef2c187b5a7f0696ff9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f530aae235fc9dda6b2c273b33e7c8c8e582ece769aa5ef2c187b5a7f0696ff9","first_computed_at":"2026-07-05T09:57:43.662368Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:57:43.662368Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Dh9RrazgWoeeppwbAFP4869RO1viXXwlD6hbD3soeB8A53kW1jmNB+P7wWG80thocPj7murhj0IUESo3LbSzBw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:57:43.662862Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.12935","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a11961646227cb37caa8998e621953df8a5050d4b5c71c2cd921141406341bd0","sha256:574050d88a7a5c35d58f43218252c1c318700e00c1e370b461eb156bd5d4da9c"],"state_sha256":"5d41384948c1e5adbf38b59f577e931f8cc7499e5b3e12160b10ff7d065e16f4"}