{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:N63GVVYCQTWAHN6QFS3PWQZTRT","short_pith_number":"pith:N63GVVYC","canonical_record":{"source":{"id":"2405.13967","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-22T20:08:48Z","cross_cats_sorted":[],"title_canon_sha256":"3d7372cc5bc96ca877ec9893bce61cc14b417a728739db58728a1579bac8a4f9","abstract_canon_sha256":"3b2d4d533d477e3722bbabf5a100f39d94968d1b7e031b9fff2b3106a94611fa"},"schema_version":"1.0"},"canonical_sha256":"6fb66ad70284ec03b7d02cb6fb43338ceabdb61f1efd65cd46fe5fc561d9ba9e","source":{"kind":"arxiv","id":"2405.13967","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.13967","created_at":"2026-07-05T10:21:59Z"},{"alias_kind":"arxiv_version","alias_value":"2405.13967v5","created_at":"2026-07-05T10:21:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.13967","created_at":"2026-07-05T10:21:59Z"},{"alias_kind":"pith_short_12","alias_value":"N63GVVYCQTWA","created_at":"2026-07-05T10:21:59Z"},{"alias_kind":"pith_short_16","alias_value":"N63GVVYCQTWAHN6Q","created_at":"2026-07-05T10:21:59Z"},{"alias_kind":"pith_short_8","alias_value":"N63GVVYC","created_at":"2026-07-05T10:21:59Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:N63GVVYCQTWAHN6QFS3PWQZTRT","target":"record","payload":{"canonical_record":{"source":{"id":"2405.13967","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-22T20:08:48Z","cross_cats_sorted":[],"title_canon_sha256":"3d7372cc5bc96ca877ec9893bce61cc14b417a728739db58728a1579bac8a4f9","abstract_canon_sha256":"3b2d4d533d477e3722bbabf5a100f39d94968d1b7e031b9fff2b3106a94611fa"},"schema_version":"1.0"},"canonical_sha256":"6fb66ad70284ec03b7d02cb6fb43338ceabdb61f1efd65cd46fe5fc561d9ba9e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:21:59.876241Z","signature_b64":"n+LMB/6KWZseWkSzlDc4mbJUxJ9/VJQVC8+pZ5ViLmxTp9J57pDXBkKx+CGxXMpTx6sCxm5rDHqMd4BPwVxYCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6fb66ad70284ec03b7d02cb6fb43338ceabdb61f1efd65cd46fe5fc561d9ba9e","last_reissued_at":"2026-07-05T10:21:59.875748Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:21:59.875748Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.13967","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:21:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CpCPkeJVEJM/Op7iAxBjpqvgWWdJy1YSKfvmm4KXmmq3CwVaYHrHba57wSwakUYQdYhphClv5FDBB3ZMVVINAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-20T10:01:36.421419Z"},"content_sha256":"9cf39318b05ad7b023791bb84e8c0f6167c83c41aa9bcfd912ca5204a2b32a6e","schema_version":"1.0","event_id":"sha256:9cf39318b05ad7b023791bb84e8c0f6167c83c41aa9bcfd912ca5204a2b32a6e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:N63GVVYCQTWAHN6QFS3PWQZTRT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Model Editing as a Robust and Denoised variant of DPO: A Case Study on Toxicity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Apratim Dey, Junjie Hu, Rheeya Uppaal, Yiqiao Zhong, Yiting He","submitted_at":"2024-05-22T20:08:48Z","abstract_excerpt":"Recent alignment algorithms such as direct preference optimization (DPO) have been developed to improve the safety of large language models (LLMs) by training these models to match human behaviors exemplified by preference data. However, these methods are both computationally intensive and lacking in controllability and transparency, inhibiting their widespread use. Furthermore, these tuning-based methods require large-scale preference data for training and are susceptible to noisy preference data. In this paper, we introduce a tuning-free alignment alternative, ProFS (Projection Filter for Su"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.13967","kind":"arxiv","version":5},"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/2405.13967/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":"2026-07-05T10:21:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lmKQXmdIYbBOGwT7vtXXAEmTetVNUcScoKIXsxLB/fEssmHku0fMFlqkcP0vOQs+lSRAv9JqCJR2dhnGFtqpCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-20T10:01:36.421806Z"},"content_sha256":"2218b0d3efe796efb095743a862759f6b7a369011843eabcda1c9b8806849610","schema_version":"1.0","event_id":"sha256:2218b0d3efe796efb095743a862759f6b7a369011843eabcda1c9b8806849610"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/N63GVVYCQTWAHN6QFS3PWQZTRT/bundle.json","state_url":"https://pith.science/pith/N63GVVYCQTWAHN6QFS3PWQZTRT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/N63GVVYCQTWAHN6QFS3PWQZTRT/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":"2026-07-20T10:01:36Z","links":{"resolver":"https://pith.science/pith/N63GVVYCQTWAHN6QFS3PWQZTRT","bundle":"https://pith.science/pith/N63GVVYCQTWAHN6QFS3PWQZTRT/bundle.json","state":"https://pith.science/pith/N63GVVYCQTWAHN6QFS3PWQZTRT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/N63GVVYCQTWAHN6QFS3PWQZTRT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:N63GVVYCQTWAHN6QFS3PWQZTRT","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":"3b2d4d533d477e3722bbabf5a100f39d94968d1b7e031b9fff2b3106a94611fa","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-22T20:08:48Z","title_canon_sha256":"3d7372cc5bc96ca877ec9893bce61cc14b417a728739db58728a1579bac8a4f9"},"schema_version":"1.0","source":{"id":"2405.13967","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.13967","created_at":"2026-07-05T10:21:59Z"},{"alias_kind":"arxiv_version","alias_value":"2405.13967v5","created_at":"2026-07-05T10:21:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.13967","created_at":"2026-07-05T10:21:59Z"},{"alias_kind":"pith_short_12","alias_value":"N63GVVYCQTWA","created_at":"2026-07-05T10:21:59Z"},{"alias_kind":"pith_short_16","alias_value":"N63GVVYCQTWAHN6Q","created_at":"2026-07-05T10:21:59Z"},{"alias_kind":"pith_short_8","alias_value":"N63GVVYC","created_at":"2026-07-05T10:21:59Z"}],"graph_snapshots":[{"event_id":"sha256:2218b0d3efe796efb095743a862759f6b7a369011843eabcda1c9b8806849610","target":"graph","created_at":"2026-07-05T10:21:59Z","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/2405.13967/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent alignment algorithms such as direct preference optimization (DPO) have been developed to improve the safety of large language models (LLMs) by training these models to match human behaviors exemplified by preference data. However, these methods are both computationally intensive and lacking in controllability and transparency, inhibiting their widespread use. Furthermore, these tuning-based methods require large-scale preference data for training and are susceptible to noisy preference data. In this paper, we introduce a tuning-free alignment alternative, ProFS (Projection Filter for Su","authors_text":"Apratim Dey, Junjie Hu, Rheeya Uppaal, Yiqiao Zhong, Yiting He","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-22T20:08:48Z","title":"Model Editing as a Robust and Denoised variant of DPO: A Case Study on Toxicity"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.13967","kind":"arxiv","version":5},"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":"sha256:9cf39318b05ad7b023791bb84e8c0f6167c83c41aa9bcfd912ca5204a2b32a6e","target":"record","created_at":"2026-07-05T10:21:59Z","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":"3b2d4d533d477e3722bbabf5a100f39d94968d1b7e031b9fff2b3106a94611fa","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-22T20:08:48Z","title_canon_sha256":"3d7372cc5bc96ca877ec9893bce61cc14b417a728739db58728a1579bac8a4f9"},"schema_version":"1.0","source":{"id":"2405.13967","kind":"arxiv","version":5}},"canonical_sha256":"6fb66ad70284ec03b7d02cb6fb43338ceabdb61f1efd65cd46fe5fc561d9ba9e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6fb66ad70284ec03b7d02cb6fb43338ceabdb61f1efd65cd46fe5fc561d9ba9e","first_computed_at":"2026-07-05T10:21:59.875748Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:21:59.875748Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"n+LMB/6KWZseWkSzlDc4mbJUxJ9/VJQVC8+pZ5ViLmxTp9J57pDXBkKx+CGxXMpTx6sCxm5rDHqMd4BPwVxYCw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:21:59.876241Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.13967","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9cf39318b05ad7b023791bb84e8c0f6167c83c41aa9bcfd912ca5204a2b32a6e","sha256:2218b0d3efe796efb095743a862759f6b7a369011843eabcda1c9b8806849610"],"state_sha256":"89326d24d2d57019c202175131ad3e87b4b11bc6509263ed1941c9f78c69436f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WGGLvKS9KcxKfym795L6McYZwaSaw6Nws84PTV/nb82I99DrlPJUIL/Ggx2Ixl/9gTHqqno/3u/k+omxNBM0BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-20T10:01:36.423969Z","bundle_sha256":"e5a83810d4b5430555695526eab3f35bca5f1958d6ebf236d5ef22c6d6830122"}}