{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:GYG2KGIROP3JZK6FWR7TANXTET","short_pith_number":"pith:GYG2KGIR","canonical_record":{"source":{"id":"2501.11621","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-20T17:36:04Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3ac4c90cf16310df117a0126f06a12f443931c95a2f65d7010729cad3433112d","abstract_canon_sha256":"66dfec5de8c9d0471adc4f0055593136464e68a4255de98e714b63c05d12c630"},"schema_version":"1.0"},"canonical_sha256":"360da5191173f69cabc5b47f3036f324dac403287e3f5233ce5b8e3c2674924e","source":{"kind":"arxiv","id":"2501.11621","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.11621","created_at":"2026-07-05T10:03:11Z"},{"alias_kind":"arxiv_version","alias_value":"2501.11621v1","created_at":"2026-07-05T10:03:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.11621","created_at":"2026-07-05T10:03:11Z"},{"alias_kind":"pith_short_12","alias_value":"GYG2KGIROP3J","created_at":"2026-07-05T10:03:11Z"},{"alias_kind":"pith_short_16","alias_value":"GYG2KGIROP3JZK6F","created_at":"2026-07-05T10:03:11Z"},{"alias_kind":"pith_short_8","alias_value":"GYG2KGIR","created_at":"2026-07-05T10:03:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:GYG2KGIROP3JZK6FWR7TANXTET","target":"record","payload":{"canonical_record":{"source":{"id":"2501.11621","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-20T17:36:04Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3ac4c90cf16310df117a0126f06a12f443931c95a2f65d7010729cad3433112d","abstract_canon_sha256":"66dfec5de8c9d0471adc4f0055593136464e68a4255de98e714b63c05d12c630"},"schema_version":"1.0"},"canonical_sha256":"360da5191173f69cabc5b47f3036f324dac403287e3f5233ce5b8e3c2674924e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:03:11.648963Z","signature_b64":"2u11+Q7XKBS8KEgZLQOHW9oNcDBwVB/R1+p7gbpdtYQHYMBHYoVMBsm95SOjqFLeNHpeqdR7+LKMLHOftUJ1CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"360da5191173f69cabc5b47f3036f324dac403287e3f5233ce5b8e3c2674924e","last_reissued_at":"2026-07-05T10:03:11.648497Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:03:11.648497Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.11621","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":"2026-07-05T10:03:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qI+jslx8OvK/e/SHJzFQyjwvMsr0DJipw4ZVY5RR7R63om8w139k5DwlzlSl7YEYj+AvNYqseDHj30ZmVyRxCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T07:34:59.560013Z"},"content_sha256":"60638a73ef59a967768cb1e3ce2a59a19d4762ea5ab51c46428b233ca853e513","schema_version":"1.0","event_id":"sha256:60638a73ef59a967768cb1e3ce2a59a19d4762ea5ab51c46428b233ca853e513"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:GYG2KGIROP3JZK6FWR7TANXTET","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Trojan Detection Through Pattern Recognition for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Matthew Yudin, Rauf Izmailov, Razvan Stefanescu, Vedant Bhasin","submitted_at":"2025-01-20T17:36:04Z","abstract_excerpt":"Trojan backdoors can be injected into large language models at various stages, including pretraining, fine-tuning, and in-context learning, posing a significant threat to the model's alignment. Due to the nature of causal language modeling, detecting these triggers is challenging given the vast search space. In this study, we propose a multistage framework for detecting Trojan triggers in large language models consisting of token filtration, trigger identification, and trigger verification. We discuss existing trigger identification methods and propose two variants of a black-box trigger inver"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.11621","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/2501.11621/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":"2026-07-05T10:03:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tra9mfrgHz9pU8+RJt3yrPZEKcgTwHtl0khMHjgB2qwxZfB+FzM1bnCiCdSIWxejJcpqjvrsTVnepkvMD0/YDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T07:34:59.560631Z"},"content_sha256":"952fd1326aa31e81d7bf4d2225826d941a5a4c81ce5c86166136003370fcdca3","schema_version":"1.0","event_id":"sha256:952fd1326aa31e81d7bf4d2225826d941a5a4c81ce5c86166136003370fcdca3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GYG2KGIROP3JZK6FWR7TANXTET/bundle.json","state_url":"https://pith.science/pith/GYG2KGIROP3JZK6FWR7TANXTET/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GYG2KGIROP3JZK6FWR7TANXTET/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":"2026-08-11T07:34:59Z","links":{"resolver":"https://pith.science/pith/GYG2KGIROP3JZK6FWR7TANXTET","bundle":"https://pith.science/pith/GYG2KGIROP3JZK6FWR7TANXTET/bundle.json","state":"https://pith.science/pith/GYG2KGIROP3JZK6FWR7TANXTET/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GYG2KGIROP3JZK6FWR7TANXTET/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:GYG2KGIROP3JZK6FWR7TANXTET","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":"66dfec5de8c9d0471adc4f0055593136464e68a4255de98e714b63c05d12c630","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-20T17:36:04Z","title_canon_sha256":"3ac4c90cf16310df117a0126f06a12f443931c95a2f65d7010729cad3433112d"},"schema_version":"1.0","source":{"id":"2501.11621","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.11621","created_at":"2026-07-05T10:03:11Z"},{"alias_kind":"arxiv_version","alias_value":"2501.11621v1","created_at":"2026-07-05T10:03:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.11621","created_at":"2026-07-05T10:03:11Z"},{"alias_kind":"pith_short_12","alias_value":"GYG2KGIROP3J","created_at":"2026-07-05T10:03:11Z"},{"alias_kind":"pith_short_16","alias_value":"GYG2KGIROP3JZK6F","created_at":"2026-07-05T10:03:11Z"},{"alias_kind":"pith_short_8","alias_value":"GYG2KGIR","created_at":"2026-07-05T10:03:11Z"}],"graph_snapshots":[{"event_id":"sha256:952fd1326aa31e81d7bf4d2225826d941a5a4c81ce5c86166136003370fcdca3","target":"graph","created_at":"2026-07-05T10:03:11Z","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/2501.11621/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Trojan backdoors can be injected into large language models at various stages, including pretraining, fine-tuning, and in-context learning, posing a significant threat to the model's alignment. Due to the nature of causal language modeling, detecting these triggers is challenging given the vast search space. In this study, we propose a multistage framework for detecting Trojan triggers in large language models consisting of token filtration, trigger identification, and trigger verification. We discuss existing trigger identification methods and propose two variants of a black-box trigger inver","authors_text":"Matthew Yudin, Rauf Izmailov, Razvan Stefanescu, Vedant Bhasin","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-20T17:36:04Z","title":"Trojan Detection Through Pattern Recognition for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.11621","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":"sha256:60638a73ef59a967768cb1e3ce2a59a19d4762ea5ab51c46428b233ca853e513","target":"record","created_at":"2026-07-05T10:03:11Z","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":"66dfec5de8c9d0471adc4f0055593136464e68a4255de98e714b63c05d12c630","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-20T17:36:04Z","title_canon_sha256":"3ac4c90cf16310df117a0126f06a12f443931c95a2f65d7010729cad3433112d"},"schema_version":"1.0","source":{"id":"2501.11621","kind":"arxiv","version":1}},"canonical_sha256":"360da5191173f69cabc5b47f3036f324dac403287e3f5233ce5b8e3c2674924e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"360da5191173f69cabc5b47f3036f324dac403287e3f5233ce5b8e3c2674924e","first_computed_at":"2026-07-05T10:03:11.648497Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:03:11.648497Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2u11+Q7XKBS8KEgZLQOHW9oNcDBwVB/R1+p7gbpdtYQHYMBHYoVMBsm95SOjqFLeNHpeqdR7+LKMLHOftUJ1CQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:03:11.648963Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.11621","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:60638a73ef59a967768cb1e3ce2a59a19d4762ea5ab51c46428b233ca853e513","sha256:952fd1326aa31e81d7bf4d2225826d941a5a4c81ce5c86166136003370fcdca3"],"state_sha256":"f89c0ae5fcd08ec5ef5ec6f9f94f9e64eec5ac912946ee90f92673240e81c3d9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cWkwROJ7FrMojUMHeIweIhxtvX4XhWFYAd+xI3hYNjsxH5D642fRrT09uA0R8qihW2HzvYbIZ2OxmTVOE6YeDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T07:34:59.566515Z","bundle_sha256":"56137da226842dfc3e069a31b78539705e0d0e337e2bf3e1bf094c609fca2939"}}