{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:GEPWEMUMYT3EDSRJRMNCVOXPFU","short_pith_number":"pith:GEPWEMUM","canonical_record":{"source":{"id":"2306.17194","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2023-06-28T17:54:04Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"c5846856978301501a8fd4d7e7d2e0bc41e1f5e1111bd475378af632516eeeef","abstract_canon_sha256":"38f2d67a0635594e5ed023dc81fe18f6c6b2b0bd48de8ecd1165f84f18620a69"},"schema_version":"1.0"},"canonical_sha256":"311f62328cc4f641ca298b1a2abaef2d0ab92c2be96737b2553882241e0c91db","source":{"kind":"arxiv","id":"2306.17194","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.17194","created_at":"2026-07-05T07:06:30Z"},{"alias_kind":"arxiv_version","alias_value":"2306.17194v2","created_at":"2026-07-05T07:06:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.17194","created_at":"2026-07-05T07:06:30Z"},{"alias_kind":"pith_short_12","alias_value":"GEPWEMUMYT3E","created_at":"2026-07-05T07:06:30Z"},{"alias_kind":"pith_short_16","alias_value":"GEPWEMUMYT3EDSRJ","created_at":"2026-07-05T07:06:30Z"},{"alias_kind":"pith_short_8","alias_value":"GEPWEMUM","created_at":"2026-07-05T07:06:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:GEPWEMUMYT3EDSRJRMNCVOXPFU","target":"record","payload":{"canonical_record":{"source":{"id":"2306.17194","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2023-06-28T17:54:04Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"c5846856978301501a8fd4d7e7d2e0bc41e1f5e1111bd475378af632516eeeef","abstract_canon_sha256":"38f2d67a0635594e5ed023dc81fe18f6c6b2b0bd48de8ecd1165f84f18620a69"},"schema_version":"1.0"},"canonical_sha256":"311f62328cc4f641ca298b1a2abaef2d0ab92c2be96737b2553882241e0c91db","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:06:30.909498Z","signature_b64":"gCnurEZsAVcjH/DtnPni3TbfKYgpUWsArzL7mqqV8XAl3R9PTQaS216AWogTMqLuBLP6z+w2suPZktshtOV4BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"311f62328cc4f641ca298b1a2abaef2d0ab92c2be96737b2553882241e0c91db","last_reissued_at":"2026-07-05T07:06:30.909011Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:06:30.909011Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.17194","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":"2026-07-05T07:06:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NKGJXw/xV1tk4qUqlvKKqTk8KxyeQRSNeNtATzB/6bqhiMbxY8OT8xQ6sq7w/GHoxx5whERfRptD0O0TKjrMAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T14:49:46.708714Z"},"content_sha256":"f28e51a127ace00e09534b93cb547a6f561770c7cc0bb8e48d9938a597195f34","schema_version":"1.0","event_id":"sha256:f28e51a127ace00e09534b93cb547a6f561770c7cc0bb8e48d9938a597195f34"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:GEPWEMUMYT3EDSRJRMNCVOXPFU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the Exploitability of Instruction Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CR","authors_text":"Chaowei Xiao, Chen Zhu, Jiongxiao Wang, Jonas Geiping, Manli Shu, Tom Goldstein","submitted_at":"2023-06-28T17:54:04Z","abstract_excerpt":"Instruction tuning is an effective technique to align large language models (LLMs) with human intents. In this work, we investigate how an adversary can exploit instruction tuning by injecting specific instruction-following examples into the training data that intentionally changes the model's behavior. For example, an adversary can achieve content injection by injecting training examples that mention target content and eliciting such behavior from downstream models. To achieve this goal, we propose \\textit{AutoPoison}, an automated data poisoning pipeline. It naturally and coherently incorpor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.17194","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/2306.17194/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":"2026-07-05T07:06:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GV1dVj5BWf4RDNysgvWByBEPkHan1SH4iEMK8zvUVDnTPwAodlP1csa7jL294oCThTUszXEnZrqo4h4Ht3r/Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T14:49:46.709619Z"},"content_sha256":"1928fa5af9b6a48a552c857faa911d9822fa8446f5e90309b44c0b342ff186c3","schema_version":"1.0","event_id":"sha256:1928fa5af9b6a48a552c857faa911d9822fa8446f5e90309b44c0b342ff186c3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GEPWEMUMYT3EDSRJRMNCVOXPFU/bundle.json","state_url":"https://pith.science/pith/GEPWEMUMYT3EDSRJRMNCVOXPFU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GEPWEMUMYT3EDSRJRMNCVOXPFU/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":"2026-08-06T14:49:46Z","links":{"resolver":"https://pith.science/pith/GEPWEMUMYT3EDSRJRMNCVOXPFU","bundle":"https://pith.science/pith/GEPWEMUMYT3EDSRJRMNCVOXPFU/bundle.json","state":"https://pith.science/pith/GEPWEMUMYT3EDSRJRMNCVOXPFU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GEPWEMUMYT3EDSRJRMNCVOXPFU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:GEPWEMUMYT3EDSRJRMNCVOXPFU","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":"38f2d67a0635594e5ed023dc81fe18f6c6b2b0bd48de8ecd1165f84f18620a69","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2023-06-28T17:54:04Z","title_canon_sha256":"c5846856978301501a8fd4d7e7d2e0bc41e1f5e1111bd475378af632516eeeef"},"schema_version":"1.0","source":{"id":"2306.17194","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.17194","created_at":"2026-07-05T07:06:30Z"},{"alias_kind":"arxiv_version","alias_value":"2306.17194v2","created_at":"2026-07-05T07:06:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.17194","created_at":"2026-07-05T07:06:30Z"},{"alias_kind":"pith_short_12","alias_value":"GEPWEMUMYT3E","created_at":"2026-07-05T07:06:30Z"},{"alias_kind":"pith_short_16","alias_value":"GEPWEMUMYT3EDSRJ","created_at":"2026-07-05T07:06:30Z"},{"alias_kind":"pith_short_8","alias_value":"GEPWEMUM","created_at":"2026-07-05T07:06:30Z"}],"graph_snapshots":[{"event_id":"sha256:1928fa5af9b6a48a552c857faa911d9822fa8446f5e90309b44c0b342ff186c3","target":"graph","created_at":"2026-07-05T07:06:30Z","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/2306.17194/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Instruction tuning is an effective technique to align large language models (LLMs) with human intents. In this work, we investigate how an adversary can exploit instruction tuning by injecting specific instruction-following examples into the training data that intentionally changes the model's behavior. For example, an adversary can achieve content injection by injecting training examples that mention target content and eliciting such behavior from downstream models. To achieve this goal, we propose \\textit{AutoPoison}, an automated data poisoning pipeline. It naturally and coherently incorpor","authors_text":"Chaowei Xiao, Chen Zhu, Jiongxiao Wang, Jonas Geiping, Manli Shu, Tom Goldstein","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2023-06-28T17:54:04Z","title":"On the Exploitability of Instruction Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.17194","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:f28e51a127ace00e09534b93cb547a6f561770c7cc0bb8e48d9938a597195f34","target":"record","created_at":"2026-07-05T07:06:30Z","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":"38f2d67a0635594e5ed023dc81fe18f6c6b2b0bd48de8ecd1165f84f18620a69","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2023-06-28T17:54:04Z","title_canon_sha256":"c5846856978301501a8fd4d7e7d2e0bc41e1f5e1111bd475378af632516eeeef"},"schema_version":"1.0","source":{"id":"2306.17194","kind":"arxiv","version":2}},"canonical_sha256":"311f62328cc4f641ca298b1a2abaef2d0ab92c2be96737b2553882241e0c91db","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"311f62328cc4f641ca298b1a2abaef2d0ab92c2be96737b2553882241e0c91db","first_computed_at":"2026-07-05T07:06:30.909011Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:06:30.909011Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gCnurEZsAVcjH/DtnPni3TbfKYgpUWsArzL7mqqV8XAl3R9PTQaS216AWogTMqLuBLP6z+w2suPZktshtOV4BA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:06:30.909498Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.17194","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f28e51a127ace00e09534b93cb547a6f561770c7cc0bb8e48d9938a597195f34","sha256:1928fa5af9b6a48a552c857faa911d9822fa8446f5e90309b44c0b342ff186c3"],"state_sha256":"767c4189230d9892d9be0d24f244820b3721f6883180cf0761b4db48152a80d1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+N2chQjOiyKPflIYrL269D2rfbFYdvmLbpUIt50C1zfTBcKjrzL83MgmNi06EXhZrgSTPkVmkGwESLO54yAZDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T14:49:46.713050Z","bundle_sha256":"2b93c730b644618c8c51a16e9749310500cfe9ab50c2f2f3490412446ead8b1e"}}