{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:3ATN6BEHMK3KS46RDWZWV4WA7B","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":"ae7d3c4bc0e43c9c4224a8a3bcb9e526f59254d112e2f3d2714f9ac0d6414917","cross_cats_sorted":["cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-02T06:13:36Z","title_canon_sha256":"597da5cfa5f0c63eb49ec395866676c5cba81c2273b1c09fe733c551db2bc423"},"schema_version":"1.0","source":{"id":"2311.01011","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.01011","created_at":"2026-07-05T07:08:19Z"},{"alias_kind":"arxiv_version","alias_value":"2311.01011v1","created_at":"2026-07-05T07:08:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.01011","created_at":"2026-07-05T07:08:19Z"},{"alias_kind":"pith_short_12","alias_value":"3ATN6BEHMK3K","created_at":"2026-07-05T07:08:19Z"},{"alias_kind":"pith_short_16","alias_value":"3ATN6BEHMK3KS46R","created_at":"2026-07-05T07:08:19Z"},{"alias_kind":"pith_short_8","alias_value":"3ATN6BEH","created_at":"2026-07-05T07:08:19Z"}],"graph_snapshots":[{"event_id":"sha256:a9bf4171b468ad1c77734f2d350e84a7069ebf61479c5fe8c3d1d2ea31da00a8","target":"graph","created_at":"2026-07-05T07:08:19Z","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/2311.01011/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While Large Language Models (LLMs) are increasingly being used in real-world applications, they remain vulnerable to prompt injection attacks: malicious third party prompts that subvert the intent of the system designer. To help researchers study this problem, we present a dataset of over 126,000 prompt injection attacks and 46,000 prompt-based \"defenses\" against prompt injection, all created by players of an online game called Tensor Trust. To the best of our knowledge, this is currently the largest dataset of human-generated adversarial examples for instruction-following LLMs. The attacks in","authors_text":"Alan Ritter, Ethan Adrian Mendes, Isaac Ong, Justin Svegliato, Karim Elmaaroufi, Luke Bailey, Olivia Watkins, Pieter Abbeel, Sam Toyer, Stuart Russell, Tiffany Wang, Trevor Darrell","cross_cats":["cs.CR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-02T06:13:36Z","title":"Tensor Trust: Interpretable Prompt Injection Attacks from an Online Game"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.01011","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:f3088b7195922b3fa12819f4ced85bae8359dd2099ed3f80101d4fab3bafff55","target":"record","created_at":"2026-07-05T07:08:19Z","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":"ae7d3c4bc0e43c9c4224a8a3bcb9e526f59254d112e2f3d2714f9ac0d6414917","cross_cats_sorted":["cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-02T06:13:36Z","title_canon_sha256":"597da5cfa5f0c63eb49ec395866676c5cba81c2273b1c09fe733c551db2bc423"},"schema_version":"1.0","source":{"id":"2311.01011","kind":"arxiv","version":1}},"canonical_sha256":"d826df048762b6a973d11db36af2c0f852602b13484f0f2ac16523cc1238599c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d826df048762b6a973d11db36af2c0f852602b13484f0f2ac16523cc1238599c","first_computed_at":"2026-07-05T07:08:19.245902Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:08:19.245902Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9PdyaoJ+Fd4ZY8gDnyN9jx+kLDlEDEb6fJf7Lju8cwWG6AEMo0HmAgAxKhKfoZqXKSBj6Zs6CsxekY4PYYMtDw==","signature_status":"signed_v1","signed_at":"2026-07-05T07:08:19.246396Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.01011","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f3088b7195922b3fa12819f4ced85bae8359dd2099ed3f80101d4fab3bafff55","sha256:a9bf4171b468ad1c77734f2d350e84a7069ebf61479c5fe8c3d1d2ea31da00a8"],"state_sha256":"d9894912b199a1dc40efaf55a4dff733da92d9df4f3f525b77f4bc99f1ce3592"}