{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:FB2ZMCBAQ5M5DS4G63DQLJPXXD","short_pith_number":"pith:FB2ZMCBA","canonical_record":{"source":{"id":"2412.15275","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-12-17T19:08:22Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"8847c84e9a5839ebdac69886da69b4394fb50ae576e8a4bf5b5d64401df27f9d","abstract_canon_sha256":"b0b1a4504cd6b351b9ec2e421d46e8371276bb8d0f7f622ea5c66dfd96f029f0"},"schema_version":"1.0"},"canonical_sha256":"28759608208759d1cb86f6c705a5f7b8c2e2f8e662fc76ee54cfcb5dff171389","source":{"kind":"arxiv","id":"2412.15275","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.15275","created_at":"2026-07-05T09:52:00Z"},{"alias_kind":"arxiv_version","alias_value":"2412.15275v1","created_at":"2026-07-05T09:52:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15275","created_at":"2026-07-05T09:52:00Z"},{"alias_kind":"pith_short_12","alias_value":"FB2ZMCBAQ5M5","created_at":"2026-07-05T09:52:00Z"},{"alias_kind":"pith_short_16","alias_value":"FB2ZMCBAQ5M5DS4G","created_at":"2026-07-05T09:52:00Z"},{"alias_kind":"pith_short_8","alias_value":"FB2ZMCBA","created_at":"2026-07-05T09:52:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:FB2ZMCBAQ5M5DS4G63DQLJPXXD","target":"record","payload":{"canonical_record":{"source":{"id":"2412.15275","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-12-17T19:08:22Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"8847c84e9a5839ebdac69886da69b4394fb50ae576e8a4bf5b5d64401df27f9d","abstract_canon_sha256":"b0b1a4504cd6b351b9ec2e421d46e8371276bb8d0f7f622ea5c66dfd96f029f0"},"schema_version":"1.0"},"canonical_sha256":"28759608208759d1cb86f6c705a5f7b8c2e2f8e662fc76ee54cfcb5dff171389","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:52:00.975279Z","signature_b64":"Q+wdxHZ2723Qxa+jqQlFtTv/oONFA1D7h9Q9hEMNlRB/uXSM+KyVjZsN3Kgk8EgQLJpN4mdtttvNWu5beBHqAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"28759608208759d1cb86f6c705a5f7b8c2e2f8e662fc76ee54cfcb5dff171389","last_reissued_at":"2026-07-05T09:52:00.974840Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:52:00.974840Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.15275","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":"2026-07-05T09:52:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oI4mPDmm4ollpUSO+EtE3Zpq9zl0mdAppcII1201VXYZuUN/cyamOVyFRns3dx6mRoqe+GozXiTJnDOqxLCXAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T19:18:40.753785Z"},"content_sha256":"f8c6385a68519634c7df627473a26eb87641fa24cbc79f868df45045213be937","schema_version":"1.0","event_id":"sha256:f8c6385a68519634c7df627473a26eb87641fa24cbc79f868df45045213be937"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:FB2ZMCBAQ5M5DS4G63DQLJPXXD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Fooling LLM graders into giving better grades through neural activity guided adversarial prompting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CR","authors_text":"Atsushi Yamamura, Surya Ganguli","submitted_at":"2024-12-17T19:08:22Z","abstract_excerpt":"The deployment of artificial intelligence (AI) in critical decision-making and evaluation processes raises concerns about inherent biases that malicious actors could exploit to distort decision outcomes. We propose a systematic method to reveal such biases in AI evaluation systems and apply it to automated essay grading as an example. Our approach first identifies hidden neural activity patterns that predict distorted decision outcomes and then optimizes an adversarial input suffix to amplify such patterns. We demonstrate that this combination can effectively fool large language model (LLM) gr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15275","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/2412.15275/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":"2026-07-05T09:52:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4WSILt3Bsc9bbY1I6MZ6OqMbTfn+ymVCPD5UaaT/RX0y9jBG2zkozTEUh8ETM6W3T3a4XWhn2ykL5bTCRe2qDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T19:18:40.754395Z"},"content_sha256":"d40f03349e5a55637a649679e8b40b01a98066b3d01aad1c5ddcf081037b9bc6","schema_version":"1.0","event_id":"sha256:d40f03349e5a55637a649679e8b40b01a98066b3d01aad1c5ddcf081037b9bc6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FB2ZMCBAQ5M5DS4G63DQLJPXXD/bundle.json","state_url":"https://pith.science/pith/FB2ZMCBAQ5M5DS4G63DQLJPXXD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FB2ZMCBAQ5M5DS4G63DQLJPXXD/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":"2026-07-12T19:18:40Z","links":{"resolver":"https://pith.science/pith/FB2ZMCBAQ5M5DS4G63DQLJPXXD","bundle":"https://pith.science/pith/FB2ZMCBAQ5M5DS4G63DQLJPXXD/bundle.json","state":"https://pith.science/pith/FB2ZMCBAQ5M5DS4G63DQLJPXXD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FB2ZMCBAQ5M5DS4G63DQLJPXXD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:FB2ZMCBAQ5M5DS4G63DQLJPXXD","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":"b0b1a4504cd6b351b9ec2e421d46e8371276bb8d0f7f622ea5c66dfd96f029f0","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-12-17T19:08:22Z","title_canon_sha256":"8847c84e9a5839ebdac69886da69b4394fb50ae576e8a4bf5b5d64401df27f9d"},"schema_version":"1.0","source":{"id":"2412.15275","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.15275","created_at":"2026-07-05T09:52:00Z"},{"alias_kind":"arxiv_version","alias_value":"2412.15275v1","created_at":"2026-07-05T09:52:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15275","created_at":"2026-07-05T09:52:00Z"},{"alias_kind":"pith_short_12","alias_value":"FB2ZMCBAQ5M5","created_at":"2026-07-05T09:52:00Z"},{"alias_kind":"pith_short_16","alias_value":"FB2ZMCBAQ5M5DS4G","created_at":"2026-07-05T09:52:00Z"},{"alias_kind":"pith_short_8","alias_value":"FB2ZMCBA","created_at":"2026-07-05T09:52:00Z"}],"graph_snapshots":[{"event_id":"sha256:d40f03349e5a55637a649679e8b40b01a98066b3d01aad1c5ddcf081037b9bc6","target":"graph","created_at":"2026-07-05T09:52:00Z","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/2412.15275/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The deployment of artificial intelligence (AI) in critical decision-making and evaluation processes raises concerns about inherent biases that malicious actors could exploit to distort decision outcomes. We propose a systematic method to reveal such biases in AI evaluation systems and apply it to automated essay grading as an example. Our approach first identifies hidden neural activity patterns that predict distorted decision outcomes and then optimizes an adversarial input suffix to amplify such patterns. We demonstrate that this combination can effectively fool large language model (LLM) gr","authors_text":"Atsushi Yamamura, Surya Ganguli","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-12-17T19:08:22Z","title":"Fooling LLM graders into giving better grades through neural activity guided adversarial prompting"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15275","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:f8c6385a68519634c7df627473a26eb87641fa24cbc79f868df45045213be937","target":"record","created_at":"2026-07-05T09:52:00Z","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":"b0b1a4504cd6b351b9ec2e421d46e8371276bb8d0f7f622ea5c66dfd96f029f0","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-12-17T19:08:22Z","title_canon_sha256":"8847c84e9a5839ebdac69886da69b4394fb50ae576e8a4bf5b5d64401df27f9d"},"schema_version":"1.0","source":{"id":"2412.15275","kind":"arxiv","version":1}},"canonical_sha256":"28759608208759d1cb86f6c705a5f7b8c2e2f8e662fc76ee54cfcb5dff171389","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"28759608208759d1cb86f6c705a5f7b8c2e2f8e662fc76ee54cfcb5dff171389","first_computed_at":"2026-07-05T09:52:00.974840Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:52:00.974840Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Q+wdxHZ2723Qxa+jqQlFtTv/oONFA1D7h9Q9hEMNlRB/uXSM+KyVjZsN3Kgk8EgQLJpN4mdtttvNWu5beBHqAw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:52:00.975279Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.15275","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f8c6385a68519634c7df627473a26eb87641fa24cbc79f868df45045213be937","sha256:d40f03349e5a55637a649679e8b40b01a98066b3d01aad1c5ddcf081037b9bc6"],"state_sha256":"5c2defebe937d07db55245df78748de6b97d0f2da5a9bc61d78bb0570bc5c0dd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tHz6yJOedWn64b0Q9rNzrdcMmgCj7MSJm7hMzaQeXC258qm6MlRo7GtLyZ0PFj/641/EXSdVIls9q/cgexqYAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-12T19:18:40.756721Z","bundle_sha256":"b438192a7afca192222f39f554933b1e112f78e2f94299cc3e2ae42e78cf57d9"}}