{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:UP2A6MXFA4JD2Z2JVF6UACCIMN","short_pith_number":"pith:UP2A6MXF","canonical_record":{"source":{"id":"2605.28591","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T15:11:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cb53a1efe8fa3a1897bf15c9d0cbe4e2402b801c09077cfdcaf2391c5a9f1fbb","abstract_canon_sha256":"d32535f5454c65c831bd0dfb28a41b24992ff8dac576611b2803aaa86beff963"},"schema_version":"1.0"},"canonical_sha256":"a3f40f32e507123d6749a97d4008486372b5cf3fd507aa1f5da31c599f8b2953","source":{"kind":"arxiv","id":"2605.28591","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.28591","created_at":"2026-05-28T02:04:57Z"},{"alias_kind":"arxiv_version","alias_value":"2605.28591v1","created_at":"2026-05-28T02:04:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28591","created_at":"2026-05-28T02:04:57Z"},{"alias_kind":"pith_short_12","alias_value":"UP2A6MXFA4JD","created_at":"2026-05-28T02:04:57Z"},{"alias_kind":"pith_short_16","alias_value":"UP2A6MXFA4JD2Z2J","created_at":"2026-05-28T02:04:57Z"},{"alias_kind":"pith_short_8","alias_value":"UP2A6MXF","created_at":"2026-05-28T02:04:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:UP2A6MXFA4JD2Z2JVF6UACCIMN","target":"record","payload":{"canonical_record":{"source":{"id":"2605.28591","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T15:11:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cb53a1efe8fa3a1897bf15c9d0cbe4e2402b801c09077cfdcaf2391c5a9f1fbb","abstract_canon_sha256":"d32535f5454c65c831bd0dfb28a41b24992ff8dac576611b2803aaa86beff963"},"schema_version":"1.0"},"canonical_sha256":"a3f40f32e507123d6749a97d4008486372b5cf3fd507aa1f5da31c599f8b2953","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T02:04:57.264938Z","signature_b64":"XEGshiucUAfgljdTicLHuy/4kwKrnTcZcEq+C4HlxqzB2XN/RFjFasI5hqIZ0c0QemmQd4UUNj2+2TfVPu55BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a3f40f32e507123d6749a97d4008486372b5cf3fd507aa1f5da31c599f8b2953","last_reissued_at":"2026-05-28T02:04:57.264511Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T02:04:57.264511Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.28591","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":"2026-05-28T02:04:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F41c4937hOX0401ip56Ys4VGDsm+MLdEALPNsSz/Tegp2+115piYw+AcHnm0inkPJVlxsH8+m25WOOA7c0nyCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T21:03:37.200640Z"},"content_sha256":"5e73f9f3148e3d29df1977a309c892a1ad4d2eedbcc6ef09367d81cf50836c39","schema_version":"1.0","event_id":"sha256:5e73f9f3148e3d29df1977a309c892a1ad4d2eedbcc6ef09367d81cf50836c39"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:UP2A6MXFA4JD2Z2JVF6UACCIMN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Models That Know How Evaluations Are Designed Score Safer","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Haritz Puerto, Jonas Geiping, Katharina Deckenbach, Sahar Abdelnabi","submitted_at":"2026-05-27T15:11:35Z","abstract_excerpt":"The validity of AI safety evaluations depends on models behaving consistently across controlled and deployment settings. Prior work has identified test-time contextual cues, such as hypothetical scenarios, as a source of verbalized evaluation awareness and subsequent behavioral shift. In this paper, we investigate a potential explanation of this phenomenon: evaluation meta-knowledge, defined as parametric knowledge about the structural traits that characterize evaluations. Similar to dataset contamination, where benchmark exposure leads to higher performance through memorization, we hypothesiz"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28591","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/2605.28591/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":"2026-05-28T02:04:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZY00EOfyVubq67LXs/Tf/1I0yyBRgqt9yO2IrEqXh5nh1cYC699rUl33se6d7Uz2qkqonL986V3Gscwn5aBEAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T21:03:37.201399Z"},"content_sha256":"5471a3008839e80cb099873ce86f711bc9cbc7922c7eb503ec5036207d06426f","schema_version":"1.0","event_id":"sha256:5471a3008839e80cb099873ce86f711bc9cbc7922c7eb503ec5036207d06426f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UP2A6MXFA4JD2Z2JVF6UACCIMN/bundle.json","state_url":"https://pith.science/pith/UP2A6MXFA4JD2Z2JVF6UACCIMN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UP2A6MXFA4JD2Z2JVF6UACCIMN/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":"2026-06-06T21:03:37Z","links":{"resolver":"https://pith.science/pith/UP2A6MXFA4JD2Z2JVF6UACCIMN","bundle":"https://pith.science/pith/UP2A6MXFA4JD2Z2JVF6UACCIMN/bundle.json","state":"https://pith.science/pith/UP2A6MXFA4JD2Z2JVF6UACCIMN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UP2A6MXFA4JD2Z2JVF6UACCIMN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UP2A6MXFA4JD2Z2JVF6UACCIMN","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":"d32535f5454c65c831bd0dfb28a41b24992ff8dac576611b2803aaa86beff963","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T15:11:35Z","title_canon_sha256":"cb53a1efe8fa3a1897bf15c9d0cbe4e2402b801c09077cfdcaf2391c5a9f1fbb"},"schema_version":"1.0","source":{"id":"2605.28591","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.28591","created_at":"2026-05-28T02:04:57Z"},{"alias_kind":"arxiv_version","alias_value":"2605.28591v1","created_at":"2026-05-28T02:04:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28591","created_at":"2026-05-28T02:04:57Z"},{"alias_kind":"pith_short_12","alias_value":"UP2A6MXFA4JD","created_at":"2026-05-28T02:04:57Z"},{"alias_kind":"pith_short_16","alias_value":"UP2A6MXFA4JD2Z2J","created_at":"2026-05-28T02:04:57Z"},{"alias_kind":"pith_short_8","alias_value":"UP2A6MXF","created_at":"2026-05-28T02:04:57Z"}],"graph_snapshots":[{"event_id":"sha256:5471a3008839e80cb099873ce86f711bc9cbc7922c7eb503ec5036207d06426f","target":"graph","created_at":"2026-05-28T02:04:57Z","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/2605.28591/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The validity of AI safety evaluations depends on models behaving consistently across controlled and deployment settings. Prior work has identified test-time contextual cues, such as hypothetical scenarios, as a source of verbalized evaluation awareness and subsequent behavioral shift. In this paper, we investigate a potential explanation of this phenomenon: evaluation meta-knowledge, defined as parametric knowledge about the structural traits that characterize evaluations. Similar to dataset contamination, where benchmark exposure leads to higher performance through memorization, we hypothesiz","authors_text":"Haritz Puerto, Jonas Geiping, Katharina Deckenbach, Sahar Abdelnabi","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T15:11:35Z","title":"Models That Know How Evaluations Are Designed Score Safer"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28591","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":"sha256:5e73f9f3148e3d29df1977a309c892a1ad4d2eedbcc6ef09367d81cf50836c39","target":"record","created_at":"2026-05-28T02:04:57Z","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":"d32535f5454c65c831bd0dfb28a41b24992ff8dac576611b2803aaa86beff963","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T15:11:35Z","title_canon_sha256":"cb53a1efe8fa3a1897bf15c9d0cbe4e2402b801c09077cfdcaf2391c5a9f1fbb"},"schema_version":"1.0","source":{"id":"2605.28591","kind":"arxiv","version":1}},"canonical_sha256":"a3f40f32e507123d6749a97d4008486372b5cf3fd507aa1f5da31c599f8b2953","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a3f40f32e507123d6749a97d4008486372b5cf3fd507aa1f5da31c599f8b2953","first_computed_at":"2026-05-28T02:04:57.264511Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T02:04:57.264511Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XEGshiucUAfgljdTicLHuy/4kwKrnTcZcEq+C4HlxqzB2XN/RFjFasI5hqIZ0c0QemmQd4UUNj2+2TfVPu55BA==","signature_status":"signed_v1","signed_at":"2026-05-28T02:04:57.264938Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.28591","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5e73f9f3148e3d29df1977a309c892a1ad4d2eedbcc6ef09367d81cf50836c39","sha256:5471a3008839e80cb099873ce86f711bc9cbc7922c7eb503ec5036207d06426f"],"state_sha256":"a42c523a44c85b116e9df24b54571f7938a5da4907995b6febed0a48ae7317dd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3zQBVRMlhiMGQavlEJ03jLvn3aU/PmZ597VS/9v771WNN8mvVPQ6+iuNck28BMTZ/Jvfv2EzparyuqH4v9BlBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-06T21:03:37.205108Z","bundle_sha256":"1162929ffe48302a1a47ef89d8510a2565062601264f47e648914aa143e32fea"}}