{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MLP6TA6FJ7ORL37MEZNQPBVAU4","short_pith_number":"pith:MLP6TA6F","schema_version":"1.0","canonical_sha256":"62dfe983c54fdd15efec265b0786a0a7036fa836909d440fc33ee7466cdb7671","source":{"kind":"arxiv","id":"2607.18867","version":1},"attestation_state":"computed","paper":{"title":"HindsightBench: A Black-Box Behavioral Audit Protocol for Parametric Hindsight in Time-Indexed LLM Decision Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Haozhe Jia","submitted_at":"2026-07-21T08:58:36Z","abstract_excerpt":"Large language models leak parametric knowledge of realized outcomes into historical financial decision tasks. Existence is settled; what users lack is a cheap way to audit a given model for it. We present HindsightBench, a black-box behavioral audit protocol that profiles parametric hindsight in any time-indexed LLM decision task at probe-level cost (no backtests, no logprobs, no corpus access). The protocol chains a four-arm date-manipulation matrix (revealed/date-only/masked/transplanted), dual memory probes (date recovery; outcome recall), and six per-model metrics -- trigger strength, tra"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.18867","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-21T08:58:36Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"80b3408468586c874cca54585aee7560117f487cf25fb891feb59d6490f1b045","abstract_canon_sha256":"4771e313aa77e1e1e9c7cf239143b247438184588c44c40627530bd42f70ec02"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-22T01:23:16.602073Z","signature_b64":"li20eStPKHZd8yEuk6rNcWQUGbO17eUdTWaqvtMdZDhoa2zi0UcS+VZ2e8NRrKnigqYsX/c5QlttnxXMLnt4CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"62dfe983c54fdd15efec265b0786a0a7036fa836909d440fc33ee7466cdb7671","last_reissued_at":"2026-07-22T01:23:16.601210Z","signature_status":"signed_v1","first_computed_at":"2026-07-22T01:23:16.601210Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HindsightBench: A Black-Box Behavioral Audit Protocol for Parametric Hindsight in Time-Indexed LLM Decision Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Haozhe Jia","submitted_at":"2026-07-21T08:58:36Z","abstract_excerpt":"Large language models leak parametric knowledge of realized outcomes into historical financial decision tasks. Existence is settled; what users lack is a cheap way to audit a given model for it. We present HindsightBench, a black-box behavioral audit protocol that profiles parametric hindsight in any time-indexed LLM decision task at probe-level cost (no backtests, no logprobs, no corpus access). The protocol chains a four-arm date-manipulation matrix (revealed/date-only/masked/transplanted), dual memory probes (date recovery; outcome recall), and six per-model metrics -- trigger strength, tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.18867","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/2607.18867/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.18867","created_at":"2026-07-22T01:23:16.601656+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.18867v1","created_at":"2026-07-22T01:23:16.601656+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.18867","created_at":"2026-07-22T01:23:16.601656+00:00"},{"alias_kind":"pith_short_12","alias_value":"MLP6TA6FJ7OR","created_at":"2026-07-22T01:23:16.601656+00:00"},{"alias_kind":"pith_short_16","alias_value":"MLP6TA6FJ7ORL37M","created_at":"2026-07-22T01:23:16.601656+00:00"},{"alias_kind":"pith_short_8","alias_value":"MLP6TA6F","created_at":"2026-07-22T01:23:16.601656+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MLP6TA6FJ7ORL37MEZNQPBVAU4","json":"https://pith.science/pith/MLP6TA6FJ7ORL37MEZNQPBVAU4.json","graph_json":"https://pith.science/api/pith-number/MLP6TA6FJ7ORL37MEZNQPBVAU4/graph.json","events_json":"https://pith.science/api/pith-number/MLP6TA6FJ7ORL37MEZNQPBVAU4/events.json","paper":"https://pith.science/paper/MLP6TA6F"},"agent_actions":{"view_html":"https://pith.science/pith/MLP6TA6FJ7ORL37MEZNQPBVAU4","download_json":"https://pith.science/pith/MLP6TA6FJ7ORL37MEZNQPBVAU4.json","view_paper":"https://pith.science/paper/MLP6TA6F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.18867&json=true","fetch_graph":"https://pith.science/api/pith-number/MLP6TA6FJ7ORL37MEZNQPBVAU4/graph.json","fetch_events":"https://pith.science/api/pith-number/MLP6TA6FJ7ORL37MEZNQPBVAU4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MLP6TA6FJ7ORL37MEZNQPBVAU4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MLP6TA6FJ7ORL37MEZNQPBVAU4/action/storage_attestation","attest_author":"https://pith.science/pith/MLP6TA6FJ7ORL37MEZNQPBVAU4/action/author_attestation","sign_citation":"https://pith.science/pith/MLP6TA6FJ7ORL37MEZNQPBVAU4/action/citation_signature","submit_replication":"https://pith.science/pith/MLP6TA6FJ7ORL37MEZNQPBVAU4/action/replication_record"}},"created_at":"2026-07-22T01:23:16.601656+00:00","updated_at":"2026-07-22T01:23:16.601656+00:00"}