{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:INWTMGGIIZ33A5PT2C32RVRKYR","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":"6c4ea3a068bca9b7ec79499ea2c6f4d2040c263bd85b8fe7f157110b754549cb","cross_cats_sorted":["cs.AI","cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-15T02:04:35Z","title_canon_sha256":"ed2b49def461ae45918c067e2517a5c25694fb2a918865276dcdaa1cef3e05a2"},"schema_version":"1.0","source":{"id":"2403.09963","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.09963","created_at":"2026-07-05T08:00:42Z"},{"alias_kind":"arxiv_version","alias_value":"2403.09963v2","created_at":"2026-07-05T08:00:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.09963","created_at":"2026-07-05T08:00:42Z"},{"alias_kind":"pith_short_12","alias_value":"INWTMGGIIZ33","created_at":"2026-07-05T08:00:42Z"},{"alias_kind":"pith_short_16","alias_value":"INWTMGGIIZ33A5PT","created_at":"2026-07-05T08:00:42Z"},{"alias_kind":"pith_short_8","alias_value":"INWTMGGI","created_at":"2026-07-05T08:00:42Z"}],"graph_snapshots":[{"event_id":"sha256:da69369fd396d9dd7ccd06971ec75956f45dbc16ced2ffed867cecab3d2c1768","target":"graph","created_at":"2026-07-05T08:00:42Z","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/2403.09963/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent research shows that pre-trained language models (PLMs) suffer from \"prompt bias\" in factual knowledge extraction, i.e., prompts tend to introduce biases toward specific labels. Prompt bias presents a significant challenge in assessing the factual knowledge within PLMs. Therefore, this paper aims to improve the reliability of existing benchmarks by thoroughly investigating and mitigating prompt bias. We show that: 1) all prompts in the experiments exhibit non-negligible bias, with gradient-based prompts like AutoPrompt and OptiPrompt displaying significantly higher levels of bias; 2) pro","authors_text":"Dacheng Tao, Keqin Peng, Liang Ding, Xiliang Lu, Ziyang Xu","cross_cats":["cs.AI","cs.IR"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-15T02:04:35Z","title":"Take Care of Your Prompt Bias! Investigating and Mitigating Prompt Bias in Factual Knowledge Extraction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.09963","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:a82bdc42a50034236868d57a7777eca0822fd228a3bca390f336a26275d0b615","target":"record","created_at":"2026-07-05T08:00:42Z","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":"6c4ea3a068bca9b7ec79499ea2c6f4d2040c263bd85b8fe7f157110b754549cb","cross_cats_sorted":["cs.AI","cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-15T02:04:35Z","title_canon_sha256":"ed2b49def461ae45918c067e2517a5c25694fb2a918865276dcdaa1cef3e05a2"},"schema_version":"1.0","source":{"id":"2403.09963","kind":"arxiv","version":2}},"canonical_sha256":"436d3618c84677b075f3d0b7a8d62ac444640aac52314d5c7ec06ce9d28dde76","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"436d3618c84677b075f3d0b7a8d62ac444640aac52314d5c7ec06ce9d28dde76","first_computed_at":"2026-07-05T08:00:42.280579Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:00:42.280579Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6tfmQwoMPYnB0myHRDIbn5JTlOmxjYwzPELm7SJv1H4jQccMtKo3CkoD58w3Mfp8nRsXePZGR2IIhTCyt09+Aw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:00:42.281060Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.09963","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a82bdc42a50034236868d57a7777eca0822fd228a3bca390f336a26275d0b615","sha256:da69369fd396d9dd7ccd06971ec75956f45dbc16ced2ffed867cecab3d2c1768"],"state_sha256":"953d0a54a4518e7eb865378a2a9d88cf76055554d9fa6af9efc16243a38f6ddf"}