{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:TSQKHCOPMAD5Y2UYDT5OGW2QSO","short_pith_number":"pith:TSQKHCOP","canonical_record":{"source":{"id":"2608.00622","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-01T12:27:13Z","cross_cats_sorted":[],"title_canon_sha256":"1bc1fd1551a997eb4fd0492e65e665672362fda22db0ccc66b7564d5f81587d2","abstract_canon_sha256":"43764b76a8061c8ecb3ebd92555ba68ab16cfe597e429d3d6ce9954d5635b979"},"schema_version":"1.0"},"canonical_sha256":"9ca0a389cf6007dc6a981cfae35b5093baa3fe944aab81e30dfb09e71fc5d964","source":{"kind":"arxiv","id":"2608.00622","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.00622","created_at":"2026-08-04T00:38:06Z"},{"alias_kind":"arxiv_version","alias_value":"2608.00622v1","created_at":"2026-08-04T00:38:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00622","created_at":"2026-08-04T00:38:06Z"},{"alias_kind":"pith_short_12","alias_value":"TSQKHCOPMAD5","created_at":"2026-08-04T00:38:06Z"},{"alias_kind":"pith_short_16","alias_value":"TSQKHCOPMAD5Y2UY","created_at":"2026-08-04T00:38:06Z"},{"alias_kind":"pith_short_8","alias_value":"TSQKHCOP","created_at":"2026-08-04T00:38:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:TSQKHCOPMAD5Y2UYDT5OGW2QSO","target":"record","payload":{"canonical_record":{"source":{"id":"2608.00622","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-01T12:27:13Z","cross_cats_sorted":[],"title_canon_sha256":"1bc1fd1551a997eb4fd0492e65e665672362fda22db0ccc66b7564d5f81587d2","abstract_canon_sha256":"43764b76a8061c8ecb3ebd92555ba68ab16cfe597e429d3d6ce9954d5635b979"},"schema_version":"1.0"},"canonical_sha256":"9ca0a389cf6007dc6a981cfae35b5093baa3fe944aab81e30dfb09e71fc5d964","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T00:38:06.915603Z","signature_b64":"V68tQ+Lb/Q3g7R1qAaEHa9u+aSlJlwlgzSdqpRDI0FA+woxfau1SHaa3tKX7fxjahqWw3iK8CELpBzQAqSLHCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ca0a389cf6007dc6a981cfae35b5093baa3fe944aab81e30dfb09e71fc5d964","last_reissued_at":"2026-08-04T00:38:06.914112Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T00:38:06.914112Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.00622","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":"2026-08-04T00:38:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"++3PUgo24AwPR5e98P0Xgm1U5eNCFonBaNaJ9ZoO8Bl3d05JnkNKtKb7dCoHsqqwFzVUQ6M27sbR7QLUG7nlBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T16:00:17.965709Z"},"content_sha256":"80aa1ec691a2bd4e7bcca6665bd3936f335affa85e3489f5706e11ddfe4839a0","schema_version":"1.0","event_id":"sha256:80aa1ec691a2bd4e7bcca6665bd3936f335affa85e3489f5706e11ddfe4839a0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:TSQKHCOPMAD5Y2UYDT5OGW2QSO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Heuristic Perspective on Debiasing Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chuancheng Shi, Guanglai Gao, Jiang Li, Tian Lan, Xiangdong Su, Xiangyu Wu, Yemin Wang, Yuan Wang, Zesheng Shi","submitted_at":"2026-08-01T12:27:13Z","abstract_excerpt":"Language models (LMs) often acquire various biases during pre-training and may express them in interactions, potentially causing social harm. Existing methods often rely on counterfactual augmentation or representation projection. These strategies remain limited in practice due to their high computational costs and difficulty in scaling to larger models. Additionally, many of these strategies require manual data annotation, narrowing their scope to specific cultures and bias categories. To overcome these limitations, we propose HEIMAT, a HEurIstic-style autoMATic debiasing framework for LMs. H"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00622","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/2608.00622/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":"2026-08-04T00:38:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xio1X80XMr81ABdq9a9nFMtv7nCC4wpHsfPdpGPiuFZ41PE4+FgyDIUskAyvZZab9iU0B4IwDEEIzzph/4iCBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T16:00:17.966045Z"},"content_sha256":"515e9702163a8408f9640dda8855879d801c974dd96c2cecc5586e334adeed0d","schema_version":"1.0","event_id":"sha256:515e9702163a8408f9640dda8855879d801c974dd96c2cecc5586e334adeed0d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TSQKHCOPMAD5Y2UYDT5OGW2QSO/bundle.json","state_url":"https://pith.science/pith/TSQKHCOPMAD5Y2UYDT5OGW2QSO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TSQKHCOPMAD5Y2UYDT5OGW2QSO/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":"2026-08-06T16:00:17Z","links":{"resolver":"https://pith.science/pith/TSQKHCOPMAD5Y2UYDT5OGW2QSO","bundle":"https://pith.science/pith/TSQKHCOPMAD5Y2UYDT5OGW2QSO/bundle.json","state":"https://pith.science/pith/TSQKHCOPMAD5Y2UYDT5OGW2QSO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TSQKHCOPMAD5Y2UYDT5OGW2QSO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TSQKHCOPMAD5Y2UYDT5OGW2QSO","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":"43764b76a8061c8ecb3ebd92555ba68ab16cfe597e429d3d6ce9954d5635b979","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-01T12:27:13Z","title_canon_sha256":"1bc1fd1551a997eb4fd0492e65e665672362fda22db0ccc66b7564d5f81587d2"},"schema_version":"1.0","source":{"id":"2608.00622","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.00622","created_at":"2026-08-04T00:38:06Z"},{"alias_kind":"arxiv_version","alias_value":"2608.00622v1","created_at":"2026-08-04T00:38:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00622","created_at":"2026-08-04T00:38:06Z"},{"alias_kind":"pith_short_12","alias_value":"TSQKHCOPMAD5","created_at":"2026-08-04T00:38:06Z"},{"alias_kind":"pith_short_16","alias_value":"TSQKHCOPMAD5Y2UY","created_at":"2026-08-04T00:38:06Z"},{"alias_kind":"pith_short_8","alias_value":"TSQKHCOP","created_at":"2026-08-04T00:38:06Z"}],"graph_snapshots":[{"event_id":"sha256:515e9702163a8408f9640dda8855879d801c974dd96c2cecc5586e334adeed0d","target":"graph","created_at":"2026-08-04T00:38:06Z","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/2608.00622/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models (LMs) often acquire various biases during pre-training and may express them in interactions, potentially causing social harm. Existing methods often rely on counterfactual augmentation or representation projection. These strategies remain limited in practice due to their high computational costs and difficulty in scaling to larger models. Additionally, many of these strategies require manual data annotation, narrowing their scope to specific cultures and bias categories. To overcome these limitations, we propose HEIMAT, a HEurIstic-style autoMATic debiasing framework for LMs. H","authors_text":"Chuancheng Shi, Guanglai Gao, Jiang Li, Tian Lan, Xiangdong Su, Xiangyu Wu, Yemin Wang, Yuan Wang, Zesheng Shi","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-01T12:27:13Z","title":"A Heuristic Perspective on Debiasing Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00622","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":"sha256:80aa1ec691a2bd4e7bcca6665bd3936f335affa85e3489f5706e11ddfe4839a0","target":"record","created_at":"2026-08-04T00:38:06Z","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":"43764b76a8061c8ecb3ebd92555ba68ab16cfe597e429d3d6ce9954d5635b979","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-01T12:27:13Z","title_canon_sha256":"1bc1fd1551a997eb4fd0492e65e665672362fda22db0ccc66b7564d5f81587d2"},"schema_version":"1.0","source":{"id":"2608.00622","kind":"arxiv","version":1}},"canonical_sha256":"9ca0a389cf6007dc6a981cfae35b5093baa3fe944aab81e30dfb09e71fc5d964","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9ca0a389cf6007dc6a981cfae35b5093baa3fe944aab81e30dfb09e71fc5d964","first_computed_at":"2026-08-04T00:38:06.914112Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-04T00:38:06.914112Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"V68tQ+Lb/Q3g7R1qAaEHa9u+aSlJlwlgzSdqpRDI0FA+woxfau1SHaa3tKX7fxjahqWw3iK8CELpBzQAqSLHCg==","signature_status":"signed_v1","signed_at":"2026-08-04T00:38:06.915603Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.00622","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:80aa1ec691a2bd4e7bcca6665bd3936f335affa85e3489f5706e11ddfe4839a0","sha256:515e9702163a8408f9640dda8855879d801c974dd96c2cecc5586e334adeed0d"],"state_sha256":"cdcf8fa75a09547d6205b75685651428714d6123c473793f27c2ac473132a41a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xEvyV4wGhVeznGAnxac8+yX3qyasAv2GTB8rQClbbSneTye9RFMVpIwpI8ERNQ0RGUbkjBHe95q8C+Ywzy3LAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T16:00:17.970156Z","bundle_sha256":"aca8503b3193a9f54fe7715724a0a2cd6609aa8b85c2bd23c5f8ed5fb4f41276"}}