{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:YHDDZ2BPBS2SFGMH76KKXVAV6O","short_pith_number":"pith:YHDDZ2BP","canonical_record":{"source":{"id":"2302.02453","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-02-05T18:35:21Z","cross_cats_sorted":["cs.CY"],"title_canon_sha256":"82db33bebd3e9a8ec7d3df206c95beeeb9d60711af8ea203cde5490fdec107af","abstract_canon_sha256":"b3a6d09cae8af5d7991999e51362a218ee18e2d5dea57dbc722b2d9861538043"},"schema_version":"1.0"},"canonical_sha256":"c1c63ce82f0cb5229987ff94abd415f3a466c68f7f634adcba7b1a6ebe77706a","source":{"kind":"arxiv","id":"2302.02453","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.02453","created_at":"2026-07-05T05:38:53Z"},{"alias_kind":"arxiv_version","alias_value":"2302.02453v1","created_at":"2026-07-05T05:38:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.02453","created_at":"2026-07-05T05:38:53Z"},{"alias_kind":"pith_short_12","alias_value":"YHDDZ2BPBS2S","created_at":"2026-07-05T05:38:53Z"},{"alias_kind":"pith_short_16","alias_value":"YHDDZ2BPBS2SFGMH","created_at":"2026-07-05T05:38:53Z"},{"alias_kind":"pith_short_8","alias_value":"YHDDZ2BP","created_at":"2026-07-05T05:38:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:YHDDZ2BPBS2SFGMH76KKXVAV6O","target":"record","payload":{"canonical_record":{"source":{"id":"2302.02453","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-02-05T18:35:21Z","cross_cats_sorted":["cs.CY"],"title_canon_sha256":"82db33bebd3e9a8ec7d3df206c95beeeb9d60711af8ea203cde5490fdec107af","abstract_canon_sha256":"b3a6d09cae8af5d7991999e51362a218ee18e2d5dea57dbc722b2d9861538043"},"schema_version":"1.0"},"canonical_sha256":"c1c63ce82f0cb5229987ff94abd415f3a466c68f7f634adcba7b1a6ebe77706a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:38:53.615227Z","signature_b64":"Maq5T6vb63Lb1Co/CxdsRzqaImH851QtoQ52bcfBsyAw+zNTDWLX+qjVoqTK2SbIZUL74lHW/uSSo1nxY42ZBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c1c63ce82f0cb5229987ff94abd415f3a466c68f7f634adcba7b1a6ebe77706a","last_reissued_at":"2026-07-05T05:38:53.614759Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:38:53.614759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2302.02453","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":"2026-07-05T05:38:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vxGLXKIsiVs9PO61pG1oEzWROwzX67+2c+BMNr7vO4W8ycQAh4YzwCp7pb5wkoLcPL743xUPYU8twnttenchAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T18:30:59.078555Z"},"content_sha256":"03abc9944cc350f35e524cfbdf8588e9797ee573e139a4ae7c848d50219160af","schema_version":"1.0","event_id":"sha256:03abc9944cc350f35e524cfbdf8588e9797ee573e139a4ae7c848d50219160af"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:YHDDZ2BPBS2SFGMH76KKXVAV6O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FineDeb: A Debiasing Framework for Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Akash Saravanan, Dhruv Mullick, Habibur Rahman, Nidhi Hegde","submitted_at":"2023-02-05T18:35:21Z","abstract_excerpt":"As language models are increasingly included in human-facing machine learning tools, bias against demographic subgroups has gained attention. We propose FineDeb, a two-phase debiasing framework for language models that starts with contextual debiasing of embeddings learned by pretrained language models. The model is then fine-tuned on a language modeling objective. Our results show that FineDeb offers stronger debiasing in comparison to other methods which often result in models as biased as the original language model. Our framework is generalizable for demographics with multiple classes, and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.02453","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/2302.02453/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":"2026-07-05T05:38:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CJMHSsDY5Xq0x3urDyXeke7TszizxfVkNsi4Ykgqclx/2PXOTu62fy1A5VUGlf/mxh3yhmYXhkdesfmXUQt9Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T18:30:59.079490Z"},"content_sha256":"913328889f0dfb7300bd7f12e047048fb532ae2766d2cc789dc88df733970534","schema_version":"1.0","event_id":"sha256:913328889f0dfb7300bd7f12e047048fb532ae2766d2cc789dc88df733970534"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YHDDZ2BPBS2SFGMH76KKXVAV6O/bundle.json","state_url":"https://pith.science/pith/YHDDZ2BPBS2SFGMH76KKXVAV6O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YHDDZ2BPBS2SFGMH76KKXVAV6O/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":"2026-08-07T18:30:59Z","links":{"resolver":"https://pith.science/pith/YHDDZ2BPBS2SFGMH76KKXVAV6O","bundle":"https://pith.science/pith/YHDDZ2BPBS2SFGMH76KKXVAV6O/bundle.json","state":"https://pith.science/pith/YHDDZ2BPBS2SFGMH76KKXVAV6O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YHDDZ2BPBS2SFGMH76KKXVAV6O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:YHDDZ2BPBS2SFGMH76KKXVAV6O","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":"b3a6d09cae8af5d7991999e51362a218ee18e2d5dea57dbc722b2d9861538043","cross_cats_sorted":["cs.CY"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-02-05T18:35:21Z","title_canon_sha256":"82db33bebd3e9a8ec7d3df206c95beeeb9d60711af8ea203cde5490fdec107af"},"schema_version":"1.0","source":{"id":"2302.02453","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.02453","created_at":"2026-07-05T05:38:53Z"},{"alias_kind":"arxiv_version","alias_value":"2302.02453v1","created_at":"2026-07-05T05:38:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.02453","created_at":"2026-07-05T05:38:53Z"},{"alias_kind":"pith_short_12","alias_value":"YHDDZ2BPBS2S","created_at":"2026-07-05T05:38:53Z"},{"alias_kind":"pith_short_16","alias_value":"YHDDZ2BPBS2SFGMH","created_at":"2026-07-05T05:38:53Z"},{"alias_kind":"pith_short_8","alias_value":"YHDDZ2BP","created_at":"2026-07-05T05:38:53Z"}],"graph_snapshots":[{"event_id":"sha256:913328889f0dfb7300bd7f12e047048fb532ae2766d2cc789dc88df733970534","target":"graph","created_at":"2026-07-05T05:38:53Z","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/2302.02453/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As language models are increasingly included in human-facing machine learning tools, bias against demographic subgroups has gained attention. We propose FineDeb, a two-phase debiasing framework for language models that starts with contextual debiasing of embeddings learned by pretrained language models. The model is then fine-tuned on a language modeling objective. Our results show that FineDeb offers stronger debiasing in comparison to other methods which often result in models as biased as the original language model. Our framework is generalizable for demographics with multiple classes, and","authors_text":"Akash Saravanan, Dhruv Mullick, Habibur Rahman, Nidhi Hegde","cross_cats":["cs.CY"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-02-05T18:35:21Z","title":"FineDeb: A Debiasing Framework for Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.02453","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":"sha256:03abc9944cc350f35e524cfbdf8588e9797ee573e139a4ae7c848d50219160af","target":"record","created_at":"2026-07-05T05:38:53Z","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":"b3a6d09cae8af5d7991999e51362a218ee18e2d5dea57dbc722b2d9861538043","cross_cats_sorted":["cs.CY"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-02-05T18:35:21Z","title_canon_sha256":"82db33bebd3e9a8ec7d3df206c95beeeb9d60711af8ea203cde5490fdec107af"},"schema_version":"1.0","source":{"id":"2302.02453","kind":"arxiv","version":1}},"canonical_sha256":"c1c63ce82f0cb5229987ff94abd415f3a466c68f7f634adcba7b1a6ebe77706a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c1c63ce82f0cb5229987ff94abd415f3a466c68f7f634adcba7b1a6ebe77706a","first_computed_at":"2026-07-05T05:38:53.614759Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:38:53.614759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Maq5T6vb63Lb1Co/CxdsRzqaImH851QtoQ52bcfBsyAw+zNTDWLX+qjVoqTK2SbIZUL74lHW/uSSo1nxY42ZBA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:38:53.615227Z","signed_message":"canonical_sha256_bytes"},"source_id":"2302.02453","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:03abc9944cc350f35e524cfbdf8588e9797ee573e139a4ae7c848d50219160af","sha256:913328889f0dfb7300bd7f12e047048fb532ae2766d2cc789dc88df733970534"],"state_sha256":"6b5c758dde94da485aedd153a514ab28ccbc4ab1ec769e48247c1e228d6ad855"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gsXjg9XgzYsYzOP8k0FALoyJ/6+lIE1ECp/nbfgEb4FuQyS9GpqX/EYUymkRFjRFpVea6HNTtzuAKyzWNGW5CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T18:30:59.091980Z","bundle_sha256":"818d8ef7204b1463f62de65bdaee992192897684e441363e399415273fcfa442"}}