{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:5BRF6DUJEHURKEY43LZK42XVDS","short_pith_number":"pith:5BRF6DUJ","canonical_record":{"source":{"id":"2303.10431","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-18T14:57:43Z","cross_cats_sorted":[],"title_canon_sha256":"8825f064adf1092bb7ee65f23f060cdef608abb105bf244b010c1a8a3965956d","abstract_canon_sha256":"db5a871654c83c89d9450b0d791e7c108dcaf751b81f22b93e06915da78c642c"},"schema_version":"1.0"},"canonical_sha256":"e8625f0e8921e915131cdaf2ae6af51c902c22108b3efbc6fac1ba07c12bfd5e","source":{"kind":"arxiv","id":"2303.10431","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.10431","created_at":"2026-07-05T05:52:32Z"},{"alias_kind":"arxiv_version","alias_value":"2303.10431v1","created_at":"2026-07-05T05:52:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.10431","created_at":"2026-07-05T05:52:32Z"},{"alias_kind":"pith_short_12","alias_value":"5BRF6DUJEHUR","created_at":"2026-07-05T05:52:32Z"},{"alias_kind":"pith_short_16","alias_value":"5BRF6DUJEHURKEY4","created_at":"2026-07-05T05:52:32Z"},{"alias_kind":"pith_short_8","alias_value":"5BRF6DUJ","created_at":"2026-07-05T05:52:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:5BRF6DUJEHURKEY43LZK42XVDS","target":"record","payload":{"canonical_record":{"source":{"id":"2303.10431","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-18T14:57:43Z","cross_cats_sorted":[],"title_canon_sha256":"8825f064adf1092bb7ee65f23f060cdef608abb105bf244b010c1a8a3965956d","abstract_canon_sha256":"db5a871654c83c89d9450b0d791e7c108dcaf751b81f22b93e06915da78c642c"},"schema_version":"1.0"},"canonical_sha256":"e8625f0e8921e915131cdaf2ae6af51c902c22108b3efbc6fac1ba07c12bfd5e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:52:32.225392Z","signature_b64":"bHI2Ov8jfD3NI00ilbEkvuT14BRW1hsTwad+RdCctwaReJjO8BxhFpx5S5AD4CeO+GL9GLE+uBB8f5ow3HzJCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e8625f0e8921e915131cdaf2ae6af51c902c22108b3efbc6fac1ba07c12bfd5e","last_reissued_at":"2026-07-05T05:52:32.225042Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:52:32.225042Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2303.10431","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":"2026-07-05T05:52:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CMV9fqaq1rgbiQk+c3BxDno5LRJfshXQZHnmBaaKp0qZfVJmDwf22irz+FhUE6qjznFLr6OfUStFkcLDSe1pAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T08:13:23.695696Z"},"content_sha256":"16971e57a73e3a54fae1a57ace145c424adf343c0f4efb1bf89695b6a312c681","schema_version":"1.0","event_id":"sha256:16971e57a73e3a54fae1a57ace145c424adf343c0f4efb1bf89695b6a312c681"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:5BRF6DUJEHURKEY43LZK42XVDS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DeAR: Debiasing Vision-Language Models with Additive Residuals","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ashish Seth, Chirag Agarwal, Mayur Hemani","submitted_at":"2023-03-18T14:57:43Z","abstract_excerpt":"Large pre-trained vision-language models (VLMs) reduce the time for developing predictive models for various vision-grounded language downstream tasks by providing rich, adaptable image and text representations. However, these models suffer from societal biases owing to the skewed distribution of various identity groups in the training data. These biases manifest as the skewed similarity between the representations for specific text concepts and images of people of different identity groups and, therefore, limit the usefulness of such models in real-world high-stakes applications. In this work"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.10431","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/2303.10431/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":"2026-07-05T05:52:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LNGxqsHhPW3MyrXO1TeUoI7mncpHoOn2rW3xc9qpEXSwh07TLz6tMXXrKscvSjytrtpBkIXP5OcyzO5jmBYYDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T08:13:23.696600Z"},"content_sha256":"2facb0a995cde9a06dbfb0f7b0467526b115cb67ca8ba1ae323b7aa8cbc1be46","schema_version":"1.0","event_id":"sha256:2facb0a995cde9a06dbfb0f7b0467526b115cb67ca8ba1ae323b7aa8cbc1be46"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5BRF6DUJEHURKEY43LZK42XVDS/bundle.json","state_url":"https://pith.science/pith/5BRF6DUJEHURKEY43LZK42XVDS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5BRF6DUJEHURKEY43LZK42XVDS/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":"2026-08-06T08:13:23Z","links":{"resolver":"https://pith.science/pith/5BRF6DUJEHURKEY43LZK42XVDS","bundle":"https://pith.science/pith/5BRF6DUJEHURKEY43LZK42XVDS/bundle.json","state":"https://pith.science/pith/5BRF6DUJEHURKEY43LZK42XVDS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5BRF6DUJEHURKEY43LZK42XVDS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:5BRF6DUJEHURKEY43LZK42XVDS","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":"db5a871654c83c89d9450b0d791e7c108dcaf751b81f22b93e06915da78c642c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-18T14:57:43Z","title_canon_sha256":"8825f064adf1092bb7ee65f23f060cdef608abb105bf244b010c1a8a3965956d"},"schema_version":"1.0","source":{"id":"2303.10431","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.10431","created_at":"2026-07-05T05:52:32Z"},{"alias_kind":"arxiv_version","alias_value":"2303.10431v1","created_at":"2026-07-05T05:52:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.10431","created_at":"2026-07-05T05:52:32Z"},{"alias_kind":"pith_short_12","alias_value":"5BRF6DUJEHUR","created_at":"2026-07-05T05:52:32Z"},{"alias_kind":"pith_short_16","alias_value":"5BRF6DUJEHURKEY4","created_at":"2026-07-05T05:52:32Z"},{"alias_kind":"pith_short_8","alias_value":"5BRF6DUJ","created_at":"2026-07-05T05:52:32Z"}],"graph_snapshots":[{"event_id":"sha256:2facb0a995cde9a06dbfb0f7b0467526b115cb67ca8ba1ae323b7aa8cbc1be46","target":"graph","created_at":"2026-07-05T05:52:32Z","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/2303.10431/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large pre-trained vision-language models (VLMs) reduce the time for developing predictive models for various vision-grounded language downstream tasks by providing rich, adaptable image and text representations. However, these models suffer from societal biases owing to the skewed distribution of various identity groups in the training data. These biases manifest as the skewed similarity between the representations for specific text concepts and images of people of different identity groups and, therefore, limit the usefulness of such models in real-world high-stakes applications. In this work","authors_text":"Ashish Seth, Chirag Agarwal, Mayur Hemani","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-18T14:57:43Z","title":"DeAR: Debiasing Vision-Language Models with Additive Residuals"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.10431","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":"sha256:16971e57a73e3a54fae1a57ace145c424adf343c0f4efb1bf89695b6a312c681","target":"record","created_at":"2026-07-05T05:52:32Z","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":"db5a871654c83c89d9450b0d791e7c108dcaf751b81f22b93e06915da78c642c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-18T14:57:43Z","title_canon_sha256":"8825f064adf1092bb7ee65f23f060cdef608abb105bf244b010c1a8a3965956d"},"schema_version":"1.0","source":{"id":"2303.10431","kind":"arxiv","version":1}},"canonical_sha256":"e8625f0e8921e915131cdaf2ae6af51c902c22108b3efbc6fac1ba07c12bfd5e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e8625f0e8921e915131cdaf2ae6af51c902c22108b3efbc6fac1ba07c12bfd5e","first_computed_at":"2026-07-05T05:52:32.225042Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:52:32.225042Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bHI2Ov8jfD3NI00ilbEkvuT14BRW1hsTwad+RdCctwaReJjO8BxhFpx5S5AD4CeO+GL9GLE+uBB8f5ow3HzJCg==","signature_status":"signed_v1","signed_at":"2026-07-05T05:52:32.225392Z","signed_message":"canonical_sha256_bytes"},"source_id":"2303.10431","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:16971e57a73e3a54fae1a57ace145c424adf343c0f4efb1bf89695b6a312c681","sha256:2facb0a995cde9a06dbfb0f7b0467526b115cb67ca8ba1ae323b7aa8cbc1be46"],"state_sha256":"a72a70a91e7692b29c9644aac6029cd7b4663625373810c23a17f50a5e765488"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6s+NS9gp0in8sWi2bmMZM8+AAX1/sBPT+M5Ltf5pdraHOExHN9/RXZv5GBPaQp77wHd6sbeyLAOtpcIIDOSGDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T08:13:23.702638Z","bundle_sha256":"62042221938b3e67c4db4be7bd8066afbd6560d1fb8beaacdf37d8e1c15c994f"}}