{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:7KN7P7CFX7VPTMVQQTK4NVKDXY","short_pith_number":"pith:7KN7P7CF","canonical_record":{"source":{"id":"2305.17804","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-28T19:36:50Z","cross_cats_sorted":[],"title_canon_sha256":"b33773aa579d61c569308ea12551176edb4a62a060c538ffc595dd30a84e750a","abstract_canon_sha256":"98a48f13f9272855a8b45432844276a7335c549a86af847068292341aff5116d"},"schema_version":"1.0"},"canonical_sha256":"fa9bf7fc45bfeaf9b2b084d5c6d543be1ac47bec554d2307dd7731bbad998cd5","source":{"kind":"arxiv","id":"2305.17804","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.17804","created_at":"2026-07-05T06:14:49Z"},{"alias_kind":"arxiv_version","alias_value":"2305.17804v1","created_at":"2026-07-05T06:14:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.17804","created_at":"2026-07-05T06:14:49Z"},{"alias_kind":"pith_short_12","alias_value":"7KN7P7CFX7VP","created_at":"2026-07-05T06:14:49Z"},{"alias_kind":"pith_short_16","alias_value":"7KN7P7CFX7VPTMVQ","created_at":"2026-07-05T06:14:49Z"},{"alias_kind":"pith_short_8","alias_value":"7KN7P7CF","created_at":"2026-07-05T06:14:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:7KN7P7CFX7VPTMVQQTK4NVKDXY","target":"record","payload":{"canonical_record":{"source":{"id":"2305.17804","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-28T19:36:50Z","cross_cats_sorted":[],"title_canon_sha256":"b33773aa579d61c569308ea12551176edb4a62a060c538ffc595dd30a84e750a","abstract_canon_sha256":"98a48f13f9272855a8b45432844276a7335c549a86af847068292341aff5116d"},"schema_version":"1.0"},"canonical_sha256":"fa9bf7fc45bfeaf9b2b084d5c6d543be1ac47bec554d2307dd7731bbad998cd5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:14:49.091656Z","signature_b64":"q+6MXM+81RNLhOIoCr2v1LqR9Me9bwHCzvQFWe7MxMUUomqnXPH2Xs1f6EkyySt++4il88PAB8EpXzo3lraaDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fa9bf7fc45bfeaf9b2b084d5c6d543be1ac47bec554d2307dd7731bbad998cd5","last_reissued_at":"2026-07-05T06:14:49.091321Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:14:49.091321Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2305.17804","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":"2026-07-05T06:14:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"utQd1azQXYb4+3C5+8g0Y+jwe45Pki0fkf7J7OTukfrg8i7ZCNQ34ikRtwOfUFfgaAiFmpepWNUwRC90lD9UDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T18:23:11.117527Z"},"content_sha256":"be985bbf8bb3bbe72fe700844dce3daac13f43675b14e29c547fde7af138a6e8","schema_version":"1.0","event_id":"sha256:be985bbf8bb3bbe72fe700844dce3daac13f43675b14e29c547fde7af138a6e8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:7KN7P7CFX7VPTMVQQTK4NVKDXY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Targeted Data Generation: Finding and Fixing Model Weaknesses","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fereshte Khani, Marco Tulio Ribeiro, Zexue He","submitted_at":"2023-05-28T19:36:50Z","abstract_excerpt":"Even when aggregate accuracy is high, state-of-the-art NLP models often fail systematically on specific subgroups of data, resulting in unfair outcomes and eroding user trust. Additional data collection may not help in addressing these weaknesses, as such challenging subgroups may be unknown to users, and underrepresented in the existing and new data. We propose Targeted Data Generation (TDG), a framework that automatically identifies challenging subgroups, and generates new data for those subgroups using large language models (LLMs) with a human in the loop. TDG estimates the expected benefit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.17804","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/2305.17804/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":"2026-07-05T06:14:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cX4271/ZAJJyabluxS24VTIWUie+YMG3wf+bZ3lpojGlJgtEC2mWl/tkYpzT+9N2bxl6l/LBChUGappFTUINCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T18:23:11.118138Z"},"content_sha256":"d23fcee6e0578945fa1aa3ef8a0dba41d55a9a3580f9338cdbd31ff1419cd296","schema_version":"1.0","event_id":"sha256:d23fcee6e0578945fa1aa3ef8a0dba41d55a9a3580f9338cdbd31ff1419cd296"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7KN7P7CFX7VPTMVQQTK4NVKDXY/bundle.json","state_url":"https://pith.science/pith/7KN7P7CFX7VPTMVQQTK4NVKDXY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7KN7P7CFX7VPTMVQQTK4NVKDXY/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":"2026-08-03T18:23:11Z","links":{"resolver":"https://pith.science/pith/7KN7P7CFX7VPTMVQQTK4NVKDXY","bundle":"https://pith.science/pith/7KN7P7CFX7VPTMVQQTK4NVKDXY/bundle.json","state":"https://pith.science/pith/7KN7P7CFX7VPTMVQQTK4NVKDXY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7KN7P7CFX7VPTMVQQTK4NVKDXY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:7KN7P7CFX7VPTMVQQTK4NVKDXY","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":"98a48f13f9272855a8b45432844276a7335c549a86af847068292341aff5116d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-28T19:36:50Z","title_canon_sha256":"b33773aa579d61c569308ea12551176edb4a62a060c538ffc595dd30a84e750a"},"schema_version":"1.0","source":{"id":"2305.17804","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.17804","created_at":"2026-07-05T06:14:49Z"},{"alias_kind":"arxiv_version","alias_value":"2305.17804v1","created_at":"2026-07-05T06:14:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.17804","created_at":"2026-07-05T06:14:49Z"},{"alias_kind":"pith_short_12","alias_value":"7KN7P7CFX7VP","created_at":"2026-07-05T06:14:49Z"},{"alias_kind":"pith_short_16","alias_value":"7KN7P7CFX7VPTMVQ","created_at":"2026-07-05T06:14:49Z"},{"alias_kind":"pith_short_8","alias_value":"7KN7P7CF","created_at":"2026-07-05T06:14:49Z"}],"graph_snapshots":[{"event_id":"sha256:d23fcee6e0578945fa1aa3ef8a0dba41d55a9a3580f9338cdbd31ff1419cd296","target":"graph","created_at":"2026-07-05T06:14:49Z","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/2305.17804/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Even when aggregate accuracy is high, state-of-the-art NLP models often fail systematically on specific subgroups of data, resulting in unfair outcomes and eroding user trust. Additional data collection may not help in addressing these weaknesses, as such challenging subgroups may be unknown to users, and underrepresented in the existing and new data. We propose Targeted Data Generation (TDG), a framework that automatically identifies challenging subgroups, and generates new data for those subgroups using large language models (LLMs) with a human in the loop. TDG estimates the expected benefit","authors_text":"Fereshte Khani, Marco Tulio Ribeiro, Zexue He","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-28T19:36:50Z","title":"Targeted Data Generation: Finding and Fixing Model Weaknesses"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.17804","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:be985bbf8bb3bbe72fe700844dce3daac13f43675b14e29c547fde7af138a6e8","target":"record","created_at":"2026-07-05T06:14:49Z","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":"98a48f13f9272855a8b45432844276a7335c549a86af847068292341aff5116d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-28T19:36:50Z","title_canon_sha256":"b33773aa579d61c569308ea12551176edb4a62a060c538ffc595dd30a84e750a"},"schema_version":"1.0","source":{"id":"2305.17804","kind":"arxiv","version":1}},"canonical_sha256":"fa9bf7fc45bfeaf9b2b084d5c6d543be1ac47bec554d2307dd7731bbad998cd5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fa9bf7fc45bfeaf9b2b084d5c6d543be1ac47bec554d2307dd7731bbad998cd5","first_computed_at":"2026-07-05T06:14:49.091321Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:14:49.091321Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"q+6MXM+81RNLhOIoCr2v1LqR9Me9bwHCzvQFWe7MxMUUomqnXPH2Xs1f6EkyySt++4il88PAB8EpXzo3lraaDg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:14:49.091656Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.17804","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:be985bbf8bb3bbe72fe700844dce3daac13f43675b14e29c547fde7af138a6e8","sha256:d23fcee6e0578945fa1aa3ef8a0dba41d55a9a3580f9338cdbd31ff1419cd296"],"state_sha256":"dc41f41d15cb80e53d902de41914408b15f3eeb68c51eebfbc93111132fb3655"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Xat7lqlTgGZGta7/QPews8eXMOk0QtPzc01UT2yno7zJkhIbPzirnlQm47yZ4HOcq8Do5WguRI2O9qllpvEzAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T18:23:11.124354Z","bundle_sha256":"487ba194f213a7e55c9555e4c15f78e3a3c7493c44a1ddecf4c4d7b0e56c1600"}}