{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:OUJRPWKWVR3JNRREBW6LU6ZYPR","short_pith_number":"pith:OUJRPWKW","canonical_record":{"source":{"id":"2105.09293","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2021-05-13T23:18:41Z","cross_cats_sorted":[],"title_canon_sha256":"e778e3d136a7364875658be55225d2db39c2ffde394b47d341d5a81c9c89184c","abstract_canon_sha256":"2421aba6845157c4922f4558278555cbbc51fd29cf50b6a9ec0a3e832331e7f4"},"schema_version":"1.0"},"canonical_sha256":"751317d956ac7696c6240dbcba7b387c749b78b142d20f960623835b58261054","source":{"kind":"arxiv","id":"2105.09293","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2105.09293","created_at":"2026-07-05T02:41:39Z"},{"alias_kind":"arxiv_version","alias_value":"2105.09293v1","created_at":"2026-07-05T02:41:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.09293","created_at":"2026-07-05T02:41:39Z"},{"alias_kind":"pith_short_12","alias_value":"OUJRPWKWVR3J","created_at":"2026-07-05T02:41:39Z"},{"alias_kind":"pith_short_16","alias_value":"OUJRPWKWVR3JNRRE","created_at":"2026-07-05T02:41:39Z"},{"alias_kind":"pith_short_8","alias_value":"OUJRPWKW","created_at":"2026-07-05T02:41:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:OUJRPWKWVR3JNRREBW6LU6ZYPR","target":"record","payload":{"canonical_record":{"source":{"id":"2105.09293","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2021-05-13T23:18:41Z","cross_cats_sorted":[],"title_canon_sha256":"e778e3d136a7364875658be55225d2db39c2ffde394b47d341d5a81c9c89184c","abstract_canon_sha256":"2421aba6845157c4922f4558278555cbbc51fd29cf50b6a9ec0a3e832331e7f4"},"schema_version":"1.0"},"canonical_sha256":"751317d956ac7696c6240dbcba7b387c749b78b142d20f960623835b58261054","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:41:39.860101Z","signature_b64":"4Fs/Q3+fxUHu3buBOkbkanjfNi8xZQvSMEH7nvSfT9EEAtWkFkRkEmvQAWeUm2QDeJ1uiIsG4svSYNvDU9ooCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"751317d956ac7696c6240dbcba7b387c749b78b142d20f960623835b58261054","last_reissued_at":"2026-07-05T02:41:39.859638Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:41:39.859638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2105.09293","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":"2026-07-05T02:41:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SHgtcSVA1yc+GeX814BewCz2D9urYOexiC5WEs+OhrA9RQtfYktKQfmKW/ggWbXHU0qpOw+KtXN0wZ4lMGJPAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T20:59:49.475777Z"},"content_sha256":"bc2fe9e43fc21099b86cd7b2a48cf793c2684055a78a727cf5fb7bae69b1ac84","schema_version":"1.0","event_id":"sha256:bc2fe9e43fc21099b86cd7b2a48cf793c2684055a78a727cf5fb7bae69b1ac84"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:OUJRPWKWVR3JNRREBW6LU6ZYPR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Lessons Learned Addressing Dataset Bias in Model-Based Candidate Generation at Twitter","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Alim Virani, Apoorv Sharma, Chenguang Yu, Dan Shiebler, Jay Baxter, Linlin Chen, Philip Gautier, Shivam Verma, Sumit Binnani, Yan Xia","submitted_at":"2021-05-13T23:18:41Z","abstract_excerpt":"Traditionally, heuristic methods are used to generate candidates for large scale recommender systems. Model-based candidate generation promises multiple potential advantages, primarily that we can explicitly optimize the same objective as the downstream ranking model. However, large scale model-based candidate generation approaches suffer from dataset bias problems caused by the infeasibility of obtaining representative data on very irrelevant candidates. Popular techniques to correct dataset bias, such as inverse propensity scoring, do not work well in the context of candidate generation. We "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.09293","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/2105.09293/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":"2026-07-05T02:41:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2aY7yDNQ5j0zabSfCJXaZWz+cqgQoy+Y75cBi5LHIY8CYjfQfvvB5IQmQiFVDolO8K8ebye8HvGsoimhVU0dBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T20:59:49.477571Z"},"content_sha256":"b9e22a8c5608dfd182f8fceef007036074cf54eae196a20e72ab1c800774fbab","schema_version":"1.0","event_id":"sha256:b9e22a8c5608dfd182f8fceef007036074cf54eae196a20e72ab1c800774fbab"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OUJRPWKWVR3JNRREBW6LU6ZYPR/bundle.json","state_url":"https://pith.science/pith/OUJRPWKWVR3JNRREBW6LU6ZYPR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OUJRPWKWVR3JNRREBW6LU6ZYPR/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":"2026-08-04T20:59:49Z","links":{"resolver":"https://pith.science/pith/OUJRPWKWVR3JNRREBW6LU6ZYPR","bundle":"https://pith.science/pith/OUJRPWKWVR3JNRREBW6LU6ZYPR/bundle.json","state":"https://pith.science/pith/OUJRPWKWVR3JNRREBW6LU6ZYPR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OUJRPWKWVR3JNRREBW6LU6ZYPR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:OUJRPWKWVR3JNRREBW6LU6ZYPR","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":"2421aba6845157c4922f4558278555cbbc51fd29cf50b6a9ec0a3e832331e7f4","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2021-05-13T23:18:41Z","title_canon_sha256":"e778e3d136a7364875658be55225d2db39c2ffde394b47d341d5a81c9c89184c"},"schema_version":"1.0","source":{"id":"2105.09293","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2105.09293","created_at":"2026-07-05T02:41:39Z"},{"alias_kind":"arxiv_version","alias_value":"2105.09293v1","created_at":"2026-07-05T02:41:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.09293","created_at":"2026-07-05T02:41:39Z"},{"alias_kind":"pith_short_12","alias_value":"OUJRPWKWVR3J","created_at":"2026-07-05T02:41:39Z"},{"alias_kind":"pith_short_16","alias_value":"OUJRPWKWVR3JNRRE","created_at":"2026-07-05T02:41:39Z"},{"alias_kind":"pith_short_8","alias_value":"OUJRPWKW","created_at":"2026-07-05T02:41:39Z"}],"graph_snapshots":[{"event_id":"sha256:b9e22a8c5608dfd182f8fceef007036074cf54eae196a20e72ab1c800774fbab","target":"graph","created_at":"2026-07-05T02:41:39Z","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/2105.09293/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Traditionally, heuristic methods are used to generate candidates for large scale recommender systems. Model-based candidate generation promises multiple potential advantages, primarily that we can explicitly optimize the same objective as the downstream ranking model. However, large scale model-based candidate generation approaches suffer from dataset bias problems caused by the infeasibility of obtaining representative data on very irrelevant candidates. Popular techniques to correct dataset bias, such as inverse propensity scoring, do not work well in the context of candidate generation. We ","authors_text":"Alim Virani, Apoorv Sharma, Chenguang Yu, Dan Shiebler, Jay Baxter, Linlin Chen, Philip Gautier, Shivam Verma, Sumit Binnani, Yan Xia","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2021-05-13T23:18:41Z","title":"Lessons Learned Addressing Dataset Bias in Model-Based Candidate Generation at Twitter"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.09293","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:bc2fe9e43fc21099b86cd7b2a48cf793c2684055a78a727cf5fb7bae69b1ac84","target":"record","created_at":"2026-07-05T02:41:39Z","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":"2421aba6845157c4922f4558278555cbbc51fd29cf50b6a9ec0a3e832331e7f4","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2021-05-13T23:18:41Z","title_canon_sha256":"e778e3d136a7364875658be55225d2db39c2ffde394b47d341d5a81c9c89184c"},"schema_version":"1.0","source":{"id":"2105.09293","kind":"arxiv","version":1}},"canonical_sha256":"751317d956ac7696c6240dbcba7b387c749b78b142d20f960623835b58261054","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"751317d956ac7696c6240dbcba7b387c749b78b142d20f960623835b58261054","first_computed_at":"2026-07-05T02:41:39.859638Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:41:39.859638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4Fs/Q3+fxUHu3buBOkbkanjfNi8xZQvSMEH7nvSfT9EEAtWkFkRkEmvQAWeUm2QDeJ1uiIsG4svSYNvDU9ooCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T02:41:39.860101Z","signed_message":"canonical_sha256_bytes"},"source_id":"2105.09293","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bc2fe9e43fc21099b86cd7b2a48cf793c2684055a78a727cf5fb7bae69b1ac84","sha256:b9e22a8c5608dfd182f8fceef007036074cf54eae196a20e72ab1c800774fbab"],"state_sha256":"f5424dddd3818414df622f8f13ec850b38653e7adac7871684d0a7c6a3f1853a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XNM4ocFrJyfQLNTIKrUhY7w/zcd7Sdgoas2ZZIwFdxvrTRBTo4A7U866XYMzlwMHZmvo5Ih2FRdvcr4RoL0DDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T20:59:49.483907Z","bundle_sha256":"5f8cc8601c7a0fa24b878cc796789ca492b9cdee0d683856f96f84522663d35d"}}