{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:6QZAOGURKHGT2VMPSPS55JO2J6","short_pith_number":"pith:6QZAOGUR","schema_version":"1.0","canonical_sha256":"f432071a9151cd3d558f93e5dea5da4fabce6c27d86741f05d27a083d7b52f9f","source":{"kind":"arxiv","id":"2205.07257","version":3},"attestation_state":"computed","paper":{"title":"Not to Overfit or Underfit the Source Domains? An Empirical Study of Domain Generalization in Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Avirup Sil, Md Arafat Sultan, Radu Florian","submitted_at":"2022-05-15T10:53:40Z","abstract_excerpt":"Machine learning models are prone to overfitting their training (source) domains, which is commonly believed to be the reason why they falter in novel target domains. Here we examine the contrasting view that multi-source domain generalization (DG) is first and foremost a problem of mitigating source domain underfitting: models not adequately learning the signal already present in their multi-domain training data. Experiments on a reading comprehension DG benchmark show that as a model learns its source domains better -- using familiar methods such as knowledge distillation (KD) from a bigger "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2205.07257","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-05-15T10:53:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"984c67f343bdc0c0ff7ec936d2bac0673becd82aa71cb3350553355af034c2e9","abstract_canon_sha256":"e17ab59e663d3ba1c1f4bcc759088fa259b40c8ac2f38ae35b8029a43a1d6bb4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:09:58.890109Z","signature_b64":"1bpw7bElrjBdrNiR2Zrxc5BuiQQlJOzCW5mMhpNhMEvYV2eq7TGyUeZjhjGgIXA8vlQYvQpVWe3OaOdOkwCeBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f432071a9151cd3d558f93e5dea5da4fabce6c27d86741f05d27a083d7b52f9f","last_reissued_at":"2026-07-05T05:09:58.889671Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:09:58.889671Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Not to Overfit or Underfit the Source Domains? An Empirical Study of Domain Generalization in Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Avirup Sil, Md Arafat Sultan, Radu Florian","submitted_at":"2022-05-15T10:53:40Z","abstract_excerpt":"Machine learning models are prone to overfitting their training (source) domains, which is commonly believed to be the reason why they falter in novel target domains. Here we examine the contrasting view that multi-source domain generalization (DG) is first and foremost a problem of mitigating source domain underfitting: models not adequately learning the signal already present in their multi-domain training data. Experiments on a reading comprehension DG benchmark show that as a model learns its source domains better -- using familiar methods such as knowledge distillation (KD) from a bigger "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.07257","kind":"arxiv","version":3},"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/2205.07257/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2205.07257","created_at":"2026-07-05T05:09:58.889719+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.07257v3","created_at":"2026-07-05T05:09:58.889719+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.07257","created_at":"2026-07-05T05:09:58.889719+00:00"},{"alias_kind":"pith_short_12","alias_value":"6QZAOGURKHGT","created_at":"2026-07-05T05:09:58.889719+00:00"},{"alias_kind":"pith_short_16","alias_value":"6QZAOGURKHGT2VMP","created_at":"2026-07-05T05:09:58.889719+00:00"},{"alias_kind":"pith_short_8","alias_value":"6QZAOGUR","created_at":"2026-07-05T05:09:58.889719+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6QZAOGURKHGT2VMPSPS55JO2J6","json":"https://pith.science/pith/6QZAOGURKHGT2VMPSPS55JO2J6.json","graph_json":"https://pith.science/api/pith-number/6QZAOGURKHGT2VMPSPS55JO2J6/graph.json","events_json":"https://pith.science/api/pith-number/6QZAOGURKHGT2VMPSPS55JO2J6/events.json","paper":"https://pith.science/paper/6QZAOGUR"},"agent_actions":{"view_html":"https://pith.science/pith/6QZAOGURKHGT2VMPSPS55JO2J6","download_json":"https://pith.science/pith/6QZAOGURKHGT2VMPSPS55JO2J6.json","view_paper":"https://pith.science/paper/6QZAOGUR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.07257&json=true","fetch_graph":"https://pith.science/api/pith-number/6QZAOGURKHGT2VMPSPS55JO2J6/graph.json","fetch_events":"https://pith.science/api/pith-number/6QZAOGURKHGT2VMPSPS55JO2J6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6QZAOGURKHGT2VMPSPS55JO2J6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6QZAOGURKHGT2VMPSPS55JO2J6/action/storage_attestation","attest_author":"https://pith.science/pith/6QZAOGURKHGT2VMPSPS55JO2J6/action/author_attestation","sign_citation":"https://pith.science/pith/6QZAOGURKHGT2VMPSPS55JO2J6/action/citation_signature","submit_replication":"https://pith.science/pith/6QZAOGURKHGT2VMPSPS55JO2J6/action/replication_record"}},"created_at":"2026-07-05T05:09:58.889719+00:00","updated_at":"2026-07-05T05:09:58.889719+00:00"}