{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:6K3BKRE6LQVXL7HYRWDVNYYKTU","short_pith_number":"pith:6K3BKRE6","schema_version":"1.0","canonical_sha256":"f2b615449e5c2b75fcf88d8756e30a9d1bda1a121fa3e852b8b298343cd38e51","source":{"kind":"arxiv","id":"2210.12517","version":1},"attestation_state":"computed","paper":{"title":"Exploring The Landscape of Distributional Robustness for Question Answering Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Anas Awadalla, Gabriel Ilharco, Hannaneh Hajishirzi, Ian Magnusson, Ludwig Schmidt, Mitchell Wortsman, Sewon Min","submitted_at":"2022-10-22T18:17:31Z","abstract_excerpt":"We conduct a large empirical evaluation to investigate the landscape of distributional robustness in question answering. Our investigation spans over 350 models and 16 question answering datasets, including a diverse set of architectures, model sizes, and adaptation methods (e.g., fine-tuning, adapter tuning, in-context learning, etc.). We find that, in many cases, model variations do not affect robustness and in-distribution performance alone determines out-of-distribution performance. Moreover, our findings indicate that i) zero-shot and in-context learning methods are more robust to distrib"},"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":"2210.12517","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-22T18:17:31Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"14ee868255c33e8bc76d67c9bff5b63fb476b159ba2e330687a737515efdb063","abstract_canon_sha256":"95a3b82f67c14f62738d0e3745c5042051da10425819c7d210d44a255c58c5a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:09:29.344896Z","signature_b64":"aNmPWEcwcmbQVi6U87zPy2FCWGTjnNIqy30mrjoYY7UHkQprbeQSSfd0fuIzW093mx7DXv6l5bUU87aWBsQTDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f2b615449e5c2b75fcf88d8756e30a9d1bda1a121fa3e852b8b298343cd38e51","last_reissued_at":"2026-07-05T05:09:29.344490Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:09:29.344490Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Exploring The Landscape of Distributional Robustness for Question Answering Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Anas Awadalla, Gabriel Ilharco, Hannaneh Hajishirzi, Ian Magnusson, Ludwig Schmidt, Mitchell Wortsman, Sewon Min","submitted_at":"2022-10-22T18:17:31Z","abstract_excerpt":"We conduct a large empirical evaluation to investigate the landscape of distributional robustness in question answering. Our investigation spans over 350 models and 16 question answering datasets, including a diverse set of architectures, model sizes, and adaptation methods (e.g., fine-tuning, adapter tuning, in-context learning, etc.). We find that, in many cases, model variations do not affect robustness and in-distribution performance alone determines out-of-distribution performance. Moreover, our findings indicate that i) zero-shot and in-context learning methods are more robust to distrib"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.12517","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/2210.12517/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":"2210.12517","created_at":"2026-07-05T05:09:29.344547+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.12517v1","created_at":"2026-07-05T05:09:29.344547+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.12517","created_at":"2026-07-05T05:09:29.344547+00:00"},{"alias_kind":"pith_short_12","alias_value":"6K3BKRE6LQVX","created_at":"2026-07-05T05:09:29.344547+00:00"},{"alias_kind":"pith_short_16","alias_value":"6K3BKRE6LQVXL7HY","created_at":"2026-07-05T05:09:29.344547+00:00"},{"alias_kind":"pith_short_8","alias_value":"6K3BKRE6","created_at":"2026-07-05T05:09:29.344547+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.12120","citing_title":"LLMs on the Line: Data Determines Loss-to-Loss Scaling Laws","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6K3BKRE6LQVXL7HYRWDVNYYKTU","json":"https://pith.science/pith/6K3BKRE6LQVXL7HYRWDVNYYKTU.json","graph_json":"https://pith.science/api/pith-number/6K3BKRE6LQVXL7HYRWDVNYYKTU/graph.json","events_json":"https://pith.science/api/pith-number/6K3BKRE6LQVXL7HYRWDVNYYKTU/events.json","paper":"https://pith.science/paper/6K3BKRE6"},"agent_actions":{"view_html":"https://pith.science/pith/6K3BKRE6LQVXL7HYRWDVNYYKTU","download_json":"https://pith.science/pith/6K3BKRE6LQVXL7HYRWDVNYYKTU.json","view_paper":"https://pith.science/paper/6K3BKRE6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.12517&json=true","fetch_graph":"https://pith.science/api/pith-number/6K3BKRE6LQVXL7HYRWDVNYYKTU/graph.json","fetch_events":"https://pith.science/api/pith-number/6K3BKRE6LQVXL7HYRWDVNYYKTU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6K3BKRE6LQVXL7HYRWDVNYYKTU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6K3BKRE6LQVXL7HYRWDVNYYKTU/action/storage_attestation","attest_author":"https://pith.science/pith/6K3BKRE6LQVXL7HYRWDVNYYKTU/action/author_attestation","sign_citation":"https://pith.science/pith/6K3BKRE6LQVXL7HYRWDVNYYKTU/action/citation_signature","submit_replication":"https://pith.science/pith/6K3BKRE6LQVXL7HYRWDVNYYKTU/action/replication_record"}},"created_at":"2026-07-05T05:09:29.344547+00:00","updated_at":"2026-07-05T05:09:29.344547+00:00"}