{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:T4KW4UGFFBYC7CRJDDM777FKYL","short_pith_number":"pith:T4KW4UGF","schema_version":"1.0","canonical_sha256":"9f156e50c528702f8a2918d9fffcaac2f3deb029935b064d70919cd6818607e1","source":{"kind":"arxiv","id":"2101.04840","version":1},"attestation_state":"computed","paper":{"title":"Robustness Gym: Unifying the NLP Evaluation Landscape","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Caiming Xiong, Christopher R\\'e, Jason Wu, Jesse Vig, Karan Goel, Mohit Bansal, Nazneen Rajani, Samson Tan, Stephan Zheng","submitted_at":"2021-01-13T02:37:54Z","abstract_excerpt":"Despite impressive performance on standard benchmarks, deep neural networks are often brittle when deployed in real-world systems. Consequently, recent research has focused on testing the robustness of such models, resulting in a diverse set of evaluation methodologies ranging from adversarial attacks to rule-based data transformations. In this work, we identify challenges with evaluating NLP systems and propose a solution in the form of Robustness Gym (RG), a simple and extensible evaluation toolkit that unifies 4 standard evaluation paradigms: subpopulations, transformations, evaluation sets"},"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":"2101.04840","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-01-13T02:37:54Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"47f9fecc24d696620042b1946a29a44c4630701c50fd305f9f2602b05694460d","abstract_canon_sha256":"d2dd40f4066f5779913d7b3a61134e0f43c244903be7e208017e46fa0f4fddda"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:06:35.620881Z","signature_b64":"baalJTel2HKRrsRZgyZ9LXplB2664rJhT8GfkiuTTgwdw1YJSHy2SSM5b+b7UG+oUx58wIyu3xHx40yixH15Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9f156e50c528702f8a2918d9fffcaac2f3deb029935b064d70919cd6818607e1","last_reissued_at":"2026-07-05T02:06:35.620471Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:06:35.620471Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Robustness Gym: Unifying the NLP Evaluation Landscape","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Caiming Xiong, Christopher R\\'e, Jason Wu, Jesse Vig, Karan Goel, Mohit Bansal, Nazneen Rajani, Samson Tan, Stephan Zheng","submitted_at":"2021-01-13T02:37:54Z","abstract_excerpt":"Despite impressive performance on standard benchmarks, deep neural networks are often brittle when deployed in real-world systems. Consequently, recent research has focused on testing the robustness of such models, resulting in a diverse set of evaluation methodologies ranging from adversarial attacks to rule-based data transformations. In this work, we identify challenges with evaluating NLP systems and propose a solution in the form of Robustness Gym (RG), a simple and extensible evaluation toolkit that unifies 4 standard evaluation paradigms: subpopulations, transformations, evaluation sets"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.04840","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/2101.04840/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":"2101.04840","created_at":"2026-07-05T02:06:35.620535+00:00"},{"alias_kind":"arxiv_version","alias_value":"2101.04840v1","created_at":"2026-07-05T02:06:35.620535+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.04840","created_at":"2026-07-05T02:06:35.620535+00:00"},{"alias_kind":"pith_short_12","alias_value":"T4KW4UGFFBYC","created_at":"2026-07-05T02:06:35.620535+00:00"},{"alias_kind":"pith_short_16","alias_value":"T4KW4UGFFBYC7CRJ","created_at":"2026-07-05T02:06:35.620535+00:00"},{"alias_kind":"pith_short_8","alias_value":"T4KW4UGF","created_at":"2026-07-05T02:06:35.620535+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22797","citing_title":"Measuring Behavior Portability in Large Language Models","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T4KW4UGFFBYC7CRJDDM777FKYL","json":"https://pith.science/pith/T4KW4UGFFBYC7CRJDDM777FKYL.json","graph_json":"https://pith.science/api/pith-number/T4KW4UGFFBYC7CRJDDM777FKYL/graph.json","events_json":"https://pith.science/api/pith-number/T4KW4UGFFBYC7CRJDDM777FKYL/events.json","paper":"https://pith.science/paper/T4KW4UGF"},"agent_actions":{"view_html":"https://pith.science/pith/T4KW4UGFFBYC7CRJDDM777FKYL","download_json":"https://pith.science/pith/T4KW4UGFFBYC7CRJDDM777FKYL.json","view_paper":"https://pith.science/paper/T4KW4UGF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2101.04840&json=true","fetch_graph":"https://pith.science/api/pith-number/T4KW4UGFFBYC7CRJDDM777FKYL/graph.json","fetch_events":"https://pith.science/api/pith-number/T4KW4UGFFBYC7CRJDDM777FKYL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T4KW4UGFFBYC7CRJDDM777FKYL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T4KW4UGFFBYC7CRJDDM777FKYL/action/storage_attestation","attest_author":"https://pith.science/pith/T4KW4UGFFBYC7CRJDDM777FKYL/action/author_attestation","sign_citation":"https://pith.science/pith/T4KW4UGFFBYC7CRJDDM777FKYL/action/citation_signature","submit_replication":"https://pith.science/pith/T4KW4UGFFBYC7CRJDDM777FKYL/action/replication_record"}},"created_at":"2026-07-05T02:06:35.620535+00:00","updated_at":"2026-07-05T02:06:35.620535+00:00"}