{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TZPLPEXEHLYUEUTVEXXLXFEXW5","short_pith_number":"pith:TZPLPEXE","schema_version":"1.0","canonical_sha256":"9e5eb792e43af142527525eebb9497b7674ece66d20a19ae49945ff1143add32","source":{"kind":"arxiv","id":"2310.19177","version":1},"attestation_state":"computed","paper":{"title":"Robustifying Language Models with Test-Time Adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chengzhi Mao, Junfeng Yang, Noah Thomas McDermott","submitted_at":"2023-10-29T22:37:54Z","abstract_excerpt":"Large-scale language models achieved state-of-the-art performance over a number of language tasks. However, they fail on adversarial language examples, which are sentences optimized to fool the language models but with similar semantic meanings for humans. While prior work focuses on making the language model robust at training time, retraining for robustness is often unrealistic for large-scale foundation models. Instead, we propose to make the language models robust at test time. By dynamically adapting the input sentence with predictions from masked words, we show that we can reverse many l"},"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":"2310.19177","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-29T22:37:54Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"74c0883cf44360a86671b4ffb69a7c42fd8d363912cb0e2e034585546981e72b","abstract_canon_sha256":"b8771212775997beb86f466263b0e97460735ac973e6b99bbee5d065109806a6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:06:47.181242Z","signature_b64":"ybJ15GiMcAh82pYM294ZOof7cqWtllwYc1IkLX70UI3TTPS+ERvewLV0EhBnHH3m9R3spuebDSstAba+r1b3AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9e5eb792e43af142527525eebb9497b7674ece66d20a19ae49945ff1143add32","last_reissued_at":"2026-07-05T07:06:47.180765Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:06:47.180765Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Robustifying Language Models with Test-Time Adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chengzhi Mao, Junfeng Yang, Noah Thomas McDermott","submitted_at":"2023-10-29T22:37:54Z","abstract_excerpt":"Large-scale language models achieved state-of-the-art performance over a number of language tasks. However, they fail on adversarial language examples, which are sentences optimized to fool the language models but with similar semantic meanings for humans. While prior work focuses on making the language model robust at training time, retraining for robustness is often unrealistic for large-scale foundation models. Instead, we propose to make the language models robust at test time. By dynamically adapting the input sentence with predictions from masked words, we show that we can reverse many l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.19177","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/2310.19177/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":"2310.19177","created_at":"2026-07-05T07:06:47.180820+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.19177v1","created_at":"2026-07-05T07:06:47.180820+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.19177","created_at":"2026-07-05T07:06:47.180820+00:00"},{"alias_kind":"pith_short_12","alias_value":"TZPLPEXEHLYU","created_at":"2026-07-05T07:06:47.180820+00:00"},{"alias_kind":"pith_short_16","alias_value":"TZPLPEXEHLYUEUTV","created_at":"2026-07-05T07:06:47.180820+00:00"},{"alias_kind":"pith_short_8","alias_value":"TZPLPEXE","created_at":"2026-07-05T07:06:47.180820+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.17257","citing_title":"Agent Identity Evals: Measuring Agentic Identity","ref_index":78,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TZPLPEXEHLYUEUTVEXXLXFEXW5","json":"https://pith.science/pith/TZPLPEXEHLYUEUTVEXXLXFEXW5.json","graph_json":"https://pith.science/api/pith-number/TZPLPEXEHLYUEUTVEXXLXFEXW5/graph.json","events_json":"https://pith.science/api/pith-number/TZPLPEXEHLYUEUTVEXXLXFEXW5/events.json","paper":"https://pith.science/paper/TZPLPEXE"},"agent_actions":{"view_html":"https://pith.science/pith/TZPLPEXEHLYUEUTVEXXLXFEXW5","download_json":"https://pith.science/pith/TZPLPEXEHLYUEUTVEXXLXFEXW5.json","view_paper":"https://pith.science/paper/TZPLPEXE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.19177&json=true","fetch_graph":"https://pith.science/api/pith-number/TZPLPEXEHLYUEUTVEXXLXFEXW5/graph.json","fetch_events":"https://pith.science/api/pith-number/TZPLPEXEHLYUEUTVEXXLXFEXW5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TZPLPEXEHLYUEUTVEXXLXFEXW5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TZPLPEXEHLYUEUTVEXXLXFEXW5/action/storage_attestation","attest_author":"https://pith.science/pith/TZPLPEXEHLYUEUTVEXXLXFEXW5/action/author_attestation","sign_citation":"https://pith.science/pith/TZPLPEXEHLYUEUTVEXXLXFEXW5/action/citation_signature","submit_replication":"https://pith.science/pith/TZPLPEXEHLYUEUTVEXXLXFEXW5/action/replication_record"}},"created_at":"2026-07-05T07:06:47.180820+00:00","updated_at":"2026-07-05T07:06:47.180820+00:00"}