{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:5SC3K3CALONNPGIRO465TP3SPI","short_pith_number":"pith:5SC3K3CA","schema_version":"1.0","canonical_sha256":"ec85b56c405b9ad79911773dd9bf727a109c990f52ccb6c329a7c4d3edbec7d7","source":{"kind":"arxiv","id":"2212.09171","version":2},"attestation_state":"computed","paper":{"title":"Rainproof: An Umbrella To Shield Text Generators From Out-Of-Distribution Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Maxime Darrin, Pablo Piantanida, Pierre Colombo","submitted_at":"2022-12-18T21:22:28Z","abstract_excerpt":"Implementing effective control mechanisms to ensure the proper functioning and security of deployed NLP models, from translation to chatbots, is essential. A key ingredient to ensure safe system behaviour is Out-Of-Distribution (OOD) detection, which aims to detect whether an input sample is statistically far from the training distribution. Although OOD detection is a widely covered topic in classification tasks, most methods rely on hidden features output by the encoder. In this work, we focus on leveraging soft-probabilities in a black-box framework, i.e. we can access the soft-predictions b"},"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":"2212.09171","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-12-18T21:22:28Z","cross_cats_sorted":[],"title_canon_sha256":"31645a4c6bd6e95d0c1087150fa44b31dd446ce196bdccc1bea1daf8743e6d7a","abstract_canon_sha256":"2aea5ffea15aac30af02b8f707b4730ba2adcbade03c73fe8bdf681afe61efd3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:50:48.593000Z","signature_b64":"oxR2eZiI8Z1e0MUsMT6Sclk8ZFYyIRN24esjToCDqFzBJ/ydtfM3YzznhNW/xqmxRlPhNZtqpjYKl2lk2rnxBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ec85b56c405b9ad79911773dd9bf727a109c990f52ccb6c329a7c4d3edbec7d7","last_reissued_at":"2026-07-05T07:50:48.592537Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:50:48.592537Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rainproof: An Umbrella To Shield Text Generators From Out-Of-Distribution Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Maxime Darrin, Pablo Piantanida, Pierre Colombo","submitted_at":"2022-12-18T21:22:28Z","abstract_excerpt":"Implementing effective control mechanisms to ensure the proper functioning and security of deployed NLP models, from translation to chatbots, is essential. A key ingredient to ensure safe system behaviour is Out-Of-Distribution (OOD) detection, which aims to detect whether an input sample is statistically far from the training distribution. Although OOD detection is a widely covered topic in classification tasks, most methods rely on hidden features output by the encoder. In this work, we focus on leveraging soft-probabilities in a black-box framework, i.e. we can access the soft-predictions b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.09171","kind":"arxiv","version":2},"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/2212.09171/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":"2212.09171","created_at":"2026-07-05T07:50:48.592596+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.09171v2","created_at":"2026-07-05T07:50:48.592596+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.09171","created_at":"2026-07-05T07:50:48.592596+00:00"},{"alias_kind":"pith_short_12","alias_value":"5SC3K3CALONN","created_at":"2026-07-05T07:50:48.592596+00:00"},{"alias_kind":"pith_short_16","alias_value":"5SC3K3CALONNPGIR","created_at":"2026-07-05T07:50:48.592596+00:00"},{"alias_kind":"pith_short_8","alias_value":"5SC3K3CA","created_at":"2026-07-05T07:50:48.592596+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.01033","citing_title":"SAUP: Situation Awareness Uncertainty Propagation on LLM Agent","ref_index":8,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5SC3K3CALONNPGIRO465TP3SPI","json":"https://pith.science/pith/5SC3K3CALONNPGIRO465TP3SPI.json","graph_json":"https://pith.science/api/pith-number/5SC3K3CALONNPGIRO465TP3SPI/graph.json","events_json":"https://pith.science/api/pith-number/5SC3K3CALONNPGIRO465TP3SPI/events.json","paper":"https://pith.science/paper/5SC3K3CA"},"agent_actions":{"view_html":"https://pith.science/pith/5SC3K3CALONNPGIRO465TP3SPI","download_json":"https://pith.science/pith/5SC3K3CALONNPGIRO465TP3SPI.json","view_paper":"https://pith.science/paper/5SC3K3CA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.09171&json=true","fetch_graph":"https://pith.science/api/pith-number/5SC3K3CALONNPGIRO465TP3SPI/graph.json","fetch_events":"https://pith.science/api/pith-number/5SC3K3CALONNPGIRO465TP3SPI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5SC3K3CALONNPGIRO465TP3SPI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5SC3K3CALONNPGIRO465TP3SPI/action/storage_attestation","attest_author":"https://pith.science/pith/5SC3K3CALONNPGIRO465TP3SPI/action/author_attestation","sign_citation":"https://pith.science/pith/5SC3K3CALONNPGIRO465TP3SPI/action/citation_signature","submit_replication":"https://pith.science/pith/5SC3K3CALONNPGIRO465TP3SPI/action/replication_record"}},"created_at":"2026-07-05T07:50:48.592596+00:00","updated_at":"2026-07-05T07:50:48.592596+00:00"}