{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:BPROKP4IQW2PZDBWPAAUIMDQJC","short_pith_number":"pith:BPROKP4I","canonical_record":{"source":{"id":"2008.09371","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-08-21T08:46:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fd6d4624b6d7006f58ccc1c64d31a4baa433d722ab46ecd9114d0d5f0a82b482","abstract_canon_sha256":"3a138a970f30f09142decececffdb0d783c533def121354e077cf769e03eef44"},"schema_version":"1.0"},"canonical_sha256":"0be2e53f8885b4fc8c36780144307048a64ce0a88ee0d59f4dfeed66325fb2fb","source":{"kind":"arxiv","id":"2008.09371","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2008.09371","created_at":"2026-07-05T04:12:14Z"},{"alias_kind":"arxiv_version","alias_value":"2008.09371v3","created_at":"2026-07-05T04:12:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.09371","created_at":"2026-07-05T04:12:14Z"},{"alias_kind":"pith_short_12","alias_value":"BPROKP4IQW2P","created_at":"2026-07-05T04:12:14Z"},{"alias_kind":"pith_short_16","alias_value":"BPROKP4IQW2PZDBW","created_at":"2026-07-05T04:12:14Z"},{"alias_kind":"pith_short_8","alias_value":"BPROKP4I","created_at":"2026-07-05T04:12:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:BPROKP4IQW2PZDBWPAAUIMDQJC","target":"record","payload":{"canonical_record":{"source":{"id":"2008.09371","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-08-21T08:46:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fd6d4624b6d7006f58ccc1c64d31a4baa433d722ab46ecd9114d0d5f0a82b482","abstract_canon_sha256":"3a138a970f30f09142decececffdb0d783c533def121354e077cf769e03eef44"},"schema_version":"1.0"},"canonical_sha256":"0be2e53f8885b4fc8c36780144307048a64ce0a88ee0d59f4dfeed66325fb2fb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:12:14.571286Z","signature_b64":"Ee3Hz+6oWd1tul6yzSmEAK49Qrbxqo8GE+PK/4gxt+GqXR+BihE1KzSb0ioOTnI8KvdAFb5b1bab7NKzBJfwAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0be2e53f8885b4fc8c36780144307048a64ce0a88ee0d59f4dfeed66325fb2fb","last_reissued_at":"2026-07-05T04:12:14.570754Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:12:14.570754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2008.09371","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:12:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"z2rviDPfg7Y7ukGi4+0b/S8zSPQJa8crbKD4J94A8h75s6u5tUVTt7dbl13Gg+5k8V3MmCLjdiOnLPKL511RCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-22T01:55:40.005839Z"},"content_sha256":"67e8d3f34438937f20244000df0b780675fda2faf82e706bd0139bf1b91248be","schema_version":"1.0","event_id":"sha256:67e8d3f34438937f20244000df0b780675fda2faf82e706bd0139bf1b91248be"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:BPROKP4IQW2PZDBWPAAUIMDQJC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Towards Improving Selective Prediction Ability of NLP Systems","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chitta Baral, Neeraj Varshney, Swaroop Mishra","submitted_at":"2020-08-21T08:46:36Z","abstract_excerpt":"It's better to say \"I can't answer\" than to answer incorrectly. This selective prediction ability is crucial for NLP systems to be reliably deployed in real-world applications. Prior work has shown that existing selective prediction techniques fail to perform well, especially in the out-of-domain setting. In this work, we propose a method that improves probability estimates of models by calibrating them using prediction confidence and difficulty score of instances. Using these two signals, we first annotate held-out instances and then train a calibrator to predict the likelihood of correctness"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.09371","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/2008.09371/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:12:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N2A5gyN3gwkK1QX9Oj+BEP2D3qxUwyXITaaheQKliPXG4aCcoBhIlKbkpGDb8SzBJSYHtRy1k2iwVUADe/pVAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-22T01:55:40.006201Z"},"content_sha256":"1dd96a57d0ea69437407daa6a793813b9a3971e72d67720575ff99736dc5757d","schema_version":"1.0","event_id":"sha256:1dd96a57d0ea69437407daa6a793813b9a3971e72d67720575ff99736dc5757d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BPROKP4IQW2PZDBWPAAUIMDQJC/bundle.json","state_url":"https://pith.science/pith/BPROKP4IQW2PZDBWPAAUIMDQJC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BPROKP4IQW2PZDBWPAAUIMDQJC/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-22T01:55:40Z","links":{"resolver":"https://pith.science/pith/BPROKP4IQW2PZDBWPAAUIMDQJC","bundle":"https://pith.science/pith/BPROKP4IQW2PZDBWPAAUIMDQJC/bundle.json","state":"https://pith.science/pith/BPROKP4IQW2PZDBWPAAUIMDQJC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BPROKP4IQW2PZDBWPAAUIMDQJC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:BPROKP4IQW2PZDBWPAAUIMDQJC","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"3a138a970f30f09142decececffdb0d783c533def121354e077cf769e03eef44","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-08-21T08:46:36Z","title_canon_sha256":"fd6d4624b6d7006f58ccc1c64d31a4baa433d722ab46ecd9114d0d5f0a82b482"},"schema_version":"1.0","source":{"id":"2008.09371","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2008.09371","created_at":"2026-07-05T04:12:14Z"},{"alias_kind":"arxiv_version","alias_value":"2008.09371v3","created_at":"2026-07-05T04:12:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.09371","created_at":"2026-07-05T04:12:14Z"},{"alias_kind":"pith_short_12","alias_value":"BPROKP4IQW2P","created_at":"2026-07-05T04:12:14Z"},{"alias_kind":"pith_short_16","alias_value":"BPROKP4IQW2PZDBW","created_at":"2026-07-05T04:12:14Z"},{"alias_kind":"pith_short_8","alias_value":"BPROKP4I","created_at":"2026-07-05T04:12:14Z"}],"graph_snapshots":[{"event_id":"sha256:1dd96a57d0ea69437407daa6a793813b9a3971e72d67720575ff99736dc5757d","target":"graph","created_at":"2026-07-05T04:12:14Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2008.09371/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"It's better to say \"I can't answer\" than to answer incorrectly. This selective prediction ability is crucial for NLP systems to be reliably deployed in real-world applications. Prior work has shown that existing selective prediction techniques fail to perform well, especially in the out-of-domain setting. In this work, we propose a method that improves probability estimates of models by calibrating them using prediction confidence and difficulty score of instances. Using these two signals, we first annotate held-out instances and then train a calibrator to predict the likelihood of correctness","authors_text":"Chitta Baral, Neeraj Varshney, Swaroop Mishra","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-08-21T08:46:36Z","title":"Towards Improving Selective Prediction Ability of NLP Systems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.09371","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:67e8d3f34438937f20244000df0b780675fda2faf82e706bd0139bf1b91248be","target":"record","created_at":"2026-07-05T04:12:14Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"3a138a970f30f09142decececffdb0d783c533def121354e077cf769e03eef44","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2020-08-21T08:46:36Z","title_canon_sha256":"fd6d4624b6d7006f58ccc1c64d31a4baa433d722ab46ecd9114d0d5f0a82b482"},"schema_version":"1.0","source":{"id":"2008.09371","kind":"arxiv","version":3}},"canonical_sha256":"0be2e53f8885b4fc8c36780144307048a64ce0a88ee0d59f4dfeed66325fb2fb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0be2e53f8885b4fc8c36780144307048a64ce0a88ee0d59f4dfeed66325fb2fb","first_computed_at":"2026-07-05T04:12:14.570754Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:12:14.570754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Ee3Hz+6oWd1tul6yzSmEAK49Qrbxqo8GE+PK/4gxt+GqXR+BihE1KzSb0ioOTnI8KvdAFb5b1bab7NKzBJfwAA==","signature_status":"signed_v1","signed_at":"2026-07-05T04:12:14.571286Z","signed_message":"canonical_sha256_bytes"},"source_id":"2008.09371","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:67e8d3f34438937f20244000df0b780675fda2faf82e706bd0139bf1b91248be","sha256:1dd96a57d0ea69437407daa6a793813b9a3971e72d67720575ff99736dc5757d"],"state_sha256":"d983459330c5c9be1a18daa60803cdef3ce7f608709b44db1a30d76ee4d4ce9f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gx5x69q0MoFagQuzx18qnNBjYKNaqLcPTulPh5APJz2aEdF1pEUJapF1XCYsCx1V79fJHTR7ixAg4GJCTzTzDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-22T01:55:40.008191Z","bundle_sha256":"84cc9ce0bec7509c88ac21b4765c140df5335fe199ddca6412713637aad469fe"}}