{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:5LZPZHSVHXBKWI5FWWZXGXCA7O","short_pith_number":"pith:5LZPZHSV","canonical_record":{"source":{"id":"2104.02145","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-05T20:36:11Z","cross_cats_sorted":[],"title_canon_sha256":"26ef90efc21489307c3471cd7281ed3e8dd1c41dd229742e2a8ca03f6106a117","abstract_canon_sha256":"0f8f89fb2776d6a62926c97f953bf638cb6077eac93b4d70acf051c1c8c31444"},"schema_version":"1.0"},"canonical_sha256":"eaf2fc9e553dc2ab23a5b5b3735c40fba312f271715110d696bb5b367a984069","source":{"kind":"arxiv","id":"2104.02145","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2104.02145","created_at":"2026-07-05T03:23:09Z"},{"alias_kind":"arxiv_version","alias_value":"2104.02145v3","created_at":"2026-07-05T03:23:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.02145","created_at":"2026-07-05T03:23:09Z"},{"alias_kind":"pith_short_12","alias_value":"5LZPZHSVHXBK","created_at":"2026-07-05T03:23:09Z"},{"alias_kind":"pith_short_16","alias_value":"5LZPZHSVHXBKWI5F","created_at":"2026-07-05T03:23:09Z"},{"alias_kind":"pith_short_8","alias_value":"5LZPZHSV","created_at":"2026-07-05T03:23:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:5LZPZHSVHXBKWI5FWWZXGXCA7O","target":"record","payload":{"canonical_record":{"source":{"id":"2104.02145","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-05T20:36:11Z","cross_cats_sorted":[],"title_canon_sha256":"26ef90efc21489307c3471cd7281ed3e8dd1c41dd229742e2a8ca03f6106a117","abstract_canon_sha256":"0f8f89fb2776d6a62926c97f953bf638cb6077eac93b4d70acf051c1c8c31444"},"schema_version":"1.0"},"canonical_sha256":"eaf2fc9e553dc2ab23a5b5b3735c40fba312f271715110d696bb5b367a984069","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:23:09.095123Z","signature_b64":"/vgxSEx5v+Me+l3WgZx8KNwZqOdQ/YtlpYiGAZycCwW+aQZrYGfSitPWR4j/mDXkttYTaD+zkJKQiCrPMbr/CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eaf2fc9e553dc2ab23a5b5b3735c40fba312f271715110d696bb5b367a984069","last_reissued_at":"2026-07-05T03:23:09.094668Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:23:09.094668Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2104.02145","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":"2026-07-05T03:23:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OUpOiJ0snumWr4jn8/2gQ2hxYXeWDNMAynsyW6fyf7F27ssvhv4pSN26iqcr6MCpFJJpbSlmv4Fz3iYToBU1BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-18T13:35:01.810190Z"},"content_sha256":"2ee16b596a7866a3f4f4039090262e157803a562b309645df34486d01f189a9b","schema_version":"1.0","event_id":"sha256:2ee16b596a7866a3f4f4039090262e157803a562b309645df34486d01f189a9b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:5LZPZHSVHXBKWI5FWWZXGXCA7O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"What Will it Take to Fix Benchmarking in Natural Language Understanding?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"George E. Dahl, Samuel R. Bowman","submitted_at":"2021-04-05T20:36:11Z","abstract_excerpt":"Evaluation for many natural language understanding (NLU) tasks is broken: Unreliable and biased systems score so highly on standard benchmarks that there is little room for researchers who develop better systems to demonstrate their improvements. The recent trend to abandon IID benchmarks in favor of adversarially-constructed, out-of-distribution test sets ensures that current models will perform poorly, but ultimately only obscures the abilities that we want our benchmarks to measure. In this position paper, we lay out four criteria that we argue NLU benchmarks should meet. We argue most curr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.02145","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/2104.02145/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":"2026-07-05T03:23:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jQnEfWWLK5t+pTidDDV3bXUGK9CeS7EGEHzVtFNUWNwdg7PbazyhiNhbYuRQaEAx1e08RduxdjhrN1cqydQtDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-18T13:35:01.810564Z"},"content_sha256":"3de670fb83d186a8e3bafe096f274edfe1faf28aab31e5eb20338848f65df05a","schema_version":"1.0","event_id":"sha256:3de670fb83d186a8e3bafe096f274edfe1faf28aab31e5eb20338848f65df05a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5LZPZHSVHXBKWI5FWWZXGXCA7O/bundle.json","state_url":"https://pith.science/pith/5LZPZHSVHXBKWI5FWWZXGXCA7O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5LZPZHSVHXBKWI5FWWZXGXCA7O/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":"2026-07-18T13:35:01Z","links":{"resolver":"https://pith.science/pith/5LZPZHSVHXBKWI5FWWZXGXCA7O","bundle":"https://pith.science/pith/5LZPZHSVHXBKWI5FWWZXGXCA7O/bundle.json","state":"https://pith.science/pith/5LZPZHSVHXBKWI5FWWZXGXCA7O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5LZPZHSVHXBKWI5FWWZXGXCA7O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:5LZPZHSVHXBKWI5FWWZXGXCA7O","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":"0f8f89fb2776d6a62926c97f953bf638cb6077eac93b4d70acf051c1c8c31444","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-05T20:36:11Z","title_canon_sha256":"26ef90efc21489307c3471cd7281ed3e8dd1c41dd229742e2a8ca03f6106a117"},"schema_version":"1.0","source":{"id":"2104.02145","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2104.02145","created_at":"2026-07-05T03:23:09Z"},{"alias_kind":"arxiv_version","alias_value":"2104.02145v3","created_at":"2026-07-05T03:23:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.02145","created_at":"2026-07-05T03:23:09Z"},{"alias_kind":"pith_short_12","alias_value":"5LZPZHSVHXBK","created_at":"2026-07-05T03:23:09Z"},{"alias_kind":"pith_short_16","alias_value":"5LZPZHSVHXBKWI5F","created_at":"2026-07-05T03:23:09Z"},{"alias_kind":"pith_short_8","alias_value":"5LZPZHSV","created_at":"2026-07-05T03:23:09Z"}],"graph_snapshots":[{"event_id":"sha256:3de670fb83d186a8e3bafe096f274edfe1faf28aab31e5eb20338848f65df05a","target":"graph","created_at":"2026-07-05T03:23:09Z","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/2104.02145/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluation for many natural language understanding (NLU) tasks is broken: Unreliable and biased systems score so highly on standard benchmarks that there is little room for researchers who develop better systems to demonstrate their improvements. The recent trend to abandon IID benchmarks in favor of adversarially-constructed, out-of-distribution test sets ensures that current models will perform poorly, but ultimately only obscures the abilities that we want our benchmarks to measure. In this position paper, we lay out four criteria that we argue NLU benchmarks should meet. We argue most curr","authors_text":"George E. Dahl, Samuel R. Bowman","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-05T20:36:11Z","title":"What Will it Take to Fix Benchmarking in Natural Language Understanding?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.02145","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":"sha256:2ee16b596a7866a3f4f4039090262e157803a562b309645df34486d01f189a9b","target":"record","created_at":"2026-07-05T03:23:09Z","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":"0f8f89fb2776d6a62926c97f953bf638cb6077eac93b4d70acf051c1c8c31444","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-04-05T20:36:11Z","title_canon_sha256":"26ef90efc21489307c3471cd7281ed3e8dd1c41dd229742e2a8ca03f6106a117"},"schema_version":"1.0","source":{"id":"2104.02145","kind":"arxiv","version":3}},"canonical_sha256":"eaf2fc9e553dc2ab23a5b5b3735c40fba312f271715110d696bb5b367a984069","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"eaf2fc9e553dc2ab23a5b5b3735c40fba312f271715110d696bb5b367a984069","first_computed_at":"2026-07-05T03:23:09.094668Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:23:09.094668Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/vgxSEx5v+Me+l3WgZx8KNwZqOdQ/YtlpYiGAZycCwW+aQZrYGfSitPWR4j/mDXkttYTaD+zkJKQiCrPMbr/CA==","signature_status":"signed_v1","signed_at":"2026-07-05T03:23:09.095123Z","signed_message":"canonical_sha256_bytes"},"source_id":"2104.02145","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2ee16b596a7866a3f4f4039090262e157803a562b309645df34486d01f189a9b","sha256:3de670fb83d186a8e3bafe096f274edfe1faf28aab31e5eb20338848f65df05a"],"state_sha256":"8f64ae991f32e655b34472274f30920955197d46420aef95528c5e4de0033d2b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Uu04wbu/vIuuaiUwUIWOZGQXqv9zsjUQcaZAzFN0wMKrRhjdAp4i03J95XAPvhNB9fZzawwS5on2YkV1eY3iCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-18T13:35:01.812661Z","bundle_sha256":"7a04ea7a92b64291612e57f92f0fb2753c6e86366e75329adf9e5f37b23077a8"}}