{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:C2OK3ZCWAJ77WDXRRWKFVRTE5A","short_pith_number":"pith:C2OK3ZCW","canonical_record":{"source":{"id":"2104.09712","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-20T01:35:16Z","cross_cats_sorted":[],"title_canon_sha256":"2f222b51fa613c1642c71fcc0d587d312b3cac42bdbfcac598e9b397b9fddd77","abstract_canon_sha256":"e21ce3505e9e71474f5da0e0d4b6d29e6e7afa60d9c37ffc77282644a12c6e7a"},"schema_version":"1.0"},"canonical_sha256":"169cade456027ffb0ef18d945ac664e81829d64d907993db6a198ab4a0244112","source":{"kind":"arxiv","id":"2104.09712","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2104.09712","created_at":"2026-07-05T02:33:36Z"},{"alias_kind":"arxiv_version","alias_value":"2104.09712v1","created_at":"2026-07-05T02:33:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.09712","created_at":"2026-07-05T02:33:36Z"},{"alias_kind":"pith_short_12","alias_value":"C2OK3ZCWAJ77","created_at":"2026-07-05T02:33:36Z"},{"alias_kind":"pith_short_16","alias_value":"C2OK3ZCWAJ77WDXR","created_at":"2026-07-05T02:33:36Z"},{"alias_kind":"pith_short_8","alias_value":"C2OK3ZCW","created_at":"2026-07-05T02:33:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:C2OK3ZCWAJ77WDXRRWKFVRTE5A","target":"record","payload":{"canonical_record":{"source":{"id":"2104.09712","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-20T01:35:16Z","cross_cats_sorted":[],"title_canon_sha256":"2f222b51fa613c1642c71fcc0d587d312b3cac42bdbfcac598e9b397b9fddd77","abstract_canon_sha256":"e21ce3505e9e71474f5da0e0d4b6d29e6e7afa60d9c37ffc77282644a12c6e7a"},"schema_version":"1.0"},"canonical_sha256":"169cade456027ffb0ef18d945ac664e81829d64d907993db6a198ab4a0244112","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:33:36.300478Z","signature_b64":"O3SzVigxpOz8seAiWhx6P6sjOCuRNScXamOssRwid53+riFLe/PwuhmNtgpZ+LvCPT1g8UEDdJm5zOjptmM/Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"169cade456027ffb0ef18d945ac664e81829d64d907993db6a198ab4a0244112","last_reissued_at":"2026-07-05T02:33:36.300053Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:33:36.300053Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2104.09712","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:33:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kUHv44xcYLYK+ll7K9KorF11vq5Lrv4fI7h+fvsEsV2Fq3TQ/D14RV2eEkxjGJEA/paDO83jp8J2EzHPxf9xBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T13:04:37.894746Z"},"content_sha256":"d81b0f8561ba3c274e899b04ddc23dbcde186fbefa965b50c4ebb731767c99ee","schema_version":"1.0","event_id":"sha256:d81b0f8561ba3c274e899b04ddc23dbcde186fbefa965b50c4ebb731767c99ee"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:C2OK3ZCWAJ77WDXRRWKFVRTE5A","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Problems and Countermeasures in Natural Language Processing Evaluation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Baobao Chang, Qingxiu Dong, Weidong Zhan, Zhifang Sui","submitted_at":"2021-04-20T01:35:16Z","abstract_excerpt":"Evaluation in natural language processing guides and promotes research on models and methods. In recent years, new evalua-tion data sets and evaluation tasks have been continuously proposed. At the same time, a series of problems exposed by ex-isting evaluation have also restricted the progress of natural language processing technology. Starting from the concept, com-position, development and meaning of natural language evaluation, this article classifies and summarizes the tasks and char-acteristics of mainstream natural language evaluation, and then summarizes the problems and causes of natu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.09712","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/2104.09712/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":"2026-07-05T02:33:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sio0dlZV+6UrAR/pVgiXGZHTG4zSHuNoJFI+S/SLObfUSDAk+8fNBduU1nKUosvyutcfUDUOfsb66YVgXgmoBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T13:04:37.895768Z"},"content_sha256":"57c0c7057fc4441c378c7851cc1b12e037f4530ed985dd45ae538ab766978592","schema_version":"1.0","event_id":"sha256:57c0c7057fc4441c378c7851cc1b12e037f4530ed985dd45ae538ab766978592"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/C2OK3ZCWAJ77WDXRRWKFVRTE5A/bundle.json","state_url":"https://pith.science/pith/C2OK3ZCWAJ77WDXRRWKFVRTE5A/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/C2OK3ZCWAJ77WDXRRWKFVRTE5A/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":"2026-08-11T13:04:37Z","links":{"resolver":"https://pith.science/pith/C2OK3ZCWAJ77WDXRRWKFVRTE5A","bundle":"https://pith.science/pith/C2OK3ZCWAJ77WDXRRWKFVRTE5A/bundle.json","state":"https://pith.science/pith/C2OK3ZCWAJ77WDXRRWKFVRTE5A/state.json","well_known_bundle":"https://pith.science/.well-known/pith/C2OK3ZCWAJ77WDXRRWKFVRTE5A/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:C2OK3ZCWAJ77WDXRRWKFVRTE5A","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":"e21ce3505e9e71474f5da0e0d4b6d29e6e7afa60d9c37ffc77282644a12c6e7a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-20T01:35:16Z","title_canon_sha256":"2f222b51fa613c1642c71fcc0d587d312b3cac42bdbfcac598e9b397b9fddd77"},"schema_version":"1.0","source":{"id":"2104.09712","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2104.09712","created_at":"2026-07-05T02:33:36Z"},{"alias_kind":"arxiv_version","alias_value":"2104.09712v1","created_at":"2026-07-05T02:33:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.09712","created_at":"2026-07-05T02:33:36Z"},{"alias_kind":"pith_short_12","alias_value":"C2OK3ZCWAJ77","created_at":"2026-07-05T02:33:36Z"},{"alias_kind":"pith_short_16","alias_value":"C2OK3ZCWAJ77WDXR","created_at":"2026-07-05T02:33:36Z"},{"alias_kind":"pith_short_8","alias_value":"C2OK3ZCW","created_at":"2026-07-05T02:33:36Z"}],"graph_snapshots":[{"event_id":"sha256:57c0c7057fc4441c378c7851cc1b12e037f4530ed985dd45ae538ab766978592","target":"graph","created_at":"2026-07-05T02:33:36Z","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/2104.09712/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluation in natural language processing guides and promotes research on models and methods. In recent years, new evalua-tion data sets and evaluation tasks have been continuously proposed. At the same time, a series of problems exposed by ex-isting evaluation have also restricted the progress of natural language processing technology. Starting from the concept, com-position, development and meaning of natural language evaluation, this article classifies and summarizes the tasks and char-acteristics of mainstream natural language evaluation, and then summarizes the problems and causes of natu","authors_text":"Baobao Chang, Qingxiu Dong, Weidong Zhan, Zhifang Sui","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-20T01:35:16Z","title":"Problems and Countermeasures in Natural Language Processing Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.09712","kind":"arxiv","version":1},"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:d81b0f8561ba3c274e899b04ddc23dbcde186fbefa965b50c4ebb731767c99ee","target":"record","created_at":"2026-07-05T02:33:36Z","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":"e21ce3505e9e71474f5da0e0d4b6d29e6e7afa60d9c37ffc77282644a12c6e7a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-20T01:35:16Z","title_canon_sha256":"2f222b51fa613c1642c71fcc0d587d312b3cac42bdbfcac598e9b397b9fddd77"},"schema_version":"1.0","source":{"id":"2104.09712","kind":"arxiv","version":1}},"canonical_sha256":"169cade456027ffb0ef18d945ac664e81829d64d907993db6a198ab4a0244112","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"169cade456027ffb0ef18d945ac664e81829d64d907993db6a198ab4a0244112","first_computed_at":"2026-07-05T02:33:36.300053Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:33:36.300053Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"O3SzVigxpOz8seAiWhx6P6sjOCuRNScXamOssRwid53+riFLe/PwuhmNtgpZ+LvCPT1g8UEDdJm5zOjptmM/Bw==","signature_status":"signed_v1","signed_at":"2026-07-05T02:33:36.300478Z","signed_message":"canonical_sha256_bytes"},"source_id":"2104.09712","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d81b0f8561ba3c274e899b04ddc23dbcde186fbefa965b50c4ebb731767c99ee","sha256:57c0c7057fc4441c378c7851cc1b12e037f4530ed985dd45ae538ab766978592"],"state_sha256":"ecc2d5fb3a32bd067fe569e027e95b9d4b0c612d7ddf5a948cd88e1a9fa35c23"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AYOH4GzBJhxuVKlGf9g/g9dJUMwFBvpS1z3QhZEVyXwSuMcFJvTtXGsgc7iDO6l1XvPplFkM/gvtaZWKMhhKDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T13:04:37.902287Z","bundle_sha256":"1f38e8b4c085d7814cf2bbb4a32af86c190887889d83497db954713164604caa"}}