{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:EOXR7XK3COD45OQ5OHYZODZGXC","short_pith_number":"pith:EOXR7XK3","canonical_record":{"source":{"id":"2109.05771","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T08:26:26Z","cross_cats_sorted":[],"title_canon_sha256":"40e05b2111a3960da64311d587af2ee3effaf1d3990a92918c9f8cb80c3d350d","abstract_canon_sha256":"cff335779f1147b8cf56b772c46ca56bfa48d8591b5d8f8a56393d25d7cac165"},"schema_version":"1.0"},"canonical_sha256":"23af1fdd5b1387ceba1d71f1970f26b898895dd6ab5a0276123fdc643476ee57","source":{"kind":"arxiv","id":"2109.05771","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.05771","created_at":"2026-07-05T03:13:42Z"},{"alias_kind":"arxiv_version","alias_value":"2109.05771v1","created_at":"2026-07-05T03:13:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.05771","created_at":"2026-07-05T03:13:42Z"},{"alias_kind":"pith_short_12","alias_value":"EOXR7XK3COD4","created_at":"2026-07-05T03:13:42Z"},{"alias_kind":"pith_short_16","alias_value":"EOXR7XK3COD45OQ5","created_at":"2026-07-05T03:13:42Z"},{"alias_kind":"pith_short_8","alias_value":"EOXR7XK3","created_at":"2026-07-05T03:13:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:EOXR7XK3COD45OQ5OHYZODZGXC","target":"record","payload":{"canonical_record":{"source":{"id":"2109.05771","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T08:26:26Z","cross_cats_sorted":[],"title_canon_sha256":"40e05b2111a3960da64311d587af2ee3effaf1d3990a92918c9f8cb80c3d350d","abstract_canon_sha256":"cff335779f1147b8cf56b772c46ca56bfa48d8591b5d8f8a56393d25d7cac165"},"schema_version":"1.0"},"canonical_sha256":"23af1fdd5b1387ceba1d71f1970f26b898895dd6ab5a0276123fdc643476ee57","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:13:42.257880Z","signature_b64":"TW0/rD+ZZPRP5kweflFxf560XRR1aeR46eS65PRzJJiLexdwphybk9FRXeLAsp4HfjwMQa+EeZt0cfhUeqjgCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"23af1fdd5b1387ceba1d71f1970f26b898895dd6ab5a0276123fdc643476ee57","last_reissued_at":"2026-07-05T03:13:42.257407Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:13:42.257407Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2109.05771","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":"2026-07-05T03:13:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6nc7qkhaSasz8Ygl/OCqBnbbcCpXCN9wv+w1RmnswMQA0NpnJmccjMwGiK7g6wp62dtmsSahTdICM3CwkP31Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T15:47:28.133498Z"},"content_sha256":"d3f217a6826113dabc4513b44caf63905cd71d377e4bef29d3125cd79187a42e","schema_version":"1.0","event_id":"sha256:d3f217a6826113dabc4513b44caf63905cd71d377e4bef29d3125cd79187a42e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:EOXR7XK3COD45OQ5OHYZODZGXC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Perturbation CheckLists for Evaluating NLG Evaluation Metrics","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ananya B. Sai, Dev Yashpal Sheth, Mitesh M. Khapra, Sreyas Mohan, Tanay Dixit","submitted_at":"2021-09-13T08:26:26Z","abstract_excerpt":"Natural Language Generation (NLG) evaluation is a multifaceted task requiring assessment of multiple desirable criteria, e.g., fluency, coherency, coverage, relevance, adequacy, overall quality, etc. Across existing datasets for 6 NLG tasks, we observe that the human evaluation scores on these multiple criteria are often not correlated. For example, there is a very low correlation between human scores on fluency and data coverage for the task of structured data to text generation. This suggests that the current recipe of proposing new automatic evaluation metrics for NLG by showing that they c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.05771","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/2109.05771/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":"2026-07-05T03:13:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ro/BSO10irecoNw+YMCl2VGnaboarYHymk0v9nD3zO0iVpt8qxCdcNVJt2Klez7VEFWBsABuCMPV0eUbzmwxDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T15:47:28.134158Z"},"content_sha256":"a7680a026064cc5834920cbb462a7ff3b82066211cbf9c2729fe8eb10191c81d","schema_version":"1.0","event_id":"sha256:a7680a026064cc5834920cbb462a7ff3b82066211cbf9c2729fe8eb10191c81d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EOXR7XK3COD45OQ5OHYZODZGXC/bundle.json","state_url":"https://pith.science/pith/EOXR7XK3COD45OQ5OHYZODZGXC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EOXR7XK3COD45OQ5OHYZODZGXC/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":"2026-08-20T15:47:28Z","links":{"resolver":"https://pith.science/pith/EOXR7XK3COD45OQ5OHYZODZGXC","bundle":"https://pith.science/pith/EOXR7XK3COD45OQ5OHYZODZGXC/bundle.json","state":"https://pith.science/pith/EOXR7XK3COD45OQ5OHYZODZGXC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EOXR7XK3COD45OQ5OHYZODZGXC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:EOXR7XK3COD45OQ5OHYZODZGXC","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":"cff335779f1147b8cf56b772c46ca56bfa48d8591b5d8f8a56393d25d7cac165","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T08:26:26Z","title_canon_sha256":"40e05b2111a3960da64311d587af2ee3effaf1d3990a92918c9f8cb80c3d350d"},"schema_version":"1.0","source":{"id":"2109.05771","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.05771","created_at":"2026-07-05T03:13:42Z"},{"alias_kind":"arxiv_version","alias_value":"2109.05771v1","created_at":"2026-07-05T03:13:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.05771","created_at":"2026-07-05T03:13:42Z"},{"alias_kind":"pith_short_12","alias_value":"EOXR7XK3COD4","created_at":"2026-07-05T03:13:42Z"},{"alias_kind":"pith_short_16","alias_value":"EOXR7XK3COD45OQ5","created_at":"2026-07-05T03:13:42Z"},{"alias_kind":"pith_short_8","alias_value":"EOXR7XK3","created_at":"2026-07-05T03:13:42Z"}],"graph_snapshots":[{"event_id":"sha256:a7680a026064cc5834920cbb462a7ff3b82066211cbf9c2729fe8eb10191c81d","target":"graph","created_at":"2026-07-05T03:13:42Z","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/2109.05771/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Natural Language Generation (NLG) evaluation is a multifaceted task requiring assessment of multiple desirable criteria, e.g., fluency, coherency, coverage, relevance, adequacy, overall quality, etc. Across existing datasets for 6 NLG tasks, we observe that the human evaluation scores on these multiple criteria are often not correlated. For example, there is a very low correlation between human scores on fluency and data coverage for the task of structured data to text generation. This suggests that the current recipe of proposing new automatic evaluation metrics for NLG by showing that they c","authors_text":"Ananya B. Sai, Dev Yashpal Sheth, Mitesh M. Khapra, Sreyas Mohan, Tanay Dixit","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T08:26:26Z","title":"Perturbation CheckLists for Evaluating NLG Evaluation Metrics"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.05771","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:d3f217a6826113dabc4513b44caf63905cd71d377e4bef29d3125cd79187a42e","target":"record","created_at":"2026-07-05T03:13:42Z","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":"cff335779f1147b8cf56b772c46ca56bfa48d8591b5d8f8a56393d25d7cac165","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-13T08:26:26Z","title_canon_sha256":"40e05b2111a3960da64311d587af2ee3effaf1d3990a92918c9f8cb80c3d350d"},"schema_version":"1.0","source":{"id":"2109.05771","kind":"arxiv","version":1}},"canonical_sha256":"23af1fdd5b1387ceba1d71f1970f26b898895dd6ab5a0276123fdc643476ee57","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"23af1fdd5b1387ceba1d71f1970f26b898895dd6ab5a0276123fdc643476ee57","first_computed_at":"2026-07-05T03:13:42.257407Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:13:42.257407Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TW0/rD+ZZPRP5kweflFxf560XRR1aeR46eS65PRzJJiLexdwphybk9FRXeLAsp4HfjwMQa+EeZt0cfhUeqjgCA==","signature_status":"signed_v1","signed_at":"2026-07-05T03:13:42.257880Z","signed_message":"canonical_sha256_bytes"},"source_id":"2109.05771","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d3f217a6826113dabc4513b44caf63905cd71d377e4bef29d3125cd79187a42e","sha256:a7680a026064cc5834920cbb462a7ff3b82066211cbf9c2729fe8eb10191c81d"],"state_sha256":"065ac2a8bf3e659355707bbd9131984d54f4b37aa41f536011b7532ad6026766"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kgyz4shPGzKcY4ox0zzGRq2YNy0NOqeeD4HuGyKOfbPwe19IHFCgILY2imJBUVwlwPuwLTcZihKLdjjB5islCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T15:47:28.140411Z","bundle_sha256":"c198044d492631f76efa79ae5977dfc6dbe842c38c16aa5596088c89bebeccf7"}}