{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:7DABMYRPMAX2W6ZHIUTYXMT3GT","short_pith_number":"pith:7DABMYRP","canonical_record":{"source":{"id":"2403.14275","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-21T10:31:11Z","cross_cats_sorted":[],"title_canon_sha256":"0fc199ae9d12076ea42ccbb9035a086db9c070d26add0596e3d79b4fd97139c7","abstract_canon_sha256":"60bf281713a25bc2a78aec4b5a75b15e8b628b72f81f4f87a3e99003240adef6"},"schema_version":"1.0"},"canonical_sha256":"f8c016622f602fab7b2745278bb27b34f1dc31903eaf89d975876f7579f2133f","source":{"kind":"arxiv","id":"2403.14275","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.14275","created_at":"2026-07-05T07:59:06Z"},{"alias_kind":"arxiv_version","alias_value":"2403.14275v1","created_at":"2026-07-05T07:59:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.14275","created_at":"2026-07-05T07:59:06Z"},{"alias_kind":"pith_short_12","alias_value":"7DABMYRPMAX2","created_at":"2026-07-05T07:59:06Z"},{"alias_kind":"pith_short_16","alias_value":"7DABMYRPMAX2W6ZH","created_at":"2026-07-05T07:59:06Z"},{"alias_kind":"pith_short_8","alias_value":"7DABMYRP","created_at":"2026-07-05T07:59:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:7DABMYRPMAX2W6ZHIUTYXMT3GT","target":"record","payload":{"canonical_record":{"source":{"id":"2403.14275","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-21T10:31:11Z","cross_cats_sorted":[],"title_canon_sha256":"0fc199ae9d12076ea42ccbb9035a086db9c070d26add0596e3d79b4fd97139c7","abstract_canon_sha256":"60bf281713a25bc2a78aec4b5a75b15e8b628b72f81f4f87a3e99003240adef6"},"schema_version":"1.0"},"canonical_sha256":"f8c016622f602fab7b2745278bb27b34f1dc31903eaf89d975876f7579f2133f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:59:06.521154Z","signature_b64":"EWvQ1sWijcDp7wvmQVUv/8VRJOqcJAoCJgCBl7H+Kq0aaFKhpqRcY+zax13OguuwBERHZXeJZH+l9wms969ABQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f8c016622f602fab7b2745278bb27b34f1dc31903eaf89d975876f7579f2133f","last_reissued_at":"2026-07-05T07:59:06.520747Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:59:06.520747Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.14275","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":"2026-07-05T07:59:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XEyWBivw7PVyEIUlZhQ9FlGBM+5vj4bzXFYx9jxByAdEN+yEkL6gj+U+NtwoEy0tmWqRQEFuXZqN4B6g8wiDAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T15:43:36.316109Z"},"content_sha256":"9026296003dd6273aa3de51c3d1a60f52fc52a1ab1617379ddff2897e513362f","schema_version":"1.0","event_id":"sha256:9026296003dd6273aa3de51c3d1a60f52fc52a1ab1617379ddff2897e513362f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:7DABMYRPMAX2W6ZHIUTYXMT3GT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Is Reference Necessary in the Evaluation of NLG Systems? When and Where?","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenghu Zhou, Jiaxin Ding, Lei Zhou, Luoyi Fu, Shuqian Sheng, Xinbing Wang, Yi Xu","submitted_at":"2024-03-21T10:31:11Z","abstract_excerpt":"The majority of automatic metrics for evaluating NLG systems are reference-based. However, the challenge of collecting human annotation results in a lack of reliable references in numerous application scenarios. Despite recent advancements in reference-free metrics, it has not been well understood when and where they can be used as an alternative to reference-based metrics. In this study, by employing diverse analytical approaches, we comprehensively assess the performance of both metrics across a wide range of NLG tasks, encompassing eight datasets and eight evaluation models. Based on solid "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.14275","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/2403.14275/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":"2026-07-05T07:59:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+F/ae8j6Jg3kg146vMyq/az3k7h3WaFLt8iyEyaX7RPGXW9PDAg0Dqs6CVbe4n3UE0+iL+S9Oi4YAXtUif1uAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T15:43:36.319645Z"},"content_sha256":"091ae8eda2eea788e59609364f31db6c4819e5e8ca3977effe3bf9b734a29db4","schema_version":"1.0","event_id":"sha256:091ae8eda2eea788e59609364f31db6c4819e5e8ca3977effe3bf9b734a29db4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7DABMYRPMAX2W6ZHIUTYXMT3GT/bundle.json","state_url":"https://pith.science/pith/7DABMYRPMAX2W6ZHIUTYXMT3GT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7DABMYRPMAX2W6ZHIUTYXMT3GT/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":"2026-08-20T15:43:36Z","links":{"resolver":"https://pith.science/pith/7DABMYRPMAX2W6ZHIUTYXMT3GT","bundle":"https://pith.science/pith/7DABMYRPMAX2W6ZHIUTYXMT3GT/bundle.json","state":"https://pith.science/pith/7DABMYRPMAX2W6ZHIUTYXMT3GT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7DABMYRPMAX2W6ZHIUTYXMT3GT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:7DABMYRPMAX2W6ZHIUTYXMT3GT","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":"60bf281713a25bc2a78aec4b5a75b15e8b628b72f81f4f87a3e99003240adef6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-21T10:31:11Z","title_canon_sha256":"0fc199ae9d12076ea42ccbb9035a086db9c070d26add0596e3d79b4fd97139c7"},"schema_version":"1.0","source":{"id":"2403.14275","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.14275","created_at":"2026-07-05T07:59:06Z"},{"alias_kind":"arxiv_version","alias_value":"2403.14275v1","created_at":"2026-07-05T07:59:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.14275","created_at":"2026-07-05T07:59:06Z"},{"alias_kind":"pith_short_12","alias_value":"7DABMYRPMAX2","created_at":"2026-07-05T07:59:06Z"},{"alias_kind":"pith_short_16","alias_value":"7DABMYRPMAX2W6ZH","created_at":"2026-07-05T07:59:06Z"},{"alias_kind":"pith_short_8","alias_value":"7DABMYRP","created_at":"2026-07-05T07:59:06Z"}],"graph_snapshots":[{"event_id":"sha256:091ae8eda2eea788e59609364f31db6c4819e5e8ca3977effe3bf9b734a29db4","target":"graph","created_at":"2026-07-05T07:59:06Z","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/2403.14275/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The majority of automatic metrics for evaluating NLG systems are reference-based. However, the challenge of collecting human annotation results in a lack of reliable references in numerous application scenarios. Despite recent advancements in reference-free metrics, it has not been well understood when and where they can be used as an alternative to reference-based metrics. In this study, by employing diverse analytical approaches, we comprehensively assess the performance of both metrics across a wide range of NLG tasks, encompassing eight datasets and eight evaluation models. Based on solid ","authors_text":"Chenghu Zhou, Jiaxin Ding, Lei Zhou, Luoyi Fu, Shuqian Sheng, Xinbing Wang, Yi Xu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-21T10:31:11Z","title":"Is Reference Necessary in the Evaluation of NLG Systems? When and Where?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.14275","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":"sha256:9026296003dd6273aa3de51c3d1a60f52fc52a1ab1617379ddff2897e513362f","target":"record","created_at":"2026-07-05T07:59:06Z","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":"60bf281713a25bc2a78aec4b5a75b15e8b628b72f81f4f87a3e99003240adef6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-21T10:31:11Z","title_canon_sha256":"0fc199ae9d12076ea42ccbb9035a086db9c070d26add0596e3d79b4fd97139c7"},"schema_version":"1.0","source":{"id":"2403.14275","kind":"arxiv","version":1}},"canonical_sha256":"f8c016622f602fab7b2745278bb27b34f1dc31903eaf89d975876f7579f2133f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f8c016622f602fab7b2745278bb27b34f1dc31903eaf89d975876f7579f2133f","first_computed_at":"2026-07-05T07:59:06.520747Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:59:06.520747Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"EWvQ1sWijcDp7wvmQVUv/8VRJOqcJAoCJgCBl7H+Kq0aaFKhpqRcY+zax13OguuwBERHZXeJZH+l9wms969ABQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:59:06.521154Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.14275","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9026296003dd6273aa3de51c3d1a60f52fc52a1ab1617379ddff2897e513362f","sha256:091ae8eda2eea788e59609364f31db6c4819e5e8ca3977effe3bf9b734a29db4"],"state_sha256":"ed228c4e2dcb7fff53c28c2e8173b1cf300c4c3e1604277ae246da389bf5a18d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Dhp1XJZ32za6+Vc8S5W0yTKkLXGgEJ3CIwrg28VsK4ZkXK+utmLkoJnozP4rpiEwLnh9AOUv1kvrY+M7oqW7CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T15:43:36.341544Z","bundle_sha256":"9ace5334538343bf12e74b703d8adff64dc27a11b27a4b9b8f2d843ee4b6362c"}}