{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:NA7BA4NSPEFQBEGJSHOSOUT73C","short_pith_number":"pith:NA7BA4NS","canonical_record":{"source":{"id":"2501.12011","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T10:05:48Z","cross_cats_sorted":[],"title_canon_sha256":"24d4e3901402c89b26d7b1f154f2894100d645040e002a47cf06d23b01485b78","abstract_canon_sha256":"cf8d6719a373a87e3e5bb4aecbd940bf84c31c69b4772f0d0f0371865a7deaab"},"schema_version":"1.0"},"canonical_sha256":"683e1071b2790b0090c991dd27527fd897ec97b9cea06bd9ed6b70d8d3dd333b","source":{"kind":"arxiv","id":"2501.12011","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.12011","created_at":"2026-07-05T10:03:30Z"},{"alias_kind":"arxiv_version","alias_value":"2501.12011v1","created_at":"2026-07-05T10:03:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.12011","created_at":"2026-07-05T10:03:30Z"},{"alias_kind":"pith_short_12","alias_value":"NA7BA4NSPEFQ","created_at":"2026-07-05T10:03:30Z"},{"alias_kind":"pith_short_16","alias_value":"NA7BA4NSPEFQBEGJ","created_at":"2026-07-05T10:03:30Z"},{"alias_kind":"pith_short_8","alias_value":"NA7BA4NS","created_at":"2026-07-05T10:03:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:NA7BA4NSPEFQBEGJSHOSOUT73C","target":"record","payload":{"canonical_record":{"source":{"id":"2501.12011","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T10:05:48Z","cross_cats_sorted":[],"title_canon_sha256":"24d4e3901402c89b26d7b1f154f2894100d645040e002a47cf06d23b01485b78","abstract_canon_sha256":"cf8d6719a373a87e3e5bb4aecbd940bf84c31c69b4772f0d0f0371865a7deaab"},"schema_version":"1.0"},"canonical_sha256":"683e1071b2790b0090c991dd27527fd897ec97b9cea06bd9ed6b70d8d3dd333b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:03:30.440597Z","signature_b64":"p93JDxPSDvDsNb7mLNLffqy4nrEJFrYpdSxr5Q/eovPAzsxRUiff8YW0cxho6u7wRtjOm1lQ4aynk84UlIGkBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"683e1071b2790b0090c991dd27527fd897ec97b9cea06bd9ed6b70d8d3dd333b","last_reissued_at":"2026-07-05T10:03:30.440044Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:03:30.440044Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.12011","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":"2026-07-05T10:03:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rY2eNNqHUVqfxqgY4vpvtBHGIoTEaQJd30++O651kQcyhuMAkXtbUoMJICWr9PHiUUaGMB7Fmy3m1J4VCkoxAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T15:54:09.525079Z"},"content_sha256":"a1599d3afc5b29893b83b86cf44734967992593d4cda01fb82f8965e0e7be010","schema_version":"1.0","event_id":"sha256:a1599d3afc5b29893b83b86cf44734967992593d4cda01fb82f8965e0e7be010"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:NA7BA4NSPEFQBEGJSHOSOUT73C","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reference-free Evaluation Metrics for Text Generation: A Survey","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jun Suzuki, Kees van Deemter, Takumi Ito","submitted_at":"2025-01-21T10:05:48Z","abstract_excerpt":"A number of automatic evaluation metrics have been proposed for natural language generation systems. The most common approach to automatic evaluation is the use of a reference-based metric that compares the model's output with gold-standard references written by humans. However, it is expensive to create such references, and for some tasks, such as response generation in dialogue, creating references is not a simple matter. Therefore, various reference-free metrics have been developed in recent years. In this survey, which intends to cover the full breadth of all NLG tasks, we investigate the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.12011","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/2501.12011/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":"2026-07-05T10:03:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6sEAZeGXqdicnJ+EJEoAWJgFSn0zIzNh7drad8kB5T9cnxEk+j+HYCepyWzkNsOrOiMFLxIcSCemmSGU6xEDAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T15:54:09.525600Z"},"content_sha256":"8f23c40d2377186428e6872cc27d7c2c4a56184408341860797fcd0d56787501","schema_version":"1.0","event_id":"sha256:8f23c40d2377186428e6872cc27d7c2c4a56184408341860797fcd0d56787501"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NA7BA4NSPEFQBEGJSHOSOUT73C/bundle.json","state_url":"https://pith.science/pith/NA7BA4NSPEFQBEGJSHOSOUT73C/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NA7BA4NSPEFQBEGJSHOSOUT73C/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":"2026-08-04T15:54:09Z","links":{"resolver":"https://pith.science/pith/NA7BA4NSPEFQBEGJSHOSOUT73C","bundle":"https://pith.science/pith/NA7BA4NSPEFQBEGJSHOSOUT73C/bundle.json","state":"https://pith.science/pith/NA7BA4NSPEFQBEGJSHOSOUT73C/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NA7BA4NSPEFQBEGJSHOSOUT73C/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:NA7BA4NSPEFQBEGJSHOSOUT73C","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":"cf8d6719a373a87e3e5bb4aecbd940bf84c31c69b4772f0d0f0371865a7deaab","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T10:05:48Z","title_canon_sha256":"24d4e3901402c89b26d7b1f154f2894100d645040e002a47cf06d23b01485b78"},"schema_version":"1.0","source":{"id":"2501.12011","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.12011","created_at":"2026-07-05T10:03:30Z"},{"alias_kind":"arxiv_version","alias_value":"2501.12011v1","created_at":"2026-07-05T10:03:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.12011","created_at":"2026-07-05T10:03:30Z"},{"alias_kind":"pith_short_12","alias_value":"NA7BA4NSPEFQ","created_at":"2026-07-05T10:03:30Z"},{"alias_kind":"pith_short_16","alias_value":"NA7BA4NSPEFQBEGJ","created_at":"2026-07-05T10:03:30Z"},{"alias_kind":"pith_short_8","alias_value":"NA7BA4NS","created_at":"2026-07-05T10:03:30Z"}],"graph_snapshots":[{"event_id":"sha256:8f23c40d2377186428e6872cc27d7c2c4a56184408341860797fcd0d56787501","target":"graph","created_at":"2026-07-05T10:03:30Z","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/2501.12011/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"A number of automatic evaluation metrics have been proposed for natural language generation systems. The most common approach to automatic evaluation is the use of a reference-based metric that compares the model's output with gold-standard references written by humans. However, it is expensive to create such references, and for some tasks, such as response generation in dialogue, creating references is not a simple matter. Therefore, various reference-free metrics have been developed in recent years. In this survey, which intends to cover the full breadth of all NLG tasks, we investigate the ","authors_text":"Jun Suzuki, Kees van Deemter, Takumi Ito","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T10:05:48Z","title":"Reference-free Evaluation Metrics for Text Generation: A Survey"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.12011","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:a1599d3afc5b29893b83b86cf44734967992593d4cda01fb82f8965e0e7be010","target":"record","created_at":"2026-07-05T10:03:30Z","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":"cf8d6719a373a87e3e5bb4aecbd940bf84c31c69b4772f0d0f0371865a7deaab","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T10:05:48Z","title_canon_sha256":"24d4e3901402c89b26d7b1f154f2894100d645040e002a47cf06d23b01485b78"},"schema_version":"1.0","source":{"id":"2501.12011","kind":"arxiv","version":1}},"canonical_sha256":"683e1071b2790b0090c991dd27527fd897ec97b9cea06bd9ed6b70d8d3dd333b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"683e1071b2790b0090c991dd27527fd897ec97b9cea06bd9ed6b70d8d3dd333b","first_computed_at":"2026-07-05T10:03:30.440044Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:03:30.440044Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"p93JDxPSDvDsNb7mLNLffqy4nrEJFrYpdSxr5Q/eovPAzsxRUiff8YW0cxho6u7wRtjOm1lQ4aynk84UlIGkBw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:03:30.440597Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.12011","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a1599d3afc5b29893b83b86cf44734967992593d4cda01fb82f8965e0e7be010","sha256:8f23c40d2377186428e6872cc27d7c2c4a56184408341860797fcd0d56787501"],"state_sha256":"4d2877535cf2019e03efb38a12e486dea1c01afba6c7cd5d780a3118eac42bfe"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5iyX2Wgqr3MdcH/SfYQfwKds5NV0/es1t1fBO/MKPbwtbMiILKyAKUIx1oowkYuQ8mkHEJHS6K/rY6AnsdpNAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T15:54:09.530541Z","bundle_sha256":"a69de3ec1bc563dd829a9de14a5a20f5f290beabe3576ad7607a45b9ea978dc7"}}