{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:5VIDZHBF5VD3C3H6WYXKBKTVY7","short_pith_number":"pith:5VIDZHBF","canonical_record":{"source":{"id":"2507.08342","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T06:44:52Z","cross_cats_sorted":[],"title_canon_sha256":"5937380b9eae998600a9fa34e7db29143c87001829a0021eb1c06640baf1cc40","abstract_canon_sha256":"869b457974e26b86adfaf3baf79b28c42c5161b1ac185d6834e6956e40b8f726"},"schema_version":"1.0"},"canonical_sha256":"ed503c9c25ed47b16cfeb62ea0aa75c7eaaf87efa5bb4a0728abb15c3f48974f","source":{"kind":"arxiv","id":"2507.08342","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.08342","created_at":"2026-07-05T11:35:21Z"},{"alias_kind":"arxiv_version","alias_value":"2507.08342v1","created_at":"2026-07-05T11:35:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.08342","created_at":"2026-07-05T11:35:21Z"},{"alias_kind":"pith_short_12","alias_value":"5VIDZHBF5VD3","created_at":"2026-07-05T11:35:21Z"},{"alias_kind":"pith_short_16","alias_value":"5VIDZHBF5VD3C3H6","created_at":"2026-07-05T11:35:21Z"},{"alias_kind":"pith_short_8","alias_value":"5VIDZHBF","created_at":"2026-07-05T11:35:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:5VIDZHBF5VD3C3H6WYXKBKTVY7","target":"record","payload":{"canonical_record":{"source":{"id":"2507.08342","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T06:44:52Z","cross_cats_sorted":[],"title_canon_sha256":"5937380b9eae998600a9fa34e7db29143c87001829a0021eb1c06640baf1cc40","abstract_canon_sha256":"869b457974e26b86adfaf3baf79b28c42c5161b1ac185d6834e6956e40b8f726"},"schema_version":"1.0"},"canonical_sha256":"ed503c9c25ed47b16cfeb62ea0aa75c7eaaf87efa5bb4a0728abb15c3f48974f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:35:21.707197Z","signature_b64":"l/4WBxqcqgGmfhoJ8eOOcOg1l2LgS6XgCCXDV6seqTzUptesn66gep48gBrJKEbGz1z+5qL6N6VpaqfNttGDBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed503c9c25ed47b16cfeb62ea0aa75c7eaaf87efa5bb4a0728abb15c3f48974f","last_reissued_at":"2026-07-05T11:35:21.706709Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:35:21.706709Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.08342","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":"2026-07-05T11:35:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KMe2lerujcneX5CUilDBCdT5r7W53pBwnWm7hlA5CSQXf4TRclZKsxRjzqafwR1sHz4DII4Qgw/YkgI2391+Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T04:14:55.660606Z"},"content_sha256":"3273233bb1146701cd2b914080b31f2b7cbf228bbb46112d07f7fd48dde7cc88","schema_version":"1.0","event_id":"sha256:3273233bb1146701cd2b914080b31f2b7cbf228bbb46112d07f7fd48dde7cc88"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:5VIDZHBF5VD3C3H6WYXKBKTVY7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Beyond N-Grams: Rethinking Evaluation Metrics and Strategies for Multilingual Abstractive Summarization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Itai Mondshine, Reut Tsarfaty, Tzuf Paz-Argaman","submitted_at":"2025-07-11T06:44:52Z","abstract_excerpt":"Automatic n-gram based metrics such as ROUGE are widely used for evaluating generative tasks such as summarization. While these metrics are considered indicative (even if imperfect) of human evaluation for English, their suitability for other languages remains unclear. To address this, we systematically assess evaluation metrics for generation both n-gram-based and neural based to evaluate their effectiveness across languages and tasks. Specifically, we design a large-scale evaluation suite across eight languages from four typological families: agglutinative, isolating, low-fusional, and high-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.08342","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/2507.08342/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":"2026-07-05T11:35:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TflGOxsAExe0sY2489lhzOCTNSw3ODkkhoX/UwGNIMn8UB7YVFgBJ6Od/vr5aBOEctcQxg0Eh7OrPBIi+AOQBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T04:14:55.661158Z"},"content_sha256":"3df802cc7f3c08ae36af4ee262f6e351cef33138f5d12d89cabc5dd205f74c87","schema_version":"1.0","event_id":"sha256:3df802cc7f3c08ae36af4ee262f6e351cef33138f5d12d89cabc5dd205f74c87"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5VIDZHBF5VD3C3H6WYXKBKTVY7/bundle.json","state_url":"https://pith.science/pith/5VIDZHBF5VD3C3H6WYXKBKTVY7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5VIDZHBF5VD3C3H6WYXKBKTVY7/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":"2026-08-09T04:14:55Z","links":{"resolver":"https://pith.science/pith/5VIDZHBF5VD3C3H6WYXKBKTVY7","bundle":"https://pith.science/pith/5VIDZHBF5VD3C3H6WYXKBKTVY7/bundle.json","state":"https://pith.science/pith/5VIDZHBF5VD3C3H6WYXKBKTVY7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5VIDZHBF5VD3C3H6WYXKBKTVY7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5VIDZHBF5VD3C3H6WYXKBKTVY7","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":"869b457974e26b86adfaf3baf79b28c42c5161b1ac185d6834e6956e40b8f726","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T06:44:52Z","title_canon_sha256":"5937380b9eae998600a9fa34e7db29143c87001829a0021eb1c06640baf1cc40"},"schema_version":"1.0","source":{"id":"2507.08342","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.08342","created_at":"2026-07-05T11:35:21Z"},{"alias_kind":"arxiv_version","alias_value":"2507.08342v1","created_at":"2026-07-05T11:35:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.08342","created_at":"2026-07-05T11:35:21Z"},{"alias_kind":"pith_short_12","alias_value":"5VIDZHBF5VD3","created_at":"2026-07-05T11:35:21Z"},{"alias_kind":"pith_short_16","alias_value":"5VIDZHBF5VD3C3H6","created_at":"2026-07-05T11:35:21Z"},{"alias_kind":"pith_short_8","alias_value":"5VIDZHBF","created_at":"2026-07-05T11:35:21Z"}],"graph_snapshots":[{"event_id":"sha256:3df802cc7f3c08ae36af4ee262f6e351cef33138f5d12d89cabc5dd205f74c87","target":"graph","created_at":"2026-07-05T11:35:21Z","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/2507.08342/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Automatic n-gram based metrics such as ROUGE are widely used for evaluating generative tasks such as summarization. While these metrics are considered indicative (even if imperfect) of human evaluation for English, their suitability for other languages remains unclear. To address this, we systematically assess evaluation metrics for generation both n-gram-based and neural based to evaluate their effectiveness across languages and tasks. Specifically, we design a large-scale evaluation suite across eight languages from four typological families: agglutinative, isolating, low-fusional, and high-","authors_text":"Itai Mondshine, Reut Tsarfaty, Tzuf Paz-Argaman","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T06:44:52Z","title":"Beyond N-Grams: Rethinking Evaluation Metrics and Strategies for Multilingual Abstractive Summarization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.08342","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":"sha256:3273233bb1146701cd2b914080b31f2b7cbf228bbb46112d07f7fd48dde7cc88","target":"record","created_at":"2026-07-05T11:35:21Z","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":"869b457974e26b86adfaf3baf79b28c42c5161b1ac185d6834e6956e40b8f726","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-11T06:44:52Z","title_canon_sha256":"5937380b9eae998600a9fa34e7db29143c87001829a0021eb1c06640baf1cc40"},"schema_version":"1.0","source":{"id":"2507.08342","kind":"arxiv","version":1}},"canonical_sha256":"ed503c9c25ed47b16cfeb62ea0aa75c7eaaf87efa5bb4a0728abb15c3f48974f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ed503c9c25ed47b16cfeb62ea0aa75c7eaaf87efa5bb4a0728abb15c3f48974f","first_computed_at":"2026-07-05T11:35:21.706709Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:35:21.706709Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"l/4WBxqcqgGmfhoJ8eOOcOg1l2LgS6XgCCXDV6seqTzUptesn66gep48gBrJKEbGz1z+5qL6N6VpaqfNttGDBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:35:21.707197Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.08342","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3273233bb1146701cd2b914080b31f2b7cbf228bbb46112d07f7fd48dde7cc88","sha256:3df802cc7f3c08ae36af4ee262f6e351cef33138f5d12d89cabc5dd205f74c87"],"state_sha256":"72c10827323605cd28bf831d818866039cefa66798a3bcb93d4887813e1a81e1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Tzwt7DJjsBMaJp7A6AI57zxQB584uc4yE3VRY3XyEWGLaLzCNu38HYhpZUnUcDG4EM0MCu5oC5av2BRwPzDuBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T04:14:55.667589Z","bundle_sha256":"f878ba4d1d8b1f2535eedb1e66dd4dd532d83c878445d1ecaeb9be3ef3244aab"}}