{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:GK6ISBHT3K4GPZVRU7LRJHSYO4","short_pith_number":"pith:GK6ISBHT","canonical_record":{"source":{"id":"2308.16797","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-08-31T15:19:28Z","cross_cats_sorted":[],"title_canon_sha256":"abe2f57c3f367364a6aceeb82d130603d7690dc2c1d844db5f9e21ed73aab769","abstract_canon_sha256":"77d22fd3a43cfdbbb2ce95a93d686f4de070ae0bf1826128679497cfb222b125"},"schema_version":"1.0"},"canonical_sha256":"32bc8904f3dab867e6b1a7d7149e587733965f30c94bcc05ed920438907951fa","source":{"kind":"arxiv","id":"2308.16797","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.16797","created_at":"2026-07-05T06:48:51Z"},{"alias_kind":"arxiv_version","alias_value":"2308.16797v2","created_at":"2026-07-05T06:48:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.16797","created_at":"2026-07-05T06:48:51Z"},{"alias_kind":"pith_short_12","alias_value":"GK6ISBHT3K4G","created_at":"2026-07-05T06:48:51Z"},{"alias_kind":"pith_short_16","alias_value":"GK6ISBHT3K4GPZVR","created_at":"2026-07-05T06:48:51Z"},{"alias_kind":"pith_short_8","alias_value":"GK6ISBHT","created_at":"2026-07-05T06:48:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:GK6ISBHT3K4GPZVRU7LRJHSYO4","target":"record","payload":{"canonical_record":{"source":{"id":"2308.16797","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-08-31T15:19:28Z","cross_cats_sorted":[],"title_canon_sha256":"abe2f57c3f367364a6aceeb82d130603d7690dc2c1d844db5f9e21ed73aab769","abstract_canon_sha256":"77d22fd3a43cfdbbb2ce95a93d686f4de070ae0bf1826128679497cfb222b125"},"schema_version":"1.0"},"canonical_sha256":"32bc8904f3dab867e6b1a7d7149e587733965f30c94bcc05ed920438907951fa","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:48:51.621443Z","signature_b64":"SBB8NEJI4DqYM0IJfbhKlRh5I3BM2qT69L8jCrXhcTNES5x+itkqZhyjIHQ25qOUxHvuWLUvknpeqUpfn2ZPAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"32bc8904f3dab867e6b1a7d7149e587733965f30c94bcc05ed920438907951fa","last_reissued_at":"2026-07-05T06:48:51.620976Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:48:51.620976Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2308.16797","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:48:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TtOmb3t8YZk1Jvh7hAxRg7IKT5pCmAHz0Z7/Aslx7Vg0L/pROhkMaMhhK/ULJNpwy96ndx9usdgGvfBLWBA4Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T19:23:15.810773Z"},"content_sha256":"13725b1b6975c3288391ef340c8e6a5339a1eeab46f73e974d33fe927f65da76","schema_version":"1.0","event_id":"sha256:13725b1b6975c3288391ef340c8e6a5339a1eeab46f73e974d33fe927f65da76"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:GK6ISBHT3K4GPZVRU7LRJHSYO4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Simple LLM Prompting is State-of-the-Art for Robust and Multilingual Dialogue Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alon Lavie, Helena Moniz, Isabel Trancoso, Jo\\~ao Paulo Carvalho, John Mendon\\c{c}a, Patr\\'icia Pereira","submitted_at":"2023-08-31T15:19:28Z","abstract_excerpt":"Despite significant research effort in the development of automatic dialogue evaluation metrics, little thought is given to evaluating dialogues other than in English. At the same time, ensuring metrics are invariant to semantically similar responses is also an overlooked topic. In order to achieve the desired properties of robustness and multilinguality for dialogue evaluation metrics, we propose a novel framework that takes advantage of the strengths of current evaluation models with the newly-established paradigm of prompting Large Language Models (LLMs). Empirical results show our framewor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.16797","kind":"arxiv","version":2},"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/2308.16797/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":"2026-07-05T06:48:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TNfq2IdIKXbzJTyz53ykju130eXBeMNsXSp4FyY0uXMfdJ13BNpR7WtCQn17QhOuT2GdGo8HEa6Zskjlh/IOBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T19:23:15.811183Z"},"content_sha256":"70687929552e825ff055086353f45aeadaa20e7fcbf277854a39750dddc08ece","schema_version":"1.0","event_id":"sha256:70687929552e825ff055086353f45aeadaa20e7fcbf277854a39750dddc08ece"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GK6ISBHT3K4GPZVRU7LRJHSYO4/bundle.json","state_url":"https://pith.science/pith/GK6ISBHT3K4GPZVRU7LRJHSYO4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GK6ISBHT3K4GPZVRU7LRJHSYO4/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":"2026-07-06T19:23:15Z","links":{"resolver":"https://pith.science/pith/GK6ISBHT3K4GPZVRU7LRJHSYO4","bundle":"https://pith.science/pith/GK6ISBHT3K4GPZVRU7LRJHSYO4/bundle.json","state":"https://pith.science/pith/GK6ISBHT3K4GPZVRU7LRJHSYO4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GK6ISBHT3K4GPZVRU7LRJHSYO4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:GK6ISBHT3K4GPZVRU7LRJHSYO4","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":"77d22fd3a43cfdbbb2ce95a93d686f4de070ae0bf1826128679497cfb222b125","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-08-31T15:19:28Z","title_canon_sha256":"abe2f57c3f367364a6aceeb82d130603d7690dc2c1d844db5f9e21ed73aab769"},"schema_version":"1.0","source":{"id":"2308.16797","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.16797","created_at":"2026-07-05T06:48:51Z"},{"alias_kind":"arxiv_version","alias_value":"2308.16797v2","created_at":"2026-07-05T06:48:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.16797","created_at":"2026-07-05T06:48:51Z"},{"alias_kind":"pith_short_12","alias_value":"GK6ISBHT3K4G","created_at":"2026-07-05T06:48:51Z"},{"alias_kind":"pith_short_16","alias_value":"GK6ISBHT3K4GPZVR","created_at":"2026-07-05T06:48:51Z"},{"alias_kind":"pith_short_8","alias_value":"GK6ISBHT","created_at":"2026-07-05T06:48:51Z"}],"graph_snapshots":[{"event_id":"sha256:70687929552e825ff055086353f45aeadaa20e7fcbf277854a39750dddc08ece","target":"graph","created_at":"2026-07-05T06:48:51Z","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/2308.16797/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Despite significant research effort in the development of automatic dialogue evaluation metrics, little thought is given to evaluating dialogues other than in English. At the same time, ensuring metrics are invariant to semantically similar responses is also an overlooked topic. In order to achieve the desired properties of robustness and multilinguality for dialogue evaluation metrics, we propose a novel framework that takes advantage of the strengths of current evaluation models with the newly-established paradigm of prompting Large Language Models (LLMs). Empirical results show our framewor","authors_text":"Alon Lavie, Helena Moniz, Isabel Trancoso, Jo\\~ao Paulo Carvalho, John Mendon\\c{c}a, Patr\\'icia Pereira","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-08-31T15:19:28Z","title":"Simple LLM Prompting is State-of-the-Art for Robust and Multilingual Dialogue Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.16797","kind":"arxiv","version":2},"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":"sha256:13725b1b6975c3288391ef340c8e6a5339a1eeab46f73e974d33fe927f65da76","target":"record","created_at":"2026-07-05T06:48:51Z","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":"77d22fd3a43cfdbbb2ce95a93d686f4de070ae0bf1826128679497cfb222b125","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-08-31T15:19:28Z","title_canon_sha256":"abe2f57c3f367364a6aceeb82d130603d7690dc2c1d844db5f9e21ed73aab769"},"schema_version":"1.0","source":{"id":"2308.16797","kind":"arxiv","version":2}},"canonical_sha256":"32bc8904f3dab867e6b1a7d7149e587733965f30c94bcc05ed920438907951fa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"32bc8904f3dab867e6b1a7d7149e587733965f30c94bcc05ed920438907951fa","first_computed_at":"2026-07-05T06:48:51.620976Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:48:51.620976Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SBB8NEJI4DqYM0IJfbhKlRh5I3BM2qT69L8jCrXhcTNES5x+itkqZhyjIHQ25qOUxHvuWLUvknpeqUpfn2ZPAw==","signature_status":"signed_v1","signed_at":"2026-07-05T06:48:51.621443Z","signed_message":"canonical_sha256_bytes"},"source_id":"2308.16797","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:13725b1b6975c3288391ef340c8e6a5339a1eeab46f73e974d33fe927f65da76","sha256:70687929552e825ff055086353f45aeadaa20e7fcbf277854a39750dddc08ece"],"state_sha256":"928872bbeca6bc67af6d0b7da0fc82f0a6101cb009f295f96accfd6712b4306f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Lr1wNqFjfwbR6eCjS/Ji0vpaordFzwIOWqjn29LSypATwXqDh26qrED5FcLjauCdUhquIR2nHhcTLW0JoKJ1DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T19:23:15.813155Z","bundle_sha256":"42e525304887722a6ba4e409afd295d6a9c77dc9df310d95b31ea942b411c13d"}}