{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KUILHKME4JRMGESOG5ETZX5ZTC","short_pith_number":"pith:KUILHKME","schema_version":"1.0","canonical_sha256":"5510b3a984e262c3124e37493cdfb9989b9bcf85da5203fd8b05fa1413146ad0","source":{"kind":"arxiv","id":"2407.02552","version":1},"attestation_state":"computed","paper":{"title":"RLHF Can Speak Many Languages: Unlocking Multilingual Preference Optimization for LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ahmet \\\"Ust\\\"un, Arash Ahmadian, John Dang, Julia Kreutzer, Kelly Marchisio, Sara Hooker","submitted_at":"2024-07-02T17:42:30Z","abstract_excerpt":"Preference optimization techniques have become a standard final stage for training state-of-art large language models (LLMs). However, despite widespread adoption, the vast majority of work to-date has focused on first-class citizen languages like English and Chinese. This captures a small fraction of the languages in the world, but also makes it unclear which aspects of current state-of-the-art research transfer to a multilingual setting. In this work, we perform an exhaustive study to achieve a new state-of-the-art in aligning multilingual LLMs. We introduce a novel, scalable method for gene"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2407.02552","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-02T17:42:30Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"2fd2066c6c54363bbb82566dd254a9f5701607acbde487121266c289ffce21df","abstract_canon_sha256":"3e703c3b5987e6b91eeaca27d8f2bad4ec334284035a47e850f6ca3659934969"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:39:38.189604Z","signature_b64":"a9ObtEmUW3VRdVbPqr7EOsHq4RCnWI4NG+iFaRl9ZHwXXhL/flqnrMGWelJq3LhUgMuCg4LxUo7xu3lVnQUVDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5510b3a984e262c3124e37493cdfb9989b9bcf85da5203fd8b05fa1413146ad0","last_reissued_at":"2026-07-05T08:39:38.189179Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:39:38.189179Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RLHF Can Speak Many Languages: Unlocking Multilingual Preference Optimization for LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ahmet \\\"Ust\\\"un, Arash Ahmadian, John Dang, Julia Kreutzer, Kelly Marchisio, Sara Hooker","submitted_at":"2024-07-02T17:42:30Z","abstract_excerpt":"Preference optimization techniques have become a standard final stage for training state-of-art large language models (LLMs). However, despite widespread adoption, the vast majority of work to-date has focused on first-class citizen languages like English and Chinese. This captures a small fraction of the languages in the world, but also makes it unclear which aspects of current state-of-the-art research transfer to a multilingual setting. In this work, we perform an exhaustive study to achieve a new state-of-the-art in aligning multilingual LLMs. We introduce a novel, scalable method for gene"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.02552","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/2407.02552/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2407.02552","created_at":"2026-07-05T08:39:38.189231+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.02552v1","created_at":"2026-07-05T08:39:38.189231+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.02552","created_at":"2026-07-05T08:39:38.189231+00:00"},{"alias_kind":"pith_short_12","alias_value":"KUILHKME4JRM","created_at":"2026-07-05T08:39:38.189231+00:00"},{"alias_kind":"pith_short_16","alias_value":"KUILHKME4JRMGESO","created_at":"2026-07-05T08:39:38.189231+00:00"},{"alias_kind":"pith_short_8","alias_value":"KUILHKME","created_at":"2026-07-05T08:39:38.189231+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2504.01919","citing_title":"Bridging the Linguistic Divide: A Survey on Leveraging Large Language Models for Machine Translation","ref_index":53,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KUILHKME4JRMGESOG5ETZX5ZTC","json":"https://pith.science/pith/KUILHKME4JRMGESOG5ETZX5ZTC.json","graph_json":"https://pith.science/api/pith-number/KUILHKME4JRMGESOG5ETZX5ZTC/graph.json","events_json":"https://pith.science/api/pith-number/KUILHKME4JRMGESOG5ETZX5ZTC/events.json","paper":"https://pith.science/paper/KUILHKME"},"agent_actions":{"view_html":"https://pith.science/pith/KUILHKME4JRMGESOG5ETZX5ZTC","download_json":"https://pith.science/pith/KUILHKME4JRMGESOG5ETZX5ZTC.json","view_paper":"https://pith.science/paper/KUILHKME","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.02552&json=true","fetch_graph":"https://pith.science/api/pith-number/KUILHKME4JRMGESOG5ETZX5ZTC/graph.json","fetch_events":"https://pith.science/api/pith-number/KUILHKME4JRMGESOG5ETZX5ZTC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KUILHKME4JRMGESOG5ETZX5ZTC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KUILHKME4JRMGESOG5ETZX5ZTC/action/storage_attestation","attest_author":"https://pith.science/pith/KUILHKME4JRMGESOG5ETZX5ZTC/action/author_attestation","sign_citation":"https://pith.science/pith/KUILHKME4JRMGESOG5ETZX5ZTC/action/citation_signature","submit_replication":"https://pith.science/pith/KUILHKME4JRMGESOG5ETZX5ZTC/action/replication_record"}},"created_at":"2026-07-05T08:39:38.189231+00:00","updated_at":"2026-07-05T08:39:38.189231+00:00"}