{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:RCQJTJEOT3OAC2UV2YLR2JJPU7","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":"2f77d85544eec36d3d6d2780d2eb5b09b5f657e9e77436aee8ecce54c9e8fddf","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-15T14:11:23Z","title_canon_sha256":"39c6359d0f8c9dadc27ce9056d049af581a6414c86214c834116b86fd0ad1d23"},"schema_version":"1.0","source":{"id":"2605.15976","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15976","created_at":"2026-05-20T00:01:47Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15976v1","created_at":"2026-05-20T00:01:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15976","created_at":"2026-05-20T00:01:47Z"},{"alias_kind":"pith_short_12","alias_value":"RCQJTJEOT3OA","created_at":"2026-05-20T00:01:47Z"},{"alias_kind":"pith_short_16","alias_value":"RCQJTJEOT3OAC2UV","created_at":"2026-05-20T00:01:47Z"},{"alias_kind":"pith_short_8","alias_value":"RCQJTJEO","created_at":"2026-05-20T00:01:47Z"}],"graph_snapshots":[{"event_id":"sha256:fb584e1f0a9b2f604649b72c8b0d866e7a20ccc1d732a4fa026e7a95eaac69e2","target":"graph","created_at":"2026-05-20T00:01:47Z","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":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T17:33:44.864180Z","status":"skipped","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T17:01:55.683966Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.15976/integrity.json","findings":[],"snapshot_sha256":"b7ba2e7b7a19dbeacc06af55080bfb9d356b99332f1a04a1ca8137255a560695","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Production machine translation relies overwhelmingly on encoder-decoder Seq2Seq models, yet reinforcement learning approaches to MT fine-tuning have largely targeted decoder-only LLMs at $\\geq$7B parameters, with limited systematic study of encoder-decoder architectures. We apply Group Relative Policy Optimization to NLLB-200 (600M and 1.3B) using a hybrid reference-free reward (LaBSE and COMET-Kiwi) that requires no parallel data at fine-tuning time, evaluating across 13 typologically diverse languages. GRPO yields consistent improvements on all 13 languages, up to $+$5.03 chrF++ for Traditio","authors_text":"Carlos Escolano, Ernesto Garcia-Estrada, Jos\\'e A. R. Fonallosa","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-15T14:11:23Z","title":"Reference-Free Reinforcement Learning Fine-Tuning for MT: A Seq2Seq Perspective"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.15976","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:f80a783c39c7c7390aa5aaaf3f045780f9ed78b40ed41e3b21234a70bda40f7c","target":"record","created_at":"2026-05-20T00:01:47Z","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":"2f77d85544eec36d3d6d2780d2eb5b09b5f657e9e77436aee8ecce54c9e8fddf","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-15T14:11:23Z","title_canon_sha256":"39c6359d0f8c9dadc27ce9056d049af581a6414c86214c834116b86fd0ad1d23"},"schema_version":"1.0","source":{"id":"2605.15976","kind":"arxiv","version":1}},"canonical_sha256":"88a099a48e9edc016a95d6171d252fa7e4b598faeb2847d8bc881e561c9b1538","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"88a099a48e9edc016a95d6171d252fa7e4b598faeb2847d8bc881e561c9b1538","first_computed_at":"2026-05-20T00:01:47.338652Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:01:47.338652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Nyb+FaZEpywBAFb9FAozsZioP12uJI8qf2mu67rA4qg7yL9EMmEyktn4+im9KEO+lZaVXeZknbCiv5BFCqJeCQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:01:47.339264Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.15976","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f80a783c39c7c7390aa5aaaf3f045780f9ed78b40ed41e3b21234a70bda40f7c","sha256:fb584e1f0a9b2f604649b72c8b0d866e7a20ccc1d732a4fa026e7a95eaac69e2"],"state_sha256":"b382618c42ce0be5c0ea039fd902b75404e4564eb62608e2e60fa003c9f86f94"}