{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XIJWZUHOVRAMXG4HYE3FOWWHWD","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":"7cee53c45dbb9d6ae6d8bf1545b0a1c0142dd80dfc8650ce461f9214283f2c7a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-02-03T10:22:55Z","title_canon_sha256":"c46dab859a6e8d16ed18f2e3a34a9c787cbd2d950c55f5e951a9971037142a5d"},"schema_version":"1.0","source":{"id":"2602.03352","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.03352","created_at":"2026-05-20T00:04:24Z"},{"alias_kind":"arxiv_version","alias_value":"2602.03352v2","created_at":"2026-05-20T00:04:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.03352","created_at":"2026-05-20T00:04:24Z"},{"alias_kind":"pith_short_12","alias_value":"XIJWZUHOVRAM","created_at":"2026-05-20T00:04:24Z"},{"alias_kind":"pith_short_16","alias_value":"XIJWZUHOVRAMXG4H","created_at":"2026-05-20T00:04:24Z"},{"alias_kind":"pith_short_8","alias_value":"XIJWZUHO","created_at":"2026-05-20T00:04:24Z"}],"graph_snapshots":[{"event_id":"sha256:f560221e1924d6704655bf6aa53ee8a148620e46ca4f849daef748bdf85a0194","target":"graph","created_at":"2026-05-20T00:04:24Z","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/2602.03352/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning (RL) has shown strong promise for LLM-based machine translation, with recent methods such as GRPO demonstrating notable gains; nevertheless, translation-oriented RL remains challenged by noisy learning signals arising from Monte Carlo return estimation, as well as a large trajectory space that favors global exploration over fine-grained local optimization. We introduce \\textbf{PEGRL}, a \\textit{two-stage} RL framework that uses post-editing as an auxiliary task to stabilize training and guide overall optimization. At each iteration, translation outputs are sampled to con","authors_text":"Hao Zhou, Junlan Feng, Shujian Huang, Xin Huang, Xue Han, Yunzhi Shen","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-02-03T10:22:55Z","title":"PEGRL: Improving Machine Translation by Post-Editing Guided Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.03352","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":"sha256:45437df6e5ef47d17da753e5cd222b2abfc5e38fe0d3a027ffcb4610855bc3b7","target":"record","created_at":"2026-05-20T00:04:24Z","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":"7cee53c45dbb9d6ae6d8bf1545b0a1c0142dd80dfc8650ce461f9214283f2c7a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-02-03T10:22:55Z","title_canon_sha256":"c46dab859a6e8d16ed18f2e3a34a9c787cbd2d950c55f5e951a9971037142a5d"},"schema_version":"1.0","source":{"id":"2602.03352","kind":"arxiv","version":2}},"canonical_sha256":"ba136cd0eeac40cb9b87c136575ac7b0eda29eee296e6b2097e7325ad0324541","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ba136cd0eeac40cb9b87c136575ac7b0eda29eee296e6b2097e7325ad0324541","first_computed_at":"2026-05-20T00:04:24.119095Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:24.119095Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eahCLaNuKUjAdU8q1rytVOauHHkhAM1rnQYoj6UlntY7xo/6d2awNrHTKC8EdyffBbGbensfElvaZi3MIds3CQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:24.119844Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.03352","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:45437df6e5ef47d17da753e5cd222b2abfc5e38fe0d3a027ffcb4610855bc3b7","sha256:f560221e1924d6704655bf6aa53ee8a148620e46ca4f849daef748bdf85a0194"],"state_sha256":"cde9cde42d7da4ec9de447bf9fc5533d64bd082c505bd3f55816e4a62da21166"}