{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RM2MY7XEXEIY7XXNCOHA6HF3UX","short_pith_number":"pith:RM2MY7XE","schema_version":"1.0","canonical_sha256":"8b34cc7ee4b9118fdeed138e0f1cbba5ea3100f24367197f91299b478410d4c3","source":{"kind":"arxiv","id":"2501.15219","version":1},"attestation_state":"computed","paper":{"title":"Faster Machine Translation Ensembling with Reinforcement Learning and Competitive Correction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Kritarth Prasad, Mohammadi Zaki, Pankaj Wasnik, Pratik Singh","submitted_at":"2025-01-25T13:50:18Z","abstract_excerpt":"Ensembling neural machine translation (NMT) models to produce higher-quality translations than the $L$ individual models has been extensively studied. Recent methods typically employ a candidate selection block (CSB) and an encoder-decoder fusion block (FB), requiring inference across \\textit{all} candidate models, leading to significant computational overhead, generally $\\Omega(L)$. This paper introduces \\textbf{SmartGen}, a reinforcement learning (RL)-based strategy that improves the CSB by selecting a small, fixed number of candidates and identifying optimal groups to pass to the fusion blo"},"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":"2501.15219","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-25T13:50:18Z","cross_cats_sorted":[],"title_canon_sha256":"719fdbaa9474ac468d4dd470256c4d3706b72272a968597631392c292cf669e0","abstract_canon_sha256":"94d73a20330561cc78fc4ce2a08407d8e662ed8f187730aa0a38ce0cf70ac8e2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:23:22.593878Z","signature_b64":"g+p2jyEsrI/+doGel8ZITdykKM8F8KNUtkjMv7SjIOeT6644LmFHwu4gd1Y9+CJcIQ5qJLmDjc6la9kOOsICAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8b34cc7ee4b9118fdeed138e0f1cbba5ea3100f24367197f91299b478410d4c3","last_reissued_at":"2026-07-05T11:23:22.593278Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:23:22.593278Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Faster Machine Translation Ensembling with Reinforcement Learning and Competitive Correction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Kritarth Prasad, Mohammadi Zaki, Pankaj Wasnik, Pratik Singh","submitted_at":"2025-01-25T13:50:18Z","abstract_excerpt":"Ensembling neural machine translation (NMT) models to produce higher-quality translations than the $L$ individual models has been extensively studied. Recent methods typically employ a candidate selection block (CSB) and an encoder-decoder fusion block (FB), requiring inference across \\textit{all} candidate models, leading to significant computational overhead, generally $\\Omega(L)$. This paper introduces \\textbf{SmartGen}, a reinforcement learning (RL)-based strategy that improves the CSB by selecting a small, fixed number of candidates and identifying optimal groups to pass to the fusion blo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.15219","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/2501.15219/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":"2501.15219","created_at":"2026-07-05T11:23:22.593365+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.15219v1","created_at":"2026-07-05T11:23:22.593365+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.15219","created_at":"2026-07-05T11:23:22.593365+00:00"},{"alias_kind":"pith_short_12","alias_value":"RM2MY7XEXEIY","created_at":"2026-07-05T11:23:22.593365+00:00"},{"alias_kind":"pith_short_16","alias_value":"RM2MY7XEXEIY7XXN","created_at":"2026-07-05T11:23:22.593365+00:00"},{"alias_kind":"pith_short_8","alias_value":"RM2MY7XE","created_at":"2026-07-05T11:23:22.593365+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.15069","citing_title":"In-Domain African Languages Translation Using LLMs and Multi-armed Bandits","ref_index":35,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RM2MY7XEXEIY7XXNCOHA6HF3UX","json":"https://pith.science/pith/RM2MY7XEXEIY7XXNCOHA6HF3UX.json","graph_json":"https://pith.science/api/pith-number/RM2MY7XEXEIY7XXNCOHA6HF3UX/graph.json","events_json":"https://pith.science/api/pith-number/RM2MY7XEXEIY7XXNCOHA6HF3UX/events.json","paper":"https://pith.science/paper/RM2MY7XE"},"agent_actions":{"view_html":"https://pith.science/pith/RM2MY7XEXEIY7XXNCOHA6HF3UX","download_json":"https://pith.science/pith/RM2MY7XEXEIY7XXNCOHA6HF3UX.json","view_paper":"https://pith.science/paper/RM2MY7XE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.15219&json=true","fetch_graph":"https://pith.science/api/pith-number/RM2MY7XEXEIY7XXNCOHA6HF3UX/graph.json","fetch_events":"https://pith.science/api/pith-number/RM2MY7XEXEIY7XXNCOHA6HF3UX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RM2MY7XEXEIY7XXNCOHA6HF3UX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RM2MY7XEXEIY7XXNCOHA6HF3UX/action/storage_attestation","attest_author":"https://pith.science/pith/RM2MY7XEXEIY7XXNCOHA6HF3UX/action/author_attestation","sign_citation":"https://pith.science/pith/RM2MY7XEXEIY7XXNCOHA6HF3UX/action/citation_signature","submit_replication":"https://pith.science/pith/RM2MY7XEXEIY7XXNCOHA6HF3UX/action/replication_record"}},"created_at":"2026-07-05T11:23:22.593365+00:00","updated_at":"2026-07-05T11:23:22.593365+00:00"}