{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XY2SJYIPBXIRDME4Q56R7IT74F","short_pith_number":"pith:XY2SJYIP","schema_version":"1.0","canonical_sha256":"be3524e10f0dd111b09c877d1fa27fe156aaed1e0e136c00afcf81d4f333201e","source":{"kind":"arxiv","id":"2506.19952","version":2},"attestation_state":"computed","paper":{"title":"CycleDistill: Bootstrapping Machine Translation using LLMs with Cyclical Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Deepon Halder, Raj Dabre, Thanmay Jayakumar","submitted_at":"2025-06-24T18:56:57Z","abstract_excerpt":"Large language models (LLMs), despite their ability to perform few-shot machine translation (MT), often lag behind dedicated MT systems trained on parallel corpora, which are crucial for high quality machine translation (MT). However, parallel corpora are often scarce or non-existent for low-resource languages. In this paper, we propose CycleDistill, a bootstrapping approach leveraging LLMs and few-shot translation to obtain high-quality MT systems. CycleDistill involves iteratively generating synthetic parallel corpora from monolingual corpora via zero- or few-shot MT, which is then used to f"},"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":"2506.19952","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-24T18:56:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fa86c23401e5f8da826b7227d13e269e76037b089248cc7d62f9be6b356cb0e2","abstract_canon_sha256":"6a730f46aa43c093b64a1911e4ea11e9171ce14ff665a7a4e946a6eb4b770820"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:30.940162Z","signature_b64":"E5hB14lfjKhkkOd7Rooj4cbGed9/XHFxUOUZY5gi/rRjcJau4fTGlnvYreSFGqY6W5Q9+8iXVBmKPtdqKkqtDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be3524e10f0dd111b09c877d1fa27fe156aaed1e0e136c00afcf81d4f333201e","last_reissued_at":"2026-07-05T11:51:30.939676Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:30.939676Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CycleDistill: Bootstrapping Machine Translation using LLMs with Cyclical Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Deepon Halder, Raj Dabre, Thanmay Jayakumar","submitted_at":"2025-06-24T18:56:57Z","abstract_excerpt":"Large language models (LLMs), despite their ability to perform few-shot machine translation (MT), often lag behind dedicated MT systems trained on parallel corpora, which are crucial for high quality machine translation (MT). However, parallel corpora are often scarce or non-existent for low-resource languages. In this paper, we propose CycleDistill, a bootstrapping approach leveraging LLMs and few-shot translation to obtain high-quality MT systems. CycleDistill involves iteratively generating synthetic parallel corpora from monolingual corpora via zero- or few-shot MT, which is then used to f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.19952","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/2506.19952/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":"2506.19952","created_at":"2026-07-05T11:51:30.939734+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.19952v2","created_at":"2026-07-05T11:51:30.939734+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.19952","created_at":"2026-07-05T11:51:30.939734+00:00"},{"alias_kind":"pith_short_12","alias_value":"XY2SJYIPBXIR","created_at":"2026-07-05T11:51:30.939734+00:00"},{"alias_kind":"pith_short_16","alias_value":"XY2SJYIPBXIRDME4","created_at":"2026-07-05T11:51:30.939734+00:00"},{"alias_kind":"pith_short_8","alias_value":"XY2SJYIP","created_at":"2026-07-05T11:51:30.939734+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XY2SJYIPBXIRDME4Q56R7IT74F","json":"https://pith.science/pith/XY2SJYIPBXIRDME4Q56R7IT74F.json","graph_json":"https://pith.science/api/pith-number/XY2SJYIPBXIRDME4Q56R7IT74F/graph.json","events_json":"https://pith.science/api/pith-number/XY2SJYIPBXIRDME4Q56R7IT74F/events.json","paper":"https://pith.science/paper/XY2SJYIP"},"agent_actions":{"view_html":"https://pith.science/pith/XY2SJYIPBXIRDME4Q56R7IT74F","download_json":"https://pith.science/pith/XY2SJYIPBXIRDME4Q56R7IT74F.json","view_paper":"https://pith.science/paper/XY2SJYIP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.19952&json=true","fetch_graph":"https://pith.science/api/pith-number/XY2SJYIPBXIRDME4Q56R7IT74F/graph.json","fetch_events":"https://pith.science/api/pith-number/XY2SJYIPBXIRDME4Q56R7IT74F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XY2SJYIPBXIRDME4Q56R7IT74F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XY2SJYIPBXIRDME4Q56R7IT74F/action/storage_attestation","attest_author":"https://pith.science/pith/XY2SJYIPBXIRDME4Q56R7IT74F/action/author_attestation","sign_citation":"https://pith.science/pith/XY2SJYIPBXIRDME4Q56R7IT74F/action/citation_signature","submit_replication":"https://pith.science/pith/XY2SJYIPBXIRDME4Q56R7IT74F/action/replication_record"}},"created_at":"2026-07-05T11:51:30.939734+00:00","updated_at":"2026-07-05T11:51:30.939734+00:00"}