{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QDT6BMMEZPTEVUAC3PRXJM52NA","short_pith_number":"pith:QDT6BMME","schema_version":"1.0","canonical_sha256":"80e7e0b184cbe64ad002dbe374b3ba6826a1e07857b3f9461e2607f8c95f6131","source":{"kind":"arxiv","id":"2505.12587","version":1},"attestation_state":"computed","paper":{"title":"CMLFormer: A Dual Decoder Transformer with Switching Point Learning for Code-Mixed Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Aditeya Baral, Allen George Ajith, Mrityunjay Abhijeet Bhanja, Roshan Nayak","submitted_at":"2025-05-19T00:50:49Z","abstract_excerpt":"Code-mixed languages, characterized by frequent within-sentence language transitions, present structural challenges that standard language models fail to address. In this work, we propose CMLFormer, an enhanced multi-layer dual-decoder Transformer with a shared encoder and synchronized decoder cross-attention, designed to model the linguistic and semantic dynamics of code-mixed text. CMLFormer is pre-trained on an augmented Hinglish corpus with switching point and translation annotations with multiple new objectives specifically aimed at capturing switching behavior, cross-lingual structure, a"},"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":"2505.12587","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-19T00:50:49Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d1a808ce9f3a8f3fe7aba4c9b86cee041542025921258416b453cfe9c10d73b6","abstract_canon_sha256":"db07f45681bece674d158c148fe58a52f6b7a356157ef27a844cb93ae43063d1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:06.467108Z","signature_b64":"LIb8GSEueXadnrsZmZvdS2LFixYNVwepQDWTJTyDWZoBdr8qJxX9ZVtQ+1uV1tV4Nl29Yr6FDM5v+4e3GRFkCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80e7e0b184cbe64ad002dbe374b3ba6826a1e07857b3f9461e2607f8c95f6131","last_reissued_at":"2026-07-05T11:05:06.466410Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:06.466410Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CMLFormer: A Dual Decoder Transformer with Switching Point Learning for Code-Mixed Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Aditeya Baral, Allen George Ajith, Mrityunjay Abhijeet Bhanja, Roshan Nayak","submitted_at":"2025-05-19T00:50:49Z","abstract_excerpt":"Code-mixed languages, characterized by frequent within-sentence language transitions, present structural challenges that standard language models fail to address. In this work, we propose CMLFormer, an enhanced multi-layer dual-decoder Transformer with a shared encoder and synchronized decoder cross-attention, designed to model the linguistic and semantic dynamics of code-mixed text. CMLFormer is pre-trained on an augmented Hinglish corpus with switching point and translation annotations with multiple new objectives specifically aimed at capturing switching behavior, cross-lingual structure, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.12587","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/2505.12587/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":"2505.12587","created_at":"2026-07-05T11:05:06.466480+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.12587v1","created_at":"2026-07-05T11:05:06.466480+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.12587","created_at":"2026-07-05T11:05:06.466480+00:00"},{"alias_kind":"pith_short_12","alias_value":"QDT6BMMEZPTE","created_at":"2026-07-05T11:05:06.466480+00:00"},{"alias_kind":"pith_short_16","alias_value":"QDT6BMMEZPTEVUAC","created_at":"2026-07-05T11:05:06.466480+00:00"},{"alias_kind":"pith_short_8","alias_value":"QDT6BMME","created_at":"2026-07-05T11:05:06.466480+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.07037","citing_title":"Beyond Monolingual Assumptions: A Survey of Code-Switched NLP in the Era of Large Language Models across Modalities","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2602.11181","citing_title":"Code Mixologist : A Practitioner's Guide to Building Code-Mixed LLMs","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QDT6BMMEZPTEVUAC3PRXJM52NA","json":"https://pith.science/pith/QDT6BMMEZPTEVUAC3PRXJM52NA.json","graph_json":"https://pith.science/api/pith-number/QDT6BMMEZPTEVUAC3PRXJM52NA/graph.json","events_json":"https://pith.science/api/pith-number/QDT6BMMEZPTEVUAC3PRXJM52NA/events.json","paper":"https://pith.science/paper/QDT6BMME"},"agent_actions":{"view_html":"https://pith.science/pith/QDT6BMMEZPTEVUAC3PRXJM52NA","download_json":"https://pith.science/pith/QDT6BMMEZPTEVUAC3PRXJM52NA.json","view_paper":"https://pith.science/paper/QDT6BMME","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.12587&json=true","fetch_graph":"https://pith.science/api/pith-number/QDT6BMMEZPTEVUAC3PRXJM52NA/graph.json","fetch_events":"https://pith.science/api/pith-number/QDT6BMMEZPTEVUAC3PRXJM52NA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QDT6BMMEZPTEVUAC3PRXJM52NA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QDT6BMMEZPTEVUAC3PRXJM52NA/action/storage_attestation","attest_author":"https://pith.science/pith/QDT6BMMEZPTEVUAC3PRXJM52NA/action/author_attestation","sign_citation":"https://pith.science/pith/QDT6BMMEZPTEVUAC3PRXJM52NA/action/citation_signature","submit_replication":"https://pith.science/pith/QDT6BMMEZPTEVUAC3PRXJM52NA/action/replication_record"}},"created_at":"2026-07-05T11:05:06.466480+00:00","updated_at":"2026-07-05T11:05:06.466480+00:00"}