{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:IO6BGMBSDNBOQ3VUQNEQGSF3Q5","short_pith_number":"pith:IO6BGMBS","schema_version":"1.0","canonical_sha256":"43bc1330321b42e86eb483490348bb87615d1b3f1a13e370942944959ed0b364","source":{"kind":"arxiv","id":"2410.11079","version":1},"attestation_state":"computed","paper":{"title":"Code-Mixer Ya Nahi: Novel Approaches to Measuring Multilingual LLMs' Code-Mixing Capabilities","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Akhil Bhogal, Ayushman Gupta, Kripabandhu Ghosh","submitted_at":"2024-10-14T20:40:36Z","abstract_excerpt":"Multilingual Large Language Models (LLMs) have demonstrated exceptional performance in Machine Translation (MT) tasks. However, their MT abilities in the context of code-switching (the practice of mixing two or more languages in an utterance) remain under-explored. In this paper, we introduce Rule-Based Prompting, a novel prompting technique to generate code-mixed sentences. We measure and compare the code-mixed MT abilities of 3 popular multilingual LLMs: GPT-3.5-turbo, GPT-4, and Gemini Pro across five language pairs: English-{Hindi, Bengali, Gujarati, French, Spanish} using $k$-shot prompti"},"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":"2410.11079","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-14T20:40:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a030e54e7f7f22f9a0e3d324aba35ceac5857da6ae41fdc23e3404b90ae84058","abstract_canon_sha256":"8bc12d169634fa6347ad8b91c33893c523714d4dbe954266acfb9e7b0a40e465"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:20:28.904589Z","signature_b64":"wgNm5vi7vIUSDQhaWN6OSD8pAwIqIUvN6DP5DhqaB7bRQVCe83vCrl28PJE6kdkFHD60hzJytuL5OD5UbOCwBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43bc1330321b42e86eb483490348bb87615d1b3f1a13e370942944959ed0b364","last_reissued_at":"2026-07-05T09:20:28.904201Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:20:28.904201Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Code-Mixer Ya Nahi: Novel Approaches to Measuring Multilingual LLMs' Code-Mixing Capabilities","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Akhil Bhogal, Ayushman Gupta, Kripabandhu Ghosh","submitted_at":"2024-10-14T20:40:36Z","abstract_excerpt":"Multilingual Large Language Models (LLMs) have demonstrated exceptional performance in Machine Translation (MT) tasks. However, their MT abilities in the context of code-switching (the practice of mixing two or more languages in an utterance) remain under-explored. In this paper, we introduce Rule-Based Prompting, a novel prompting technique to generate code-mixed sentences. We measure and compare the code-mixed MT abilities of 3 popular multilingual LLMs: GPT-3.5-turbo, GPT-4, and Gemini Pro across five language pairs: English-{Hindi, Bengali, Gujarati, French, Spanish} using $k$-shot prompti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.11079","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/2410.11079/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":"2410.11079","created_at":"2026-07-05T09:20:28.904258+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.11079v1","created_at":"2026-07-05T09:20:28.904258+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.11079","created_at":"2026-07-05T09:20:28.904258+00:00"},{"alias_kind":"pith_short_12","alias_value":"IO6BGMBSDNBO","created_at":"2026-07-05T09:20:28.904258+00:00"},{"alias_kind":"pith_short_16","alias_value":"IO6BGMBSDNBOQ3VU","created_at":"2026-07-05T09:20:28.904258+00:00"},{"alias_kind":"pith_short_8","alias_value":"IO6BGMBS","created_at":"2026-07-05T09:20:28.904258+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30790","citing_title":"Indi-RomCoM: Code-Mixed Benchmark for Evaluating LLMs on Romanized Indic-English Instructions","ref_index":80,"is_internal_anchor":false},{"citing_arxiv_id":"2511.10670","citing_title":"Towards Fine-Grained Code-Switch Speech Translation with Semantic Space Alignment","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2602.11181","citing_title":"Code Mixologist : A Practitioner's Guide to Building Code-Mixed LLMs","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IO6BGMBSDNBOQ3VUQNEQGSF3Q5","json":"https://pith.science/pith/IO6BGMBSDNBOQ3VUQNEQGSF3Q5.json","graph_json":"https://pith.science/api/pith-number/IO6BGMBSDNBOQ3VUQNEQGSF3Q5/graph.json","events_json":"https://pith.science/api/pith-number/IO6BGMBSDNBOQ3VUQNEQGSF3Q5/events.json","paper":"https://pith.science/paper/IO6BGMBS"},"agent_actions":{"view_html":"https://pith.science/pith/IO6BGMBSDNBOQ3VUQNEQGSF3Q5","download_json":"https://pith.science/pith/IO6BGMBSDNBOQ3VUQNEQGSF3Q5.json","view_paper":"https://pith.science/paper/IO6BGMBS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.11079&json=true","fetch_graph":"https://pith.science/api/pith-number/IO6BGMBSDNBOQ3VUQNEQGSF3Q5/graph.json","fetch_events":"https://pith.science/api/pith-number/IO6BGMBSDNBOQ3VUQNEQGSF3Q5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IO6BGMBSDNBOQ3VUQNEQGSF3Q5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IO6BGMBSDNBOQ3VUQNEQGSF3Q5/action/storage_attestation","attest_author":"https://pith.science/pith/IO6BGMBSDNBOQ3VUQNEQGSF3Q5/action/author_attestation","sign_citation":"https://pith.science/pith/IO6BGMBSDNBOQ3VUQNEQGSF3Q5/action/citation_signature","submit_replication":"https://pith.science/pith/IO6BGMBSDNBOQ3VUQNEQGSF3Q5/action/replication_record"}},"created_at":"2026-07-05T09:20:28.904258+00:00","updated_at":"2026-07-05T09:20:28.904258+00:00"}