{"id":35124,"date":"2026-04-13T14:10:10","date_gmt":"2026-04-13T14:10:10","guid":{"rendered":"https:\/\/lamarr-institute.org\/publication\/is-continuous-cot-better-suited-for-multi-lingual-reasoning\/"},"modified":"2026-06-08T13:17:35","modified_gmt":"2026-06-08T13:17:35","slug":"is-continuous-cot-better-suited-for-multi-lingual-reasoning","status":"publish","type":"publication","link":"https:\/\/lamarr-institute.org\/de\/publication\/is-continuous-cot-better-suited-for-multi-lingual-reasoning\/","title":{"rendered":"Is continuous {CoT} better suited for multi-lingual reasoning?"},"content":{"rendered":"<p>We investigate whether performing reasoning in a continuous latent space leads to more robust multilingual capabilities. We compare Continuous Chain-of-Thought (using the {CODI} framework) against standard supervised fine-tuning across five typologically diverse languages: English, Chinese, German, French, and Urdu. Our experiments on {GSM}8k and {CommonsenseQA} demonstrate that continuous reasoning significantly outperforms explicit reasoning on low-resource languages, particularly in zero-shot settings where the target language was not seen during training. Additionally, this approach achieves extreme efficiency, compressing reasoning traces by approximately $29\\times$ to $50\\times$. These findings indicate that continuous latent representations naturally exhibit greater language invariance, offering a scalable solution for cross-lingual reasoning.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>We investigate whether performing reasoning in a continuous latent space leads to more robust multilingual capabilities. We compare Continuous Chain-of-Thought (using the {CODI} framework) against standard supervised fine-tuning across five typologically diverse languages: English, Chinese, German, French, and Urdu. Our experiments on {GSM}8k and {CommonsenseQA} demonstrate that continuous reasoning significantly outperforms explicit reasoning on low-resource languages, particularly in zero-shot settings where the target language was not seen during training. Additionally, [&hellip;]<\/p>\n","protected":false},"author":12,"featured_media":0,"template":"","meta":{"_acf_changed":false,"footnotes":""},"publication-type":[30],"class_list":["post-35124","publication","type-publication","status-publish","hentry","publication-type-article"],"acf":[],"publishpress_future_workflow_manual_trigger":{"enabledWorkflows":[]},"_links":{"self":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication\/35124","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication"}],"about":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/types\/publication"}],"author":[{"embeddable":true,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/users\/12"}],"version-history":[{"count":0,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication\/35124\/revisions"}],"wp:attachment":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/media?parent=35124"}],"wp:term":[{"taxonomy":"publication-type","embeddable":true,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication-type?post=35124"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}