aiqbench analysis
This research introduces CHAI, a framework that significantly boosts large language model performance in code-mixed translation by combining LLM-based annotation with reinforcement learning from AI feedback. By generating synthetic preference data and incorporating domain knowledge as a constitutional guide, the method achieves a 68.45% win rate over existing open-source models on real-world datasets. The work addresses a critical gap in multilingual NLP, offering a scalable path toward more inclusive and robust language models for diverse linguistic environments.
CHAI for LLMs: Improving Code-Mixed Translation in Large Language Models through Reinforcement Learning with AI Feedback
arXiv:2411.09073v4 Announce Type: replace-cross Abstract: Large language models (LLMs) show strong performance across many tasks but remain weak at understanding code-mixed (CM) language. Despite this limitation, improving LLMs for CM tasks has received little attention. To address this gap, we propose CHAI, a general-purpose framework for enhancing LLM performance on CM tasks, focusing on CM translation. CHAI leverages four key ideas. First, we investigate the use of LLMs as annotators to address the scarcity of high-quality CM datasets. Second, we leverage these LLM annotations to generate large-scale preference data and apply reinforcement learning from AI feedback (RLAIF) to improve CM translation. Third, we incorporate LLM-generated domain knowledge as a constitution, enabling iterative response refinement. Fourth, we perform extensive evaluations on real-world datasets and settings. Results show that CHAI-powered LLMs outperform state-of-the-art open-source models by 68.45% on average in human-adjudicated win rate on CM translation tasks. This work is a step toward more inclusive open-source code-mixed LLMs.
Abstract:Large language models (LLMs) show strong performance across many tasks but remain weak at understanding code-mixed (CM) language. Despite this limitation, improving LLMs for CM tasks has received little attention. To address this gap, we propose CHAI, a general-purpose framework for enhancing LLM performance on CM tasks, focusing on CM translation. CHAI leverages four key ideas. First, we investigate the use of LLMs as annotators to address the scarcity of high-quality CM datasets. Second, we leverage these LLM annotations to generate large-scale preference data and apply reinforcement learning from AI feedback (RLAIF) to improve CM translation. Third, we incorporate LLM-generated domain knowledge as a constitution, enabling iterative response refinement. Fourth, we perform extensive evaluations on real-world datasets and settings. Results show that CHAI-powered LLMs outperform state-of-the-art open-source models by 68.45% on average in human-adjudicated win rate on CM translation tasks. This work is a step toward more inclusive open-source code-mixed LLMs.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2411.09073 [cs.CL]
(or arXiv:2411.09073v4 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2411.09073 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 250 words
Related coverage
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · Sep 25, 2026
Reviewing failure modes in science as we prepare for AI researchers
LessWrong · Oct 8, 2026
Building and evaluating model diffing agents
AI Alignment Forum · Jun 12, 2026
RAG Strategies for Natural Language-Based SQL Query and REST API Call Generation
arXiv · Aug 5, 2026