Language-Coupled Reinforcement Learning for Multilingual Retrieval-Augmented Generation
This addresses suboptimal performance in multilingual AI systems for users needing accurate cross-lingual knowledge integration, though it appears incremental as it builds on existing retrieval-augmented generation methods.
The paper tackles the problem of knowledge bias and conflict in multilingual retrieval-augmented generation by proposing LcRL, a reinforcement learning framework that integrates language-coupled group sampling and anti-consistency penalties, achieving competitive performance in scenarios like constrained training data and large multilingual collections.
Multilingual retrieval-augmented generation (MRAG) requires models to effectively acquire and integrate beneficial external knowledge from multilingual collections. However, most existing studies employ a unitive process where queries of equivalent semantics across different languages are processed through a single-turn retrieval and subsequent optimization. Such a ``one-size-fits-all'' strategy is often suboptimal in multilingual settings, as the models occur to knowledge bias and conflict during the interaction with the search engine. To alleviate the issues, we propose LcRL, a multilingual search-augmented reinforcement learning framework that integrates a language-coupled Group Relative Policy Optimization into the policy and reward models. We adopt the language-coupled group sampling in the rollout module to reduce knowledge bias, and regularize an auxiliary anti-consistency penalty in the reward models to mitigate the knowledge conflict. Experimental results demonstrate that LcRL not only achieves competitive performance but is also appropriate for various practical scenarios such as constrained training data and retrieval over collections encompassing a large number of languages. Our code is available at https://github.com/Cherry-qwq/LcRL-Open.