RRAML: Reinforced Retrieval Augmented Machine Learning

TL;DR

This paper introduces RRAML, a framework combining reinforcement learning, retrieval, and large language models to improve reasoning, reduce hallucinations, and overcome access limitations of LLMs.

Contribution

It presents a novel reinforcement learning-based approach that integrates retrieval with LLMs, addressing access, retraining, and hallucination issues.

Findings

Effective integration of retrieval and reasoning in LLMs.

Reduced hallucinations and irrelevant information.

No need for gradient access or retraining of LLMs.

Abstract

The emergence of large language models (LLMs) has revolutionized machine learning and related fields, showcasing remarkable abilities in comprehending, generating, and manipulating human language. However, their conventional usage through API-based text prompt submissions imposes certain limitations in terms of context constraints and external source availability. To address these challenges, we propose a novel framework called Reinforced Retrieval Augmented Machine Learning (RRAML). RRAML integrates the reasoning capabilities of LLMs with supporting information retrieved by a purpose-built retriever from a vast user-provided database. By leveraging recent advancements in reinforcement learning, our method effectively addresses several critical challenges. Firstly, it circumvents the need for accessing LLM gradients. Secondly, our method alleviates the burden of retraining LLMs for specific tasks, as it is often impractical or impossible due to restricted access to the model and the computational intensity involved. Additionally we seamlessly link the retriever's task with the reasoner, mitigating hallucinations and reducing irrelevant, and potentially damaging retrieved documents. We believe that the research agenda outlined in this paper has the potential to profoundly impact the field of AI, democratizing access to and utilization of LLMs for a wide range of entities.