Mistral-C2F: Coarse to Fine Actor for Analytical and Reasoning Enhancement in RLHF and Effective-Merged LLMs

TL;DR

Mistral-C2F introduces a two-step coarse-to-fine actor approach that enhances small LLMs' analytical and reasoning capabilities through RLHF, continuous maximization, and knowledge residue merging, outperforming larger models.

Contribution

The paper proposes a novel coarse-to-fine actor framework with continuous maximization and knowledge residue merging to significantly improve small LLMs' analytical and conversational performance.

Findings

Outperforms similar-scale models on 11 language tasks.

Excels in MT-Bench Dialogue task.

Enhances reasoning and reduces redundancies in small LLMs.

Abstract

Despite the advances in Large Language Models (LLMs), exemplified by models like GPT-4 and Claude, smaller-scale LLMs such as Llama and Mistral often struggle with generating in-depth and coherent dialogues. This paper presents a novel two-step Coarse-to-Fine Actor model to address the inherent limitations in conversational and analytical capabilities of small-sized LLMs. Our approach begins with the Policy-based Coarse Actor, employing a technique we term "Continuous Maximization". The Coarse Actor establishes an enhanced, knowledge-rich pool adept at aligning with human preference styles in analysis and reasoning. Through the RLHF process, it employs Continuous Maximization, a strategy that dynamically and adaptively extends the output length limit, enabling the generation of more detailed and analytical content. Subsequently, the Fine Actor refines this analytical content, addressing the generation of excessively redundant information from the Coarse Actor. We introduce a "Knowledge Residue Merger" approach, refining the content from the Coarse Actor and merging it with an existing Instruction model to improve quality, correctness, and reduce redundancies. We applied our methodology to the popular Mistral model, creating Mistral-C2F, which has demonstrated exceptional performance across 11 general language tasks and the MT-Bench Dialogue task, outperforming similar-scale models and even larger models with 13B and 30B parameters. Our model has significantly improved conversational and analytical reasoning abilities.