Reinforced Latent Reasoning for LLM-based Recommendation

TL;DR

This paper introduces LatentR3, a reinforcement learning framework that enables large language models to perform efficient, implicit reasoning for recommendation tasks without explicit chain-of-thought data, improving performance and inference speed.

Contribution

It proposes a novel end-to-end RL-based training method for latent reasoning in LLMs, eliminating the need for explicit reasoning data and enhancing recommendation accuracy.

Findings

LatentR3 improves recommendation performance across various LLMs.

The framework reduces inference latency by avoiding explicit reasoning generation.

Reinforcement learning effectively trains latent reasoning modules without supervision.

Abstract

Large Language Models (LLMs) have demonstrated impressive reasoning capabilities in complex problem-solving tasks, sparking growing interest in their application to preference reasoning in recommendation systems. Existing methods typically rely on fine-tuning with explicit chain-of-thought (CoT) data. However, these methods face significant practical limitations due to (1) the difficulty of obtaining high-quality CoT data in recommendation and (2) the high inference latency caused by generating CoT reasoning. In this work, we explore an alternative approach that shifts from explicit CoT reasoning to compact, information-dense latent reasoning. This approach eliminates the need for explicit CoT generation and improves inference efficiency, as few latent tokens can effectively capture the entire reasoning process. Building on this idea, we propose \textit{\underline{R}einforced \underline{Latent} \underline{R}easoning for \underline{R}ecommendation} (LatentR$^3$), a novel end-to-end training framework that leverages reinforcement learning (RL) to optimize latent reasoning without relying on any CoT data. LatentR$^3$ adopts a two-stage training strategy: first, supervised fine-tuning to initialize the latent reasoning module, followed by pure RL training to encourage exploration through a rule-based reward design. Our RL implementation is based on a modified GRPO algorithm, which reduces computational overhead during training and introduces continuous reward signals for more efficient learning. Extensive experiments demonstrate that LatentR$^3$ enables effective latent reasoning without any direct supervision of the reasoning process, significantly improving performance when integrated with different LLM-based recommendation methods. Our codes are available at https://github.com/xuwenxinedu/R3.