Enhancing LLM Problem Solving via Tutor-Student Multi-Agent Interaction

2026-04-10

AI Summary

This work addresses the challenge of enhancing large language models’ (LLMs’) performance on complex tasks such as programming without relying on stronger models or heterogeneous ensembles. It introduces PETITE, a novel framework that adapts the mentor–student interaction paradigm from developmental psychology into an LLM-based multi-agent system. PETITE instantiates two role-differentiated agents from the same LLM: a student agent that generates and iteratively refines solutions, and a mentor agent that provides structured feedback without access to ground-truth answers. Through multiple rounds of collaborative interaction, the framework unlocks the latent capabilities of a single model. Evaluated on the APPS programming benchmark, PETITE achieves accuracy comparable to or better than state-of-the-art methods like Self-Consistency and Multi-Agent Debate, while substantially reducing token consumption.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Multiagent Learning
• Natural Language Processing: (Large) Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Human cognitive development is shaped not only by individual effort but by structured social interaction, where role-based exchanges such as those between a tutor and a learner, enable solutions that neither could achieve alone. Inspired by these developmental principles, we ask the question whether a tutor-student multi-agent system can create a synergistic effect by pushing Large Language Model (LLM) beyond what it can do within existing frameworks. To test the idea, we adopt autonomous coding problem domain where two agents instantiated from the same LLM assigned asymmetric roles: a student agent generates and iteratively refines solutions, while a tutor agent provides structured evaluative feedback without access to ground-truth answers. In our proposed framework (PETITE), we aim to extract better problem-solving performance from one model by structuring its interaction through complementary roles, rather than relying on stronger supervisory models or heterogeneous ensembles. Our model is evaluated on the APPS coding benchmark against state-of-the-art approaches of Self-Consistency, Self-Refine, Multi-Agent Debate, and Multi-Agent Review. The results show that our model achieves similar or higher accuracy while consuming significantly fewer tokens. These results suggest that developmentally grounded role-differentiated interaction structures provide a principled and resource-efficient paradigm for enhancing LLM problem-solving through structured peer-like interactions. Index Terms- Peer Tutoring, Scaffolding, Large Language Models, Multi-Agent Systems, Code Generation

Problem

• Large Language Models
• Multi-Agent Systems
• Peer Tutoring
• Code Generation
• Problem Solving

Innovation

• Tutor-Student Interaction
• Multi-Agent Systems
• Large Language Models
• Role-Differentiated Interaction
• Code Generation