# Toward Safe and Human-Aligned Game Conversational Recommendation via Multi-Agent Decomposition

arXiv: https://arxiv.org/abs/2504.20094 · HTML: https://mlsanity.com/paper/2504.20094/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Agentic recommender systems, one of 16 directions in LLM agents and tool use.
- Builds on 24 of its 49 references that are in the map.
- First posted on arXiv 26 Apr 2025.

## Placed in

- LLM agents and tool use / [Agentic recommender systems](https://mlsanity.com/agents/agentic-recommender-systems/)

## Builds on (most cited first, 24 of 24)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [A Survey on Large Language Model based Autonomous Agents](https://mlsanity.com/paper/2308.11432/) (2023)
- [A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions](https://mlsanity.com/paper/2311.05232/) (2023)
- [Wide & Deep Learning for Recommender Systems](https://mlsanity.com/paper/1606.07792/) (2016)
- [Can Large Language Models Be an Alternative to Human Evaluations?](https://mlsanity.com/paper/2305.01937/) (2023)
- [Prompt Injection attack against LLM-integrated Applications](https://mlsanity.com/paper/2306.05499/) (2023)
- [Windows Agent Arena: Evaluating Multi-Modal OS Agents at Scale](https://mlsanity.com/paper/2409.08264/) (2024)
- [Defending Against Alignment-Breaking Attacks via Robustly Aligned LLM](https://mlsanity.com/paper/2309.14348/) (2023)
- [Towards Deep Conversational Recommendations](https://mlsanity.com/paper/1812.07617/) (2018)
- [A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systems](https://mlsanity.com/paper/2402.18013/) (2024)
- [A Survey on Conversational Recommender Systems](https://mlsanity.com/paper/2004.00646/) (2020)
- [Can LLM be a Personalized Judge?](https://mlsanity.com/paper/2406.11657/) (2024)
- [Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems](https://mlsanity.com/paper/2002.09102/) (2020)
- [Bias and Unfairness in Information Retrieval Systems: New Challenges in the LLM Era](https://mlsanity.com/paper/2404.11457/) (2024)
- [MACRec: a Multi-Agent Collaboration Framework for Recommendation](https://mlsanity.com/paper/2402.15235/) (2024)
- [A Multi-Agent Conversational Recommender System](https://mlsanity.com/paper/2402.01135/) (2024)
- [Knowledge Solver: Teaching LLMs to Search for Domain Knowledge from Knowledge Graphs](https://mlsanity.com/paper/2309.03118/) (2023)
- [A Survey of Personalization: From RAG to Agent](https://mlsanity.com/paper/2504.10147/) (2025)
- [ToxiCraft: A Novel Framework for Synthetic Generation of Harmful Information](https://mlsanity.com/paper/2409.14740/) (2024)
- [Improving Conversational Recommendation Systems via Counterfactual Data Simulation](https://mlsanity.com/paper/2306.02842/) (2023)
- [PANGeA: Procedural Artificial Narrative using Generative AI for Turn-Based Video Games](https://mlsanity.com/paper/2404.19721/) (2024)
- [Empowering Retrieval-based Conversational Recommendation with Contrasting User Preferences](https://mlsanity.com/paper/2503.22005/) (2025)
- [OMuleT: Orchestrating Multiple Tools for Practicable Conversational Recommendation](https://mlsanity.com/paper/2411.19352/) (2024)

## Built on by (most cited first, 9 of 9)

- [Value of Information: A Framework for Human-Agent Communication](https://mlsanity.com/paper/2601.06407/) (2026)
- [TRIDENT: Benchmarking LLM Safety in Finance, Medicine, and Law](https://mlsanity.com/paper/2507.21134/) (2025)
- [Privacy-R1: Privacy-Aware Multi-LLM Agent Collaboration via Reinforcement Learning](https://mlsanity.com/paper/2510.16054/) (2025)
- [Retrieval Augmented Conversational Recommendation with Reinforcement Learning](https://mlsanity.com/paper/2604.04457/) (2026)
- [τ-Rec: A Verifiable Benchmark for Agentic Recommender Systems](https://mlsanity.com/paper/2606.10156/) (2026)
- [Steer Model beyond Assistant: Controlling System Prompt Strength via Contrastive Decoding](https://mlsanity.com/paper/2601.06403/) (2026)
- [Agentic Evaluation of Copyright Law Compliance](https://mlsanity.com/paper/2607.21799/) (2026)
- [Can We Steer the Black-Box? Towards Controllability-Centric Evaluation of Recommender Systems with Collaborative Agents](https://mlsanity.com/paper/2607.13418/) (2026)
- [Attacking and Defending Multi-Agent Collaborative Filtering Systems Through Connectivity](https://mlsanity.com/paper/2608.03272/) (2026)

## Abstract (arXiv)

Conversational recommender systems (CRS) have advanced with large language models, showing strong results in domains like movies. These domains typically involve fixed content and passive consumption, where user preferences can be matched by genre or theme. In contrast, games present distinct challenges: fast-evolving catalogs, interaction-driven preferences (e.g., skill level, mechanics, hardware), and increased risk of unsafe responses in open-ended conversation. We propose MATCHA, a multi-agent framework for CRS that assigns specialized agents for intent parsing, tool-augmented retrieval, multi-LLM ranking with reflection, explanation, and risk control which enabling finer personalization, long-tail coverage, and stronger safety. Evaluated on real user request dataset, MATCHA outperforms six baselines across eight metrics, improving Hit@5 by 20%, reducing popularity bias by 24%, and achieving 97.9% adversarial defense. Human and virtual-judge evaluations confirm improved explanation quality and user alignment.