论文分享 | 智能体相关研究进展

我们从2025-01-01到2025-01-06的25篇文章中精选出5篇优秀的工作分享给读者。

1. Proactive Conversational Agents with Inner Thoughts
2. MAIN-RAG: Multi-Agent Filtering Retrieval-Augmented Generation
3. PSYCHE: A Multi-faceted Patient Simulation Framework for Evaluation of Psychiatric Assessment Conversational Agents
4. Deterministic Model of Incremental Multi-Agent Boltzmann Q-Learning: Transient Cooperation, Metastability, and Oscillations
5. Modelling and Control of Spatial Behaviours in Multi-Agent Systems with Applications to Biology and Robotics

1.Proactive Conversational Agents with Inner Thoughts

Authors: Xingyu Bruce Liu, Shitao Fang, Weiyan Shi, Chien-Sheng Wu, Takeo Igarashi, Xiang 'Anthony' Chen

https://arxiv.org/abs/2501.00383

论文摘要

Oneofthelong-standing aspirations in conversational AI is to allow them to autonomously take initiatives in conversations, i.e., being proactive. This is especially challenging for multi-party conversa tions. Prior NLP research focused mainly on predicting the next speaker from contexts like preceding conversations. In this paper, we demonstrate the limitations of such methods and rethink what  it means for AI to be proactive in multi-party, human-AI conversa tions. We propose that just like humans, rather than merely reacting to turn-taking cues, a proactive AI formulates its own inner thoughts during a conversation, and seeks the right moment to contribute. Through a formative study with 24 participants and inspiration from linguistics and cognitive psychology, we introduce the Inner Thoughts framework. Our framework equips AI with a continuous, covert train of thoughts in parallel to the overt communication process, which enables it to proactively engage by modeling its intrinsic motivation to express these thoughts. We instantiated this framework into two real-time systems: an AI playground web app and a chatbot. Through a technical evaluation and user studies with human participants, our framework significantly surpasses existing baselines on aspects like anthropomorphism, coherence, intelligence, and turn-taking appropriateness.

论文简评

这篇论文深入探讨了内省框架这一概念，其目的是使人工智能（AI）生成内部思考，以便基于内在动机参与对话，而非仅仅响应提示。该框架包含一个形成性研究以及两个实施案例，这些都展示了传统模型与新模型之间的显著差异。文章的创新之处在于其引入的内省概念，这一概念源于认知心理学，并通过实证研究得到了验证。此外，论文还提出了两种实际应用的框架实施，为AI在对话设置中的实践提供了指导原则。总体而言，这篇文章对于推动AI技术的发展具有重要意义，因为它揭示了一种新的、更人性化的交互方式，能够更好地满足用户的需求。

2.MAIN-RAG: Multi-Agent Filtering Retrieval-Augmented Generation

Authors: Chia-Yuan Chang, Zhimeng Jiang, Vineeth Rakesh, Menghai Pan, Chin-Chia Michael Yeh, Guanchu Wang, Mingzhi Hu, Zhichao Xu, Yan Zheng, Mahashweta Das, Na Zou

https://arxiv.org/abs/2501.00332

论文摘要

Large Language Models (LLMs) are becoming essential tools for various natural language processing tasks but often suffer from generating outdated or incorrect information. Retrieval Augmented Generation (RAG) addresses this issue by incorporating external, real-time in formation retrieval to ground LLM responses. However, the existing RAG systems frequently struggle with the quality of retrieval documents, as irrelevant or noisy documents degrade per formance, increase computational overhead, and undermine response reliability. To tackle this problem, we propose Multi-Agent Filtering Retrieval-Augmented Generation (MAIN-RAG), a training-free RAG framework that leverages multiple LLM agents to collaboratively fil ter and score retrieved documents. Specifi cally, MAIN-RAG introduces an adaptive filter ing mechanism that dynamically adjusts the relevance filtering threshold based on score dis tributions, effectively minimizing noise while maintaining high recall of relevant documents. The proposed approach leverages inter-agent consensus to ensure robust document selection without requiring additional training data or f ine-tuning. Experimental results across four QA benchmarks demonstrate that MAIN-RAG consistently outperforms traditional RAG ap proaches, achieving a 2–11% improvement in answer accuracy while reducing the number of irrelevant retrieved documents. Quantita tive analysis further reveals that our approach achieves superior response consistency and an swer accuracy over baseline methods, offer ing a competitive and practical alternative to training-based solutions.

论文简评

这篇论文提出了一个名为MAIN-RAG的新框架，旨在通过利用多个LLM（语言模型）代理过滤和评分检索到的文档，以提高检索增强生成（RAG）系统的性能。该研究提出了一种自适应过滤机制，可以根据文档得分分布调整相关性阈值，从而通过最小化文档检索中的噪声来提升回答的准确性。实验结果表明，在多个问答基准上取得了显著的性能提升，展示了潜在的实际应用潜力。总体来看，这篇论文提供了新颖的方法论，并证明了其在实际应用中的可行性。

3.PSYCHE: A Multi-faceted Patient Simulation Framework for Evaluation of Psychiatric Assessment Conversational Agents

Authors: Jingoo Lee, Kyungho Lim, Young-Chul Jung, Byung-Hoon Kim

https://arxiv.org/abs/2501.01594

论文摘要

Recent advances in large language models (LLMs) have accelerated the development of conversational agents capable of generating human-like responses. Since psychiatric assessments typically involve complex conversational interactions between psychiatrists and patients, there is growing interest in developing LLM-based psychiatric assessment conversational agents (PACAs) that aim to simulate the role of psychiatrists in clinical evaluations. However, standardized methods for benchmarking the clinical appropriateness of PACAs’ interaction with patients still remain underexplored. Here, we propose PSYCHE, a novel framework designed to enable the 1) clinically relevant, 2) ethically safe, 3) cost-efficient, and 4) quantitative evaluation of PACAs. This is achieved by simulating psychiatric patients based on a multi-faceted psychiatric construct that defines the simulated patients’ profiles, histories, and behaviors, which PACAs are expected to assess. We validate the effectiveness of PSYCHE through a study with 10 board-certified psychiatrists, supported by an in-depth analysis of the simulated patient utterances.

论文简评

本文探讨了评估精神病学诊断对话代理（PACAs）有效性的一种框架。该框架被设计为确保临床相关性和伦理安全，并且具有成本效率和定量评估能力。通过专家评估和深入研究，PSYCHE的效力得到了验证。此外，论文详细描述了PSYCHE框架及其实施步骤，有助于理解该框架的应用过程。这些特点使得PSYCHE成为评估PACAs有效性的重要工具，特别是在当前对医疗技术需求日益增长的背景下。

4.Deterministic Model of Incremental Multi-Agent Boltzmann Q-Learning: Transient Cooperation, Metastability, and Oscillations

Authors: David Goll, Jobst Heitzig, Wolfram Barfuss

https://arxiv.org/abs/2501.00160

论文摘要

Multi-Agent Reinforcement Learning involves interacting agents whose learning processes are coupled through their shared environment, giving rise to emergent, collective dynamics that are sensitive to initial conditions and parameter variations. A Dynamical Systems approach, which studies the evolution of multi-component systems over time, has uncovered some of the underlying dynamics by constructing deterministic approximation models of stochastic algorithms. In this work, we demonstrate that even in the simplest case of independent Q-learning with a Boltzmann exploration policy, significant discrepancies arise between the actual algorithm and previous approximations. We elaborate why these models actually approximate interesting variants, simplifying the learning dynamics, rather than the original incremental algorithm. To explain the discrepancies, we introduce a new discrete-time approximation model that explicitly accounts for agents’ update frequencies within the learning process, and show that its dynamics fundamentally differ from the simplified dynamics of prior models. We illustrate the usefulness of our approach by applying it to the question of spontaneous cooperation in social dilemmas, specifically the Prisoner’s Dilemma as the simplest case study. We identify conditions under which the learning behaviour appears as long-term stable cooperation from an external perspective. However, our model shows that this behaviour is merely a metastable transient phase and not a true equilibrium, making it exploitable. We further exemplify how specific parameter settings can significantly exacerbate the moving target problem in independent learning. Through a systematic analysis of our model, we show that increasing the discount factor induces oscillations, preventing convergence to a joint policy. These oscillations arise from a supercritical Neimark-Sacker bifurcation, which transforms the unique stable fixed point into an unstable focus surrounded by a stable limit cycle.

论文简评

该论文是关于多智能体强化学习（MARL）中独立Q-learning动态分析的一项创新研究。作者通过提出一个新的解析模型，解决了这一领域的重要问题——独立Q-learning在多智能体环境下的行为模式。此外，他们还探讨了更新频率对学习行为的影响，并特别关注如囚徒困境等场景下合作的学习表现。研究结果表明，在某些情况下，看似稳定的合作可能是暂时的稳定状态，而非真正的均衡点。这篇论文不仅提出了新颖的解析模型，还深入分析了学习过程中出现的动态不稳定现象，为理解和解释MARL中的复杂行为提供了有力支持。因此，此文整体上是一个非常有价值的学术成果，其理论意义和实际应用价值均值得高度赞赏。

5.Modelling and Control of Spatial Behaviours in Multi-Agent Systems with Applications to Biology and Robotics

Authors: Andrea Giusti

https://arxiv.org/abs/2501.00110

论文摘要

Large-Scale Multi-Agent Systems (LS-MAS) consist of several autonomous components, interacting in a non-trivial way, so that the emerging be haviour of the ensemble depends on the individual dynamics of the com ponents and their reciprocal interactions. These models can describe a rich variety of natural systems, as well as artificial ones, characterised by unpar alleled scalability, robustness, and flexibility. Indeed, a crucial objective is devising efficient strategies to model and control the spatial behaviours of LS-MAS to achieve specific goals. However, the inherent complexity of these systems and the wide spectrum of their emerging behaviours pose significant challenges. The overarching goal of this thesis is, therefore, to advance methods for modelling, analyzing and controlling the spatial be haviours of LS-MAS, with applications to cellular populations and swarm robotics. The thesis begins with an overview of the existing Literature, and is then organized into two distinct parts. In the context of swarm robotics, Part I deals with distributed control algorithms to spatially organize agents on geometric patterns. The contribution is twofold, encompassing both the development of original control algorithms, and providing a novel formal analysis, which allows to guarantee the emergence of specific geometric patterns. In Part II, looking at the spatial behaviours of biological agents, experiments are carried out to study the movement of microorganisms and their response to light stimuli. This allows the derivation and parametriza tion of mathematical models that capture these behaviours, and pave the way for the development of innovative approaches for the spatial control of microorganisms. The results presented in the thesis were developed by leveraging formal analytical tools, simulations, and experiments, using in novative platforms and original computational frameworks.

论文简评

这篇论文关注大规模多智能体系统（Large-Scale Multi-Agent Systems）的研究，特别是在生物学和机器人领域的应用。论文的主要贡献在于提出了一种新型的几何模式形成控制算法，并探讨了微生物的空间行为。论文内容分为两个部分，其中一部分通过形式分析、模拟和实验验证来展示理论成果。此外，论文还提供了详细的实验证据，以增强其结论的有效性。整篇论文结构清晰，逻辑严谨，充分展示了作者在该领域的重要研究工作，并为未来的研究提供了一定的参考价值。

我们欢迎您在评论区中留下宝贵的建议！包括但不限于：

• 可以提出推文中论文简评的不足！
• 可以分享最近更值得推荐的论文并给出理由！

推荐阅读