The increasing reliance on web interfaces presents many challenges for visually impaired users, showcasing the need for more advanced assistive technologies. This paper introduces WebNav, a voice-controlled web navigation agent that leverages a ReAct-inspired architecture and generative AI to provide this framework. WebNav comprises of a hierarchical structure: a Digital Navigation Module (DIGNAV) for high-level strategic planning, an Assistant Module for translating abstract commands into executable actions, and an Inference Module for low-level interaction. A key component is a dynamic labeling engine, implemented as a browser extension, that generates real-time labels for interactive elements, creating mapping between voice commands and Document Object Model (DOM) components. Preliminary evaluations show that WebNav outperforms traditional screen readers in response time and task completion accuracy for the visually impaired. Future work will focus on extensive user evaluations, benchmark development, and refining the agent's adaptive capabilities for real-world deployment.

Strategies for orchestrating the interactions between multiple agents, both human and artificial, can wildly overestimate performance and underestimate the cost of orchestration. We design a framework to orchestrate agents under realistic conditions, such as inference costs or availability constraints. We show theoretically that orchestration is only effective if there are performance or cost differentials between agents. We then empirically demonstrate how orchestration between multiple agents can be helpful for selecting agents in a simulated environment, picking a learning strategy in the infamous Rogers' Paradox from social science, and outsourcing tasks to other agents during a question-answer task in a user study.

Image restoration (IR) is challenging due to the complexity of real-world degradations. While many specialized and all-in-one IR models have been developed, they fail to effectively handle complex, mixed degradations. Recent agentic methods RestoreAgent and AgenticIR leverage intelligent, autonomous workflows to alleviate this issue, yet they suffer from suboptimal results and inefficiency due to their resource-intensive finetunings, and ineffective searches and tool execution trials for satisfactory outputs. In this paper, we propose MAIR, a novel Multi-Agent approach for complex IR problems. We introduce a real-world degradation prior, categorizing degradations into three types: (1) scene, (2) imaging, and (3) compression, which are observed to occur sequentially in real world, and reverse them in the opposite order. Built upon this three-stage restoration framework, MAIR emulates a team of collaborative human specialists, including a "scheduler" for overall planning and multiple "experts" dedicated to specific degradations. This design minimizes search space and trial efforts, improving image quality while reducing inference costs. In addition, a registry mechanism is introduced to enable easy integration of new tools. Experiments on both synthetic and real-world datasets show that proposed MAIR achieves competitive performance and improved efficiency over the previous agentic IR system. Code and models will be made available.

Large Language Models (LLMs) have made significant progress in various fields. However, challenges remain in Multi-Disciplinary Team (MDT) medical consultations. Current research enhances reasoning through role assignment, task decomposition, and accumulation of medical experience. Multi-role collaboration in MDT consultations often results in excessively long dialogue histories. This increases the model's cognitive burden and degrades both efficiency and accuracy. Some methods only store treatment histories. They do not extract effective experience or reflect on errors. This limits knowledge generalization and system evolution. We propose a multi-agent MDT medical consultation framework based on LLMs to address these issues. Our framework uses consensus aggregation and a residual discussion structure for multi-round consultations. It also employs a Correct Answer Knowledge Base (CorrectKB) and a Chain-of-Thought Knowledge Base (ChainKB) to accumulate consultation experience. These mechanisms enable the framework to evolve and continually improve diagnosis rationality and accuracy. Experimental results on the MedQA and PubMedQA datasets demonstrate that our framework achieves accuracies of 90.1% and 83.9%, respectively, and that the constructed knowledge bases generalize effectively across test sets from both datasets.

Large Language Models (LLMs) are probabilistic language generation models that do not incorporate explicit reasoning systems or logical planning modules. Consequently, in tasks that require synthesizing information over multiple steps, the reasoning performed at each stage is not clearly delineated, and intermediate reasoning occurs implicitly, making the process susceptible to errors. Furthermore, the difficulty of rigorously validating each step exacerbates the accumulation of errors throughout the overall process. To overcome these challenges, it is often necessary to retrieve external knowledge that compensates for the inherent limitations of LLMs, especially in real-world scenarios. Approaches such as Retrieval Augmented Generation (RAG) play a significant role by acquiring information not contained within the model in real time, thereby enabling more precise responses. Multi-step question answering (QA) is a representative challenge that demands both high precision in intermediate reasoning and the integration of diverse information. It not only exposes the limitations of LLMs but has also emerged as an important benchmark for real-world problems that seek to transcend these limitations. In this context, we propose Knowledge-Aware Iterative Retrieval for Multi-Agent Systems, a retrieval optimization system that employs an agent-based framework. It iteratively optimizes search queries through agent-guided knowledge accumulation, with a focus on query refinement, the iterative process of modifying or enhancing an initial query to improve search results.

We present PORTAL, a novel framework for developing artificial intelligence agents capable of playing thousands of 3D video games through language-guided policy generation. By transforming decision-making problems into language modeling tasks, our approach leverages large language models (LLMs) to generate behavior trees represented in domain-specific language (DSL). This method eliminates the computational burden associated with traditional reinforcement learning approaches while preserving strategic depth and rapid adaptability. Our framework introduces a hybrid policy structure that combines rule-based nodes with neural network components, enabling both high-level strategic reasoning and precise low-level control. A dual-feedback mechanism incorporating quantitative game metrics and vision-language model analysis facilitates iterative policy improvement at both tactical and strategic levels. The resulting policies are instantaneously deployable, human-interpretable, and capable of generalizing across diverse gaming environments. Experimental results demonstrate PORTAL's effectiveness across thousands of first-person shooter (FPS) games, showcasing significant improvements in development efficiency, policy generalization, and behavior diversity compared to traditional approaches. PORTAL represents a significant advancement in game AI development, offering a practical solution for creating sophisticated agents that can operate across thousands of commercial video games with minimal development overhead. Experiment results on the 3D video games are best viewed on https://zhongwen.one/projects/portal .

Reasoning and strategic behavior in social interactions is a hallmark of intelligence. This form of reasoning is significantly more sophisticated than isolated planning or reasoning tasks in static settings (e.g., math problem solving). In this paper, we present Strategic Planning, Interaction, and Negotiation (SPIN-Bench), a new multi-domain evaluation designed to measure the intelligence of strategic planning and social reasoning. While many existing benchmarks focus on narrow planning or single-agent reasoning, SPIN-Bench combines classical PDDL tasks, competitive board games, cooperative card games, and multi-agent negotiation scenarios in one unified framework. The framework includes both a benchmark as well as an arena to simulate and evaluate the variety of social settings to test reasoning and strategic behavior of AI agents. We formulate the benchmark SPIN-Bench by systematically varying action spaces, state complexity, and the number of interacting agents to simulate a variety of social settings where success depends on not only methodical and step-wise decision making, but also conceptual inference of other (adversarial or cooperative) participants. Our experiments reveal that while contemporary LLMs handle basic fact retrieval and short-range planning reasonably well, they encounter significant performance bottlenecks in tasks requiring deep multi-hop reasoning over large state spaces and socially adept coordination under uncertainty. We envision SPIN-Bench as a catalyst for future research on robust multi-agent planning, social reasoning, and human--AI teaming. Project Website: https://spinbench.github.io/

Hierarchical decision-making is a natural paradigm for coordinating multi-agent systems in complex environments such as air traffic management. In this paper, we present a bilevel framework for game-theoretic hierarchical routing, where a high-level router assigns discrete routes to multiple vehicles who seek to optimize potentially noncooperative objectives that depend upon the assigned routes. To address computational challenges, we propose a reformulation that preserves the convexity of each agent's feasible set. This convex reformulation enables a solution to be identified efficiently via a customized branch-and-bound algorithm. Our approach ensures global optimality while capturing strategic interactions between agents at the lower level. We demonstrate the solution concept of our framework in two-vehicle and three-vehicle routing scenarios.

Despite growing enthusiasm for Multi-Agent Systems (MAS), where multiple LLM agents collaborate to accomplish tasks, their performance gains across popular benchmarks remain minimal compared to single-agent frameworks. This gap highlights the need to analyze the challenges hindering MAS effectiveness. In this paper, we present the first comprehensive study of MAS challenges. We analyze five popular MAS frameworks across over 150 tasks, involving six expert human annotators. We identify 14 unique failure modes and propose a comprehensive taxonomy applicable to various MAS frameworks. This taxonomy emerges iteratively from agreements among three expert annotators per study, achieving a Cohen's Kappa score of 0.88. These fine-grained failure modes are organized into 3 categories, (i) specification and system design failures, (ii) inter-agent misalignment, and (iii) task verification and termination. To support scalable evaluation, we integrate MASFT with LLM-as-a-Judge. We also explore if identified failures could be easily prevented by proposing two interventions: improved specification of agent roles and enhanced orchestration strategies. Our findings reveal that identified failures require more complex solutions, highlighting a clear roadmap for future research. We open-source our dataset and LLM annotator.

In complex multi-agent environments, achieving efficient learning and desirable behaviours is a significant challenge for Multi-Agent Reinforcement Learning (MARL) systems. This work explores the potential of combining MARL with Large Language Model (LLM)-mediated interventions to guide agents toward more desirable behaviours. Specifically, we investigate how LLMs can be used to interpret and facilitate interventions that shape the learning trajectories of multiple agents. We experimented with two types of interventions, referred to as controllers: a Natural Language (NL) Controller and a Rule-Based (RB) Controller. The NL Controller, which uses an LLM to simulate human-like interventions, showed a stronger impact than the RB Controller. Our findings indicate that agents particularly benefit from early interventions, leading to more efficient training and higher performance. Both intervention types outperform the baseline without interventions, highlighting the potential of LLM-mediated guidance to accelerate training and enhance MARL performance in challenging environments.