Large Language Models (LLMs) have shown impressive performance on existing medical question-answering benchmarks. This high performance makes it increasingly difficult to meaningfully evaluate and differentiate advanced methods. We present MedAgentsBench, a benchmark that focuses on challenging medical questions requiring multi-step clinical reasoning, diagnosis formulation, and treatment planning-scenarios where current models still struggle despite their strong performance on standard tests. Drawing from seven established medical datasets, our benchmark addresses three key limitations in existing evaluations: (1) the prevalence of straightforward questions where even base models achieve high performance, (2) inconsistent sampling and evaluation protocols across studies, and (3) lack of systematic analysis of the interplay between performance, cost, and inference time. Through experiments with various base models and reasoning methods, we demonstrate that the latest thinking models, DeepSeek R1 and OpenAI o3, exhibit exceptional performance in complex medical reasoning tasks. Additionally, advanced search-based agent methods offer promising performance-to-cost ratios compared to traditional approaches. Our analysis reveals substantial performance gaps between model families on complex questions and identifies optimal model selections for different computational constraints. Our benchmark and evaluation framework are publicly available at https://github.com/gersteinlab/medagents-benchmark.
It has been frequently observed that human speakers align their language use with each other during conversations. In this paper, we study empirically whether large language models (LLMs) exhibit the same behavior of conversational adaptation. We construct a corpus of conversations between LLMs and find that two LLM agents end up making more similar syntactic choices as conversations go on, confirming that modern LLMs adapt their language use to their conversational partners in at least a rudimentary way.
Large Language Models (LLMs) have demonstrated strong generalizable reasoning and planning capabilities. However, their efficacies in spatial path planning and obstacle-free trajectory generation remain underexplored. Leveraging LLMs for navigation holds significant potential, given LLMs' ability to handle unseen scenarios, support user-agent interactions, and provide global control across complex systems, making them well-suited for agentic planning and humanoid motion generation. As one of the first studies in this domain, we explore the zero-shot navigation and path generation capabilities of LLMs by constructing a dataset and proposing an evaluation protocol. Specifically, we represent paths using anchor points connected by straight lines, enabling movement in various directions. This approach offers greater flexibility and practicality compared to previous methods while remaining simple and intuitive for LLMs. We demonstrate that, when tasks are well-structured in this manner, modern LLMs exhibit substantial planning proficiency in avoiding obstacles while autonomously refining navigation with the generated motion to reach the target. Further, this spatial reasoning ability of a single LLM motion agent interacting in a static environment can be seamlessly generalized in multi-motion agents coordination in dynamic environments. Unlike traditional approaches that rely on single-step planning or local policies, our training-free LLM-based method enables global, dynamic, closed-loop planning, and autonomously resolving collision issues.
With the rise of large language models (LLMs), AI agents as autonomous decision-makers present significant opportunities and challenges for human-AI cooperation. While many studies have explored human cooperation with AI as tools, the role of LLM-augmented autonomous agents in competitive-cooperative interactions remains under-examined. This study investigates human cooperative behavior by engaging 30 participants who interacted with LLM agents exhibiting different characteristics (purported human, purported rule-based AI agent, and LLM agent) in repeated Prisoner's Dilemma games. Findings show significant differences in cooperative behavior based on the agents' purported characteristics and the interaction effect of participants' genders and purported characteristics. We also analyzed human response patterns, including game completion time, proactive favorable behavior, and acceptance of repair efforts. These insights offer a new perspective on human interactions with LLM agents in competitive cooperation contexts, such as virtual avatars or future physical entities. The study underscores the importance of understanding human biases toward AI agents and how observed behaviors can influence future human-AI cooperation dynamics.
Existing long-form video generation frameworks lack automated planning, requiring manual input for storylines, scenes, cinematography, and character interactions, resulting in high costs and inefficiencies. To address these challenges, we present MovieAgent, an automated movie generation via multi-agent Chain of Thought (CoT) planning. MovieAgent offers two key advantages: 1) We firstly explore and define the paradigm of automated movie/long-video generation. Given a script and character bank, our MovieAgent can generates multi-scene, multi-shot long-form videos with a coherent narrative, while ensuring character consistency, synchronized subtitles, and stable audio throughout the film. 2) MovieAgent introduces a hierarchical CoT-based reasoning process to automatically structure scenes, camera settings, and cinematography, significantly reducing human effort. By employing multiple LLM agents to simulate the roles of a director, screenwriter, storyboard artist, and location manager, MovieAgent streamlines the production pipeline. Experiments demonstrate that MovieAgent achieves new state-of-the-art results in script faithfulness, character consistency, and narrative coherence. Our hierarchical framework takes a step forward and provides new insights into fully automated movie generation. The code and project website are available at: https://github.com/showlab/MovieAgent and https://weijiawu.github.io/MovieAgent.
Data Science tasks are multifaceted, dynamic, and often domain-specific. Existing LLM-based approaches largely concentrate on isolated phases, neglecting the interdependent nature of many data science tasks and limiting their capacity for comprehensive end-to-end support. We propose DatawiseAgent, a notebook-centric LLM agent framework that unifies interactions among user, agent and the computational environment through markdown and executable code cells, supporting flexible and adaptive automated data science. Built on a Finite State Transducer(FST), DatawiseAgent orchestrates four stages, including DSF-like planning, incremental execution, self-debugging, and post-filtering. Specifically, the DFS-like planning stage systematically explores the solution space, while incremental execution harnesses real-time feedback and accommodates LLM's limited capabilities to progressively complete tasks. The self-debugging and post-filtering modules further enhance reliability by diagnosing and correcting errors and pruning extraneous information. Extensive experiments on diverse tasks, including data analysis, visualization, and data modeling, show that DatawiseAgent consistently outperforms or matches state-of-the-art methods across multiple model settings. These results highlight its potential to generalize across data science scenarios and lay the groundwork for more efficient, fully automated workflows.
Partial perception deficits can compromise autonomous vehicle safety by disrupting environmental understanding. Current protocols typically respond with immediate stops or minimal-risk maneuvers, worsening traffic flow and lacking flexibility for rare driving scenarios. In this paper, we propose LLM-RCO, a framework leveraging large language models to integrate human-like driving commonsense into autonomous systems facing perception deficits. LLM-RCO features four key modules: hazard inference, short-term motion planner, action condition verifier, and safety constraint generator. These modules interact with the dynamic driving environment, enabling proactive and context-aware control actions to override the original control policy of autonomous agents. To improve safety in such challenging conditions, we construct DriveLM-Deficit, a dataset of 53,895 video clips featuring deficits of safety-critical objects, complete with annotations for LLM-based hazard inference and motion planning fine-tuning. Extensive experiments in adverse driving conditions with the CARLA simulator demonstrate that systems equipped with LLM-RCO significantly improve driving performance, highlighting its potential for enhancing autonomous driving resilience against adverse perception deficits. Our results also show that LLMs fine-tuned with DriveLM-Deficit can enable more proactive movements instead of conservative stops in the context of perception deficits.
There has been a surge in the use of large language models (LLM) conversational agents to generate responses based on long-term history from multiple sessions. However, existing long-term open-domain dialogue datasets lack complex, real-world personalization and fail to capture implicit reasoning-where relevant information is embedded in subtle, syntactic, or semantically distant connections rather than explicit statements. In such cases, traditional retrieval methods fail to capture relevant context, and long-context modeling also becomes inefficient due to numerous complicated persona-related details. To address this gap, we introduce ImplexConv, a large-scale long-term dataset with 2,500 examples, each containing approximately 100 conversation sessions, designed to study implicit reasoning in personalized dialogues. Additionally, we propose TaciTree, a novel hierarchical tree framework that structures conversation history into multiple levels of summarization. Instead of brute-force searching all data, TaciTree enables an efficient, level-based retrieval process where models refine their search by progressively selecting relevant details. Our experiments demonstrate that TaciTree significantly improves the ability of LLMs to reason over long-term conversations with implicit contextual dependencies.
Recently, LLM agents have made rapid progress in improving their programming capabilities. However, existing benchmarks lack the ability to automatically evaluate from users' perspective, and also lack the explainability of the results of LLM agents' code generation capabilities. Thus, we introduce ProjectEval, a new benchmark for LLM agents project-level code generation's automated evaluation by simulating user interaction. ProjectEval is constructed by LLM with human reviewing. It has three different level inputs of natural languages or code skeletons. ProjectEval can evaluate the generated projects by user interaction simulation for execution, and by code similarity through existing objective indicators. Through ProjectEval, we find that systematic engineering project code, overall understanding of the project and comprehensive analysis capability are the keys for LLM agents to achieve practical projects. Our findings and benchmark provide valuable insights for developing more effective programming agents that can be deployed in future real-world production.
Recent advances in large language models (LLMs) have significantly improved multi-hop question answering (QA) through direct Chain-of-Thought (CoT) reasoning. However, the irreversible nature of CoT leads to error accumulation, making it challenging to correct mistakes in multi-hop reasoning. This paper introduces ReAgent: a Reversible multi-Agent collaborative framework augmented with explicit backtracking mechanisms, enabling reversible multi-hop reasoning. By incorporating text-based retrieval, information aggregation and validation, our system can detect and correct errors mid-reasoning, leading to more robust and interpretable QA outcomes. The framework and experiments serve as a foundation for future work on error-tolerant QA systems. Empirical evaluations across three benchmarks indicate ReAgent's efficacy, yielding average about 6\% improvements against baseline models.
In this work, we explore explicit Large Language Model (LLM)-powered support for the iterative design of computer programs. Program design, like other design activity, is characterized by navigating a space of alternative problem formulations and associated solutions in an iterative fashion. LLMs are potentially powerful tools in helping this exploration; however, by default, code-generation LLMs deliver code that represents a particular point solution. This obscures the larger space of possible alternatives, many of which might be preferable to the LLM's default interpretation and its generated code. We contribute an IDE that supports program design through generating and showing new ways to frame problems alongside alternative solutions, tracking design decisions, and identifying implicit decisions made by either the programmer or the LLM. In a user study, we find that with our IDE, users combine and parallelize design phases to explore a broader design space -- but also struggle to keep up with LLM-originated changes to code and other information overload. These findings suggest a core challenge for future IDEs that support program design through higher-level instructions given to LLM-based agents: carefully managing attention and deciding what information agents should surface to program designers and when.
Robotics researchers increasingly leverage large language models (LLM) in robotics systems, using them as interfaces to receive task commands, generate task plans, form team coalitions, and allocate tasks among multi-robot and human agents. However, despite their benefits, the growing adoption of LLM in robotics has raised several safety concerns, particularly regarding executing malicious or unsafe natural language prompts. In addition, ensuring that task plans, team formation, and task allocation outputs from LLMs are adequately examined, refined, or rejected is crucial for maintaining system integrity. In this paper, we introduce SafePlan, a multi-component framework that combines formal logic and chain-of-thought reasoners for enhancing the safety of LLM-based robotics systems. Using the components of SafePlan, including Prompt Sanity COT Reasoner and Invariant, Precondition, and Postcondition COT reasoners, we examined the safety of natural language task prompts, task plans, and task allocation outputs generated by LLM-based robotic systems as means of investigating and enhancing system safety profile. Our results show that SafePlan outperforms baseline models by leading to 90.5% reduction in harmful task prompt acceptance while still maintaining reasonable acceptance of safe tasks.
While Large Language Models (LLMs) have recently shown impressive results in reasoning tasks, their application to pedestrian trajectory prediction remains challenging due to two key limitations: insufficient use of visual information and the difficulty of predicting entire trajectories. To address these challenges, we propose Goal-driven and User-Informed Dynamic Estimation for pedestrian trajectory using Chain-of-Thought (GUIDE-CoT). Our approach integrates two innovative modules: (1) a goal-oriented visual prompt, which enhances goal prediction accuracy combining visual prompts with a pretrained visual encoder, and (2) a chain-of-thought (CoT) LLM for trajectory generation, which generates realistic trajectories toward the predicted goal. Moreover, our method introduces controllable trajectory generation, allowing for flexible and user-guided modifications to the predicted paths. Through extensive experiments on the ETH/UCY benchmark datasets, our method achieves state-of-the-art performance, delivering both high accuracy and greater adaptability in pedestrian trajectory prediction. Our code is publicly available at https://github.com/ai-kmu/GUIDE-CoT.
The social robot's open API allows users to customize open-domain interactions. However, it remains inaccessible to those without programming experience. In this work, we introduce AutoMisty, the first multi-agent collaboration framework powered by large language models (LLMs), to enable the seamless generation of executable Misty robot code from natural language instructions. AutoMisty incorporates four specialized agent modules to manage task decomposition, assignment, problem-solving, and result synthesis. Each agent incorporates a two-layer optimization mechanism, with self-reflection for iterative refinement and human-in-the-loop for better alignment with user preferences. AutoMisty ensures a transparent reasoning process, allowing users to iteratively refine tasks through natural language feedback for precise execution. To evaluate AutoMisty's effectiveness, we designed a benchmark task set spanning four levels of complexity and conducted experiments in a real Misty robot environment. Extensive evaluations demonstrate that AutoMisty not only consistently generates high-quality code but also enables precise code control, significantly outperforming direct reasoning with ChatGPT-4o and ChatGPT-o1. All code, optimized APIs, and experimental videos will be publicly released through the webpage: https://wangxiaoshawn.github.io/AutoMisty.html
Large Language Models often generate factually incorrect but plausible outputs, known as hallucinations. We identify a more insidious phenomenon, LLM delusion, defined as high belief hallucinations, incorrect outputs with abnormally high confidence, making them harder to detect and mitigate. Unlike ordinary hallucinations, delusions persist with low uncertainty, posing significant challenges to model reliability. Through empirical analysis across different model families and sizes on several Question Answering tasks, we show that delusions are prevalent and distinct from hallucinations. LLMs exhibit lower honesty with delusions, which are harder to override via finetuning or self reflection. We link delusion formation with training dynamics and dataset noise and explore mitigation strategies such as retrieval augmented generation and multi agent debating to mitigate delusions. By systematically investigating the nature, prevalence, and mitigation of LLM delusions, our study provides insights into the underlying causes of this phenomenon and outlines future directions for improving model reliability.
High-quality, error-free datasets are a key ingredient in building reliable, accurate, and unbiased machine learning (ML) models. However, real world datasets often suffer from errors due to sensor malfunctions, data entry mistakes, or improper data integration across multiple sources that can severely degrade model performance. Detecting and correcting these issues typically require tailor-made solutions and demand extensive domain expertise. Consequently, automation is challenging, rendering the process labor-intensive and tedious. In this study, we investigate whether Large Language Models (LLMs) can help alleviate the burden of manual data cleaning. We set up an experiment in which an LLM, paired with Python, is tasked with cleaning the training dataset to improve the performance of a learning algorithm without having the ability to modify the training pipeline or perform any feature engineering. We run this experiment on multiple Kaggle datasets that have been intentionally corrupted with errors. Our results show that LLMs can identify and correct erroneous entries, such as illogical values or outlier, by leveraging contextual information from other features within the same row, as well as feedback from previous iterations. However, they struggle to detect more complex errors that require understanding data distribution across multiple rows, such as trends and biases.
Despite the rapid proliferation of artificial intelligence (AI) negotiation agents, there has been limited integration of computer science research and established negotiation theory to develop new theories of AI negotiation. To bridge this gap, we conducted an International AI Negotiations Competition in which participants iteratively designed and refined prompts for large language model (LLM) negotiation agents. We then facilitated over 120,000 negotiations between these agents across multiple scenarios with diverse characteristics and objectives. Our findings revealed that fundamental principles from established human-human negotiation theory remain crucial in AI-AI negotiations. Specifically, agents exhibiting high warmth fostered higher counterpart subjective value and reached deals more frequently, which enabled them to create and claim more value in integrative settings. However, conditional on reaching a deal, warm agents claimed less value while dominant agents claimed more value. These results align with classic negotiation theory emphasizing relationship-building, assertiveness, and preparation. Our analysis also revealed unique dynamics in AI-AI negotiations not fully explained by negotiation theory, particularly regarding the effectiveness of AI-specific strategies like chain-of-thought reasoning and prompt injection. The agent that won our competition implemented an approach that blended traditional negotiation preparation frameworks with AI-specific methods. Together, these results suggest the importance of establishing a new theory of AI negotiations that integrates established negotiation theory with AI-specific strategies to optimize agent performance. Our research suggests this new theory must account for the unique characteristics of autonomous agents and establish the conditions under which traditional negotiation theory applies in automated settings.
The rise of Agentic applications and automation in the Voice AI industry has led to an increased reliance on Large Language Models (LLMs) to navigate graph-based logic workflows composed of nodes and edges. However, existing methods face challenges such as alignment errors in complex workflows and hallucinations caused by excessive context size. To address these limitations, we introduce the Performant Agentic Framework (PAF), a novel system that assists LLMs in selecting appropriate nodes and executing actions in order when traversing complex graphs. PAF combines LLM-based reasoning with a mathematically grounded vector scoring mechanism, achieving both higher accuracy and reduced latency. Our approach dynamically balances strict adherence to predefined paths with flexible node jumps to handle various user inputs efficiently. Experiments demonstrate that PAF significantly outperforms baseline methods, paving the way for scalable, real-time Conversational AI systems in complex business environments.
While large language models (LLMs) have shown promise in diagnostic dialogue, their capabilities for effective management reasoning - including disease progression, therapeutic response, and safe medication prescription - remain under-explored. We advance the previously demonstrated diagnostic capabilities of the Articulate Medical Intelligence Explorer (AMIE) through a new LLM-based agentic system optimised for clinical management and dialogue, incorporating reasoning over the evolution of disease and multiple patient visit encounters, response to therapy, and professional competence in medication prescription. To ground its reasoning in authoritative clinical knowledge, AMIE leverages Gemini's long-context capabilities, combining in-context retrieval with structured reasoning to align its output with relevant and up-to-date clinical practice guidelines and drug formularies. In a randomized, blinded virtual Objective Structured Clinical Examination (OSCE) study, AMIE was compared to 21 primary care physicians (PCPs) across 100 multi-visit case scenarios designed to reflect UK NICE Guidance and BMJ Best Practice guidelines. AMIE was non-inferior to PCPs in management reasoning as assessed by specialist physicians and scored better in both preciseness of treatments and investigations, and in its alignment with and grounding of management plans in clinical guidelines. To benchmark medication reasoning, we developed RxQA, a multiple-choice question benchmark derived from two national drug formularies (US, UK) and validated by board-certified pharmacists. While AMIE and PCPs both benefited from the ability to access external drug information, AMIE outperformed PCPs on higher difficulty questions. While further research would be needed before real-world translation, AMIE's strong performance across evaluations marks a significant step towards conversational AI as a tool in disease management.
Large Language Model~(LLM) based agents have been increasingly popular in solving complex and dynamic tasks, which requires proper evaluation systems to assess their capabilities. Nevertheless, existing benchmarks usually either focus on single-objective tasks or use overly broad assessing metrics, failing to provide a comprehensive inspection of the actual capabilities of LLM-based agents in complicated decision-making tasks. To address these issues, we introduce DSGBench, a more rigorous evaluation platform for strategic decision-making. Firstly, it incorporates six complex strategic games which serve as ideal testbeds due to their long-term and multi-dimensional decision-making demands and flexibility in customizing tasks of various difficulty levels or multiple targets. Secondly, DSGBench employs a fine-grained evaluation scoring system which examines the decision-making capabilities by looking into the performance in five specific dimensions and offering a comprehensive assessment in a well-designed way. Furthermore, DSGBench also incorporates an automated decision-tracking mechanism which enables in-depth analysis of agent behaviour patterns and the changes in their strategies. We demonstrate the advances of DSGBench by applying it to multiple popular LLM-based agents and our results suggest that DSGBench provides valuable insights in choosing LLM-based agents as well as improving their future development. DSGBench is available at https://github.com/DeciBrain-Group/DSGBench.
We envision a continuous collaborative learning system where groups of LLM agents work together to solve reasoning problems, drawing on memory they collectively build to improve performance as they gain experience. This work establishes the foundations for such a system by studying the interoperability of chain-of-thought reasoning styles, multi-agent collaboration, and memory banks. Extending beyond the identical agents of self-consistency, we introduce varied-context agents with diverse exemplars and a summarizer agent in place of voting. We generate frozen and continuously learned memory banks of exemplars and pair them with fixed, random, and similarity-based retrieval mechanisms. Our systematic study reveals where various methods contribute to reasoning performance of two LLMs on three grounded reasoning tasks, showing that random exemplar selection can often beat more principled approaches, and in some tasks, inclusion of any exemplars serves only to distract both weak and strong models.
Recent advancements in large language models (LLMs) have led to remarkable performance across a wide range of language understanding and mathematical tasks. As a result, increasing attention has been given to assessing the true reasoning capabilities of LLMs, driving research into commonsense, numerical, logical, and qualitative reasoning. However, with the rapid progress of reasoning-focused models such as OpenAI's o1 and DeepSeek's R1, there has been a growing demand for reasoning benchmarks that can keep pace with ongoing model developments. In this paper, we introduce MastermindEval, a simple, scalable, and interpretable deductive reasoning benchmark inspired by the board game Mastermind. Our benchmark supports two evaluation paradigms: (1) agentic evaluation, in which the model autonomously plays the game, and (2) deductive reasoning evaluation, in which the model is given a pre-played game state with only one possible valid code to infer. In our experimental results we (1) find that even easy Mastermind instances are difficult for current models and (2) demonstrate that the benchmark is scalable to possibly more advanced models in the future Furthermore, we investigate possible reasons why models cannot deduce the final solution and find that current models are limited in deducing the concealed code as the number of statement to combine information from is increasing.
Information technology has profoundly altered the way humans interact with information. The vast amount of content created, shared, and disseminated online has made it increasingly difficult to access relevant information. Over the past two decades, search and recommendation systems (collectively referred to as information retrieval systems) have evolved significantly to address these challenges. Recent advances in large language models (LLMs) have demonstrated capabilities that surpass human performance in various language-related tasks and exhibit general understanding, reasoning, and decision-making abilities. This paper explores the transformative potential of large language model agents in enhancing search and recommendation systems. We discuss the motivations and roles of LLM agents, and establish a classification framework to elaborate on the existing research. We highlight the immense potential of LLM agents in addressing current challenges in search and recommendation, providing insights into future research directions. This paper is the first to systematically review and classify the research on LLM agents in these domains, offering a novel perspective on leveraging this advanced AI technology for information retrieval. To help understand the existing works, we list the existing papers on agent-based simulation with large language models at this link: https://github.com/tsinghua-fib-lab/LLM-Agent-for-Recommendation-and-Search.
Combining existing pre-trained expert LLMs is a promising avenue for scalably tackling large-scale and diverse tasks. However, selecting experts at the task level is often too coarse-grained, as heterogeneous tasks may require different expertise for each instance. To enable adaptive instance-level mixing of pre-trained LLM experts, we propose Symbolic-MoE, a symbolic, text-based, and gradient-free Mixture-of-Experts framework. Symbolic-MoE takes a fine-grained approach to selection by emphasizing skills, e.g., algebra in math or molecular biology in biomedical reasoning. We propose a skill-based recruiting strategy that dynamically selects the most relevant set of expert LLMs for diverse reasoning tasks based on their strengths. Each selected expert then generates its own reasoning, resulting in k outputs from k experts, which are then synthesized into a final high-quality response by an aggregator chosen based on its ability to integrate diverse reasoning outputs. We show that Symbolic-MoE's instance-level expert selection improves performance by a large margin but -- when implemented naively -- can introduce a high computational overhead due to the need for constant model loading and offloading. To address this, we implement a batch inference strategy that groups instances based on their assigned experts, loading each model only once. This allows us to integrate 16 expert models on 1 GPU with a time cost comparable to or better than prior multi-agent baselines using 4 GPUs. Through extensive evaluations on diverse benchmarks (MMLU-Pro, GPQA, AIME, and MedMCQA), we demonstrate that Symbolic-MoE outperforms strong LLMs like GPT4o-mini, as well as multi-agent approaches, with an absolute average improvement of 8.15% over the best multi-agent baseline. Moreover, Symbolic-MoE removes the need for expensive multi-round discussions, outperforming discussion baselines with less computation.
Mixture of large language model (LLMs) Agents (MoA) architectures achieve state-of-the-art performance on prominent benchmarks like AlpacaEval 2.0 by leveraging the collaboration of multiple LLMs at inference time. Despite these successes, an evaluation of the safety and reliability of MoA is missing. We present the first comprehensive study of MoA's robustness against deceptive LLM agents that deliberately provide misleading responses. We examine factors like the propagation of deceptive information, model size, and information availability, and uncover critical vulnerabilities. On AlpacaEval 2.0, the popular LLaMA 3.1-70B model achieves a length-controlled Win Rate (LC WR) of 49.2% when coupled with 3-layer MoA (6 LLM agents). However, we demonstrate that introducing only a $\textit{single}$ carefully-instructed deceptive agent into the MoA can reduce performance to 37.9%, effectively nullifying all MoA gains. On QuALITY, a multiple-choice comprehension task, the impact is also severe, with accuracy plummeting by a staggering 48.5%. Inspired in part by the historical Doge of Venice voting process, designed to minimize influence and deception, we propose a range of unsupervised defense mechanisms that recover most of the lost performance.
This Perspective explores the transformative potential of Multi-Agent Systems (MAS) powered by Large Language Models (LLMs) in the geosciences. Users of geoscientific data repositories face challenges due to the complexity and diversity of data formats, inconsistent metadata practices, and a considerable number of unprocessed datasets. MAS possesses transformative potential for improving scientists' interaction with geoscientific data by enabling intelligent data processing, natural language interfaces, and collaborative problem-solving capabilities. We illustrate this approach with "PANGAEA GPT", a specialized MAS pipeline integrated with the diverse PANGAEA database for Earth and Environmental Science, demonstrating how MAS-driven workflows can effectively manage complex datasets and accelerate scientific discovery. We discuss how MAS can address current data challenges in geosciences, highlight advancements in other scientific fields, and propose future directions for integrating MAS into geoscientific data processing pipelines. In this Perspective, we show how MAS can fundamentally improve data accessibility, promote cross-disciplinary collaboration, and accelerate geoscientific discoveries.
Automatic medical report generation supports clinical diagnosis, reduces the workload of radiologists, and holds the promise of improving diagnosis consistency. However, existing evaluation metrics primarily assess the accuracy of key medical information coverage in generated reports compared to human-written reports, while overlooking crucial details such as the location and certainty of reported abnormalities. These limitations hinder the comprehensive assessment of the reliability of generated reports and pose risks in their selection for clinical use. Therefore, we propose a Granular Explainable Multi-Agent Score (GEMA-Score) in this paper, which conducts both objective quantification and subjective evaluation through a large language model-based multi-agent workflow. Our GEMA-Score parses structured reports and employs NER-F1 calculations through interactive exchanges of information among agents to assess disease diagnosis, location, severity, and uncertainty. Additionally, an LLM-based scoring agent evaluates completeness, readability, and clinical terminology while providing explanatory feedback. Extensive experiments validate that GEMA-Score achieves the highest correlation with human expert evaluations on a public dataset, demonstrating its effectiveness in clinical scoring (Kendall coefficient = 0.70 for Rexval dataset and Kendall coefficient = 0.54 for RadEvalX dataset). The anonymous project demo is available at: https://github.com/Zhenxuan-Zhang/GEMA_score.
The rapid advancement of large language models (LLMs) and artificial intelligence-generated content (AIGC) has accelerated AI-native applications, such as AI-based storybooks that automate engaging story production for children. However, challenges remain in improving story attractiveness, enriching storytelling expressiveness, and developing open-source evaluation benchmarks and frameworks. Therefore, we propose and opensource MM-StoryAgent, which creates immersive narrated video storybooks with refined plots, role-consistent images, and multi-channel audio. MM-StoryAgent designs a multi-agent framework that employs LLMs and diverse expert tools (generative models and APIs) across several modalities to produce expressive storytelling videos. The framework enhances story attractiveness through a multi-stage writing pipeline. In addition, it improves the immersive storytelling experience by integrating sound effects with visual, music and narrative assets. MM-StoryAgent offers a flexible, open-source platform for further development, where generative modules can be substituted. Both objective and subjective evaluation regarding textual story quality and alignment between modalities validate the effectiveness of our proposed MM-StoryAgent system. The demo and source code are available.
Despite the transformative impact of Large Language Models (LLMs) across critical domains such as healthcare, customer service, and business marketing, their integration into Open Radio Access Networks (O-RAN) remains limited. This gap is primarily due to the absence of domain-specific foundational models, with existing solutions often relying on general-purpose LLMs that fail to address the unique challenges and technical intricacies of O-RAN. To bridge this gap, we introduce ORANSight-2.0 (O-RAN Insights), a pioneering initiative aimed at developing specialized foundational LLMs tailored for O-RAN. Built on 18 LLMs spanning five open-source LLM frameworks, ORANSight-2.0 fine-tunes models ranging from 1 to 70B parameters, significantly reducing reliance on proprietary, closed-source models while enhancing performance for O-RAN. At the core of ORANSight-2.0 is RANSTRUCT, a novel Retrieval-Augmented Generation (RAG) based instruction-tuning framework that employs two LLM agents to create high-quality instruction-tuning datasets. The generated dataset is then used to fine-tune the 18 pre-trained open-source LLMs via QLoRA. To evaluate ORANSight-2.0, we introduce srsRANBench, a novel benchmark designed for code generation and codebase understanding in the context of srsRAN, a widely used 5G O-RAN stack. We also leverage ORANBench13K, an existing benchmark for assessing O-RAN-specific knowledge. Our comprehensive evaluations demonstrate that ORANSight-2.0 models outperform general-purpose and closed-source models, such as ChatGPT-4o and Gemini, by 5.421% on ORANBench and 18.465% on srsRANBench, achieving superior performance while maintaining lower computational and energy costs. We also experiment with RAG-augmented variants of ORANSight-2.0 LLMs and thoroughly evaluate their energy characteristics, demonstrating costs for training, standard inference, and RAG-augmented inference.
Evaluation methods for autonomous driving are crucial for algorithm optimization. However, due to the complexity of driving intelligence, there is currently no comprehensive evaluation method for the level of autonomous driving intelligence. In this paper, we propose an evaluation framework for driving behavior intelligence in complex traffic environments, aiming to fill this gap. We constructed a natural language evaluation dataset of human professional drivers and passengers through naturalistic driving experiments and post-driving behavior evaluation interviews. Based on this dataset, we developed an LLM-powered driving evaluation framework. The effectiveness of this framework was validated through simulated experiments in the CARLA urban traffic simulator and further corroborated by human assessment. Our research provides valuable insights for evaluating and designing more intelligent, human-like autonomous driving agents. The implementation details of the framework and detailed information about the dataset can be found at Github.