Recent advancements in Web AI agents have demonstrated remarkable capabilities in addressing complex web navigation tasks. However, emerging research shows that these agents exhibit greater vulnerability compared to standalone Large Language Models (LLMs), despite both being built upon the same safety-aligned models. This discrepancy is particularly concerning given the greater flexibility of Web AI Agent compared to standalone LLMs, which may expose them to a wider range of adversarial user inputs. To build a scaffold that addresses these concerns, this study investigates the underlying factors that contribute to the increased vulnerability of Web AI agents. Notably, this disparity stems from the multifaceted differences between Web AI agents and standalone LLMs, as well as the complex signals - nuances that simple evaluation metrics, such as success rate, often fail to capture. To tackle these challenges, we propose a component-level analysis and a more granular, systematic evaluation framework. Through this fine-grained investigation, we identify three critical factors that amplify the vulnerability of Web AI agents; (1) embedding user goals into the system prompt, (2) multi-step action generation, and (3) observational capabilities. Our findings highlights the pressing need to enhance security and robustness in AI agent design and provide actionable insights for targeted defense strategies.
By utilizing more computational resources at test-time, large language models (LLMs) can improve without additional training. One common strategy uses verifiers to evaluate candidate outputs. In this work, we propose a novel scaling dimension for test-time compute: scaling the number of verifiers. We introduce Multi-Agent Verification (MAV) as a test-time compute paradigm that combines multiple verifiers to improve performance. We propose using Aspect Verifiers (AVs), off-the-shelf LLMs prompted to verify different aspects of outputs, as one possible choice for the verifiers in a MAV system. AVs are a convenient building block for MAV since they can be easily combined without additional training. Moreover, we introduce BoN-MAV, a simple multi-agent verification algorithm that combines best-of-n sampling with multiple verifiers. BoN-MAV demonstrates stronger scaling patterns than self-consistency and reward model verification, and we demonstrate both weak-to-strong generalization, where combining weak verifiers improves even stronger LLMs, and self-improvement, where the same base model is used to both generate and verify outputs. Our results establish scaling the number of verifiers as a promising new dimension for improving language model performance at test-time.
Agentic Generative AI, powered by Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG), Knowledge Graphs (KGs), and Vector Stores (VSs), represents a transformative technology applicable to specialized domains such as legal systems, research, recommender systems, cybersecurity, and global security, including proliferation research. This technology excels at inferring relationships within vast unstructured or semi-structured datasets. The legal domain here comprises complex data characterized by extensive, interrelated, and semi-structured knowledge systems with complex relations. It comprises constitutions, statutes, regulations, and case law. Extracting insights and navigating the intricate networks of legal documents and their relations is crucial for effective legal research. Here, we introduce a generative AI system that integrates RAG, VS, and KG, constructed via Non-Negative Matrix Factorization (NMF), to enhance legal information retrieval and AI reasoning and minimize hallucinations. In the legal system, these technologies empower AI agents to identify and analyze complex connections among cases, statutes, and legal precedents, uncovering hidden relationships and predicting legal trends-challenging tasks that are essential for ensuring justice and improving operational efficiency. Our system employs web scraping techniques to systematically collect legal texts, such as statutes, constitutional provisions, and case law, from publicly accessible platforms like Justia. It bridges the gap between traditional keyword-based searches and contextual understanding by leveraging advanced semantic representations, hierarchical relationships, and latent topic discovery. This framework supports legal document clustering, summarization, and cross-referencing, for scalable, interpretable, and accurate retrieval for semi-structured data while advancing computational law and AI.
Telephone surveys remain a valuable tool for gathering insights but typically require substantial resources in training and coordinating human interviewers. This work presents an AI-driven telephone survey system integrating text-to-speech (TTS), a large language model (LLM), and speech-to-text (STT) that mimics the versatility of human-led interviews on scale. We tested the system across two populations, a pilot study in the United States (n = 75) and a large-scale deployment in Peru (n = 2,739), inviting participants via web-based links and contacting them via direct phone calls. The AI agent successfully administered open-ended and closed-ended questions, handled basic clarifications, and dynamically navigated branching logic, allowing fast large-scale survey deployment without interviewer recruitment or training. Our findings demonstrate that while the AI system's probing for qualitative depth was more limited than human interviewers, overall data quality approached human-led standards for structured items. This study represents one of the first successful large-scale deployments of an LLM-based telephone interviewer in a real-world survey context. The AI-powered telephone survey system has the potential for expanding scalable, consistent data collecting across market research, social science, and public opinion studies, thus improving operational efficiency while maintaining appropriate data quality for research.
Large language models (LLMs) based agent systems have made great strides in real-world applications beyond traditional NLP tasks. This paper proposes a new LLM-powered Multi-Agent System (LLM-MAS) benchmark, Collab-Overcooked, built on the popular Overcooked-AI game with more applicable and challenging tasks in interactive environments. Collab-Overcooked extends existing benchmarks from two novel perspectives. First, it provides a multi-agent framework supporting diverse tasks and objectives and encourages collaboration through natural language communication. Second, it introduces a spectrum of process-oriented evaluation metrics to assess the fine-grained collaboration capabilities of different LLM agents, a dimension often overlooked in prior work. We conduct extensive experiments over 10 popular LLMs and show that, while the LLMs present a strong ability in goal interpretation, there is a significant discrepancy in active collaboration and continuous adaption that are critical for efficiently fulfilling complicated tasks. Notably, we highlight the strengths and weaknesses in LLM-MAS and provide insights for improving and evaluating LLM-MAS on a unified and open-sourced benchmark. Environments, 30 open-ended tasks, and an integrated evaluation package are now publicly available at https://github.com/YusaeMeow/Collab-Overcooked.
To improve Multimodal Large Language Models' (MLLMs) ability to process images and complex instructions, researchers predominantly curate large-scale visual instruction tuning datasets, which are either sourced from existing vision tasks or synthetically generated using LLMs and image descriptions. However, they often suffer from critical flaws, including misaligned instruction-image pairs and low-quality images. Such issues hinder training efficiency and limit performance improvements, as models waste resources on noisy or irrelevant data with minimal benefit to overall capability. To address this issue, we propose a \textbf{Vi}sual-Centric \textbf{S}election approach via \textbf{A}gents Collaboration (ViSA), which centers on image quality assessment and image-instruction relevance evaluation. Specifically, our approach consists of 1) an image information quantification method via visual agents collaboration to select images with rich visual information, and 2) a visual-centric instruction quality assessment method to select high-quality instruction data related to high-quality images. Finally, we reorganize 80K instruction data from large open-source datasets. Extensive experiments demonstrate that ViSA outperforms or is comparable to current state-of-the-art models on seven benchmarks, using only 2.5\% of the original data, highlighting the efficiency of our data selection approach. Moreover, we conduct ablation studies to validate the effectiveness of each component of our method. The code is available at https://github.com/HITsz-TMG/ViSA.
Mental health issues are worsening in today's competitive society, such as depression and anxiety. Traditional healings like counseling and chatbots fail to engage effectively, they often provide generic responses lacking emotional depth. Although large language models (LLMs) have the potential to create more human-like interactions, they still struggle to capture subtle emotions. This requires LLMs to be equipped with human-like adaptability and warmth. To fill this gap, we propose the MIND (Multi-agent INner Dialogue), a novel paradigm that provides more immersive psychological healing environments. Considering the strong generative and role-playing ability of LLM agents, we predefine an interactive healing framework and assign LLM agents different roles within the framework to engage in interactive inner dialogues with users, thereby providing an immersive healing experience. We conduct extensive human experiments in various real-world healing dimensions, and find that MIND provides a more user-friendly experience than traditional paradigms. This demonstrates that MIND effectively leverages the significant potential of LLMs in psychological healing.
The vast and complex materials design space demands innovative strategies to integrate multidisciplinary scientific knowledge and optimize materials discovery. While large language models (LLMs) have demonstrated promising reasoning and automation capabilities across various domains, their application in materials science remains limited due to a lack of benchmarking standards and practical implementation frameworks. To address these challenges, we introduce Mixture-of-Workflows for Self-Corrective Retrieval-Augmented Generation (CRAG-MoW) - a novel paradigm that orchestrates multiple agentic workflows employing distinct CRAG strategies using open-source LLMs. Unlike prior approaches, CRAG-MoW synthesizes diverse outputs through an orchestration agent, enabling direct evaluation of multiple LLMs across the same problem domain. We benchmark CRAG-MoWs across small molecules, polymers, and chemical reactions, as well as multi-modal nuclear magnetic resonance (NMR) spectral retrieval. Our results demonstrate that CRAG-MoWs achieve performance comparable to GPT-4o while being preferred more frequently in comparative evaluations, highlighting the advantage of structured retrieval and multi-agent synthesis. By revealing performance variations across data types, CRAG-MoW provides a scalable, interpretable, and benchmark-driven approach to optimizing AI architectures for materials discovery. These insights are pivotal in addressing fundamental gaps in benchmarking LLMs and autonomous AI agents for scientific applications.
Recent advances in Large Language Models (LLMs) have raised interest in their formal reasoning capabilities, particularly in mathematics. While closed LLMs like GPT-4 perform well on mathematical benchmarks, e.g., GSM8K, it remains unclear whether small to medium-sized open LLMs can achieve similar performance, questioning their reliability. To close this gap, we propose a post-training approach leveraging a mixture of opinions (MoO) from weaker ancillary LLMs to enhance a (relatively) stronger LLM's reasoning. For that, each post-training sample is augmented with Chain-of-Thought (CoT) reasoning steps and answers from ancillary LLMs, enabling the main LLM to learn from diverse perspectives. We compare MoO with standard supervised fine-tuning (SFT), few-shot prompting, and the Mixture of Agents (MoA) method on mathematical reasoning benchmarks. Our results show that incorporating weaker LLMs' opinions improves mathematical reasoning by an average of 5%, highlighting the value of diverse perspectives in reasoning tasks.
Multi-agent debate - multiple instances of large language models discussing problems in turn-based interaction - has shown promise for solving knowledge and reasoning tasks. However, these methods show limitations, particularly when scaling them to longer reasoning chains. In this study, we unveil a new issue of multi-agent debate: discussions drift away from the initial problem over multiple turns. We define this phenomenon as problem drift and quantify its presence across ten tasks (i.e., three generative, three knowledge, three reasoning, and one instruction-following task). To identify the reasons for this issue, we perform a human study with eight experts on discussions suffering from problem drift, who find the most common issues are a lack of progress (35% of cases), low-quality feedback (26% of cases), and a lack of clarity (25% of cases). To systematically address the issue of problem drift, we propose DRIFTJudge, a method based on LLM-as-a-judge, to detect problem drift at test-time. We further propose DRIFTPolicy, a method to mitigate 31% of problem drift cases. Our study can be seen as a first step to understanding a key limitation of multi-agent debate, highlighting pathways for improving their effectiveness in the future.
The deployment of Large Language Models (LLMs) in customer support is constrained by hallucination-generating false information-and the high cost of proprietary models. To address these challenges, we propose a retrieval-augmented question-answering (QA) pipeline and explore how to balance human input and automation. Using a dataset of questions about a Samsung Smart TV user manual, we demonstrate that synthetic data generated by LLMs outperforms crowdsourced data in reducing hallucination in finetuned models. We also compare self-training (fine-tuning models on their own outputs) and knowledge distillation (fine-tuning on stronger models' outputs, e.g., GPT-4o), and find that self-training achieves comparable hallucination reduction. We conjecture that this surprising finding can be attributed to increased exposure bias issues in the knowledge distillation case and support this conjecture with post hoc analysis. We also improve robustness to unanswerable questions and retrieval failures with contextualized "I don't know" responses. These findings show that scalable, cost-efficient QA systems can be built using synthetic data and self-training with open-source models, reducing reliance on proprietary tools or costly human annotations.
The language generation and reasoning capabilities of large language models (LLMs) have enabled conversational systems with impressive performance in a variety of tasks, from code generation, to composing essays, to passing STEM and legal exams, to a new paradigm for knowledge search. Besides those short-term use applications, LLMs are increasingly used to help with real-life goals or tasks that take a long time to complete, involving multiple sessions across days, weeks, months, or even years. Thus to enable conversational systems for long term interactions and tasks, we need language-based agents that can plan for long horizons. Traditionally, such capabilities were addressed by reinforcement learning agents with hierarchical planning capabilities. In this work, we explore a novel architecture where the LLM acts as the meta-controller deciding the agent's next macro-action, and tool use augmented LLM-based option policies execute the selected macro-action. We instantiate this framework for a specific set of macro-actions enabling adaptive planning for users' personal plans through conversation and follow-up questions collecting user feedback. We show how this paradigm can be applicable in scenarios ranging from tutoring for academic and non-academic tasks to conversational coaching for personal health plans.
Understanding domain-specific theorems often requires more than just text-based reasoning; effective communication through structured visual explanations is crucial for deeper comprehension. While large language models (LLMs) demonstrate strong performance in text-based theorem reasoning, their ability to generate coherent and pedagogically meaningful visual explanations remains an open challenge. In this work, we introduce TheoremExplainAgent, an agentic approach for generating long-form theorem explanation videos (over 5 minutes) using Manim animations. To systematically evaluate multimodal theorem explanations, we propose TheoremExplainBench, a benchmark covering 240 theorems across multiple STEM disciplines, along with 5 automated evaluation metrics. Our results reveal that agentic planning is essential for generating detailed long-form videos, and the o3-mini agent achieves a success rate of 93.8% and an overall score of 0.77. However, our quantitative and qualitative studies show that most of the videos produced exhibit minor issues with visual element layout. Furthermore, multimodal explanations expose deeper reasoning flaws that text-based explanations fail to reveal, highlighting the importance of multimodal explanations.
Reward models (RMs) are crucial for the training and inference-time scaling up of large language models (LLMs). However, existing reward models primarily focus on human preferences, neglecting verifiable correctness signals which have shown strong potential in training LLMs. In this paper, we propose agentic reward modeling, a reward system that combines reward models with verifiable correctness signals from different aspects to provide reliable rewards. We empirically implement a reward agent, named RewardAgent, that combines human preference rewards with two verifiable signals: factuality and instruction following, to provide more reliable rewards. We conduct comprehensive experiments on existing reward model benchmarks and inference time best-of-n searches on real-world downstream tasks. RewardAgent significantly outperforms vanilla reward models, demonstrating its effectiveness. We further construct training preference pairs using RewardAgent and train an LLM with the DPO objective, achieving superior performance on various NLP benchmarks compared to conventional reward models. Our codes are publicly released to facilitate further research (https://github.com/THU-KEG/Agentic-Reward-Modeling).
As large language models (LLMs) become more specialized, we envision a future where millions of expert LLMs exist, each trained on proprietary data and excelling in specific domains. In such a system, answering a query requires selecting a small subset of relevant models, querying them efficiently, and synthesizing their responses. This paper introduces a framework for agent-centric information access, where LLMs function as knowledge agents that are dynamically ranked and queried based on their demonstrated expertise. Unlike traditional document retrieval, this approach requires inferring expertise on the fly, rather than relying on static metadata or predefined model descriptions. This shift introduces several challenges, including efficient expert selection, cost-effective querying, response aggregation across multiple models, and robustness against adversarial manipulation. To address these issues, we propose a scalable evaluation framework that leverages retrieval-augmented generation and clustering techniques to construct and assess thousands of specialized models, with the potential to scale toward millions.
Social media platforms frequently impose restrictive policies to moderate user content, prompting the emergence of creative evasion language strategies. This paper presents a multi-agent framework based on Large Language Models (LLMs) to simulate the iterative evolution of language strategies under regulatory constraints. In this framework, participant agents, as social media users, continuously evolve their language expression, while supervisory agents emulate platform-level regulation by assessing policy violations. To achieve a more faithful simulation, we employ a dual design of language strategies (constraint and expression) to differentiate conflicting goals and utilize an LLM-driven GA (Genetic Algorithm) for the selection, mutation, and crossover of language strategies. The framework is evaluated using two distinct scenarios: an abstract password game and a realistic simulated illegal pet trade scenario. Experimental results demonstrate that as the number of dialogue rounds increases, both the number of uninterrupted dialogue turns and the accuracy of information transmission improve significantly. Furthermore, a user study with 40 participants validates the real-world relevance of the generated dialogues and strategies. Moreover, ablation studies validate the importance of the GA, emphasizing its contribution to long-term adaptability and improved overall results.
Differential Diagnosis (DDx) is a fundamental yet complex aspect of clinical decision-making, in which physicians iteratively refine a ranked list of possible diseases based on symptoms, antecedents, and medical knowledge. While recent advances in large language models have shown promise in supporting DDx, existing approaches face key limitations, including single-dataset evaluations, isolated optimization of components, unrealistic assumptions about complete patient profiles, and single-attempt diagnosis. We introduce a Modular Explainable DDx Agent (MEDDxAgent) framework designed for interactive DDx, where diagnostic reasoning evolves through iterative learning, rather than assuming a complete patient profile is accessible. MEDDxAgent integrates three modular components: (1) an orchestrator (DDxDriver), (2) a history taking simulator, and (3) two specialized agents for knowledge retrieval and diagnosis strategy. To ensure robust evaluation, we introduce a comprehensive DDx benchmark covering respiratory, skin, and rare diseases. We analyze single-turn diagnostic approaches and demonstrate the importance of iterative refinement when patient profiles are not available at the outset. Our broad evaluation demonstrates that MEDDxAgent achieves over 10% accuracy improvements in interactive DDx across both large and small LLMs, while offering critical explainability into its diagnostic reasoning process.
This paper presents a novel framework, called PLANTOR (PLanning with Natural language for Task-Oriented Robots), that integrates Large Language Models (LLMs) with Prolog-based knowledge management and planning for multi-robot tasks. The system employs a two-phase generation of a robot-oriented knowledge base, ensuring reusability and compositional reasoning, as well as a three-step planning procedure that handles temporal dependencies, resource constraints, and parallel task execution via mixed-integer linear programming. The final plan is converted into a Behaviour Tree for direct use in ROS2. We tested the framework in multi-robot assembly tasks within a block world and an arch-building scenario. Results demonstrate that LLMs can produce accurate knowledge bases with modest human feedback, while Prolog guarantees formal correctness and explainability. This approach underscores the potential of LLM integration for advanced robotics tasks requiring flexible, scalable, and human-understandable planning.
Understanding how opinions evolve is crucial for addressing issues such as polarization, radicalization, and consensus in social systems. While much research has focused on identifying factors influencing opinion change, the role of language and argumentative fallacies remains underexplored. This paper aims to fill this gap by investigating how language - along with social dynamics - influences opinion evolution through LODAS, a Language-Driven Opinion Dynamics Model for Agent-Based Simulations. The model simulates debates around the "Ship of Theseus" paradox, in which agents with discrete opinions interact with each other and evolve their opinions by accepting, rejecting, or ignoring the arguments presented. We study three different scenarios: balanced, polarized, and unbalanced opinion distributions. Agreeableness and sycophancy emerge as two main characteristics of LLM agents, and consensus around the presented statement emerges almost in any setting. Moreover, such AI agents are often producers of fallacious arguments in the attempt of persuading their peers and - for their complacency - they are also highly influenced by arguments built on logical fallacies. These results highlight the potential of this framework not only for simulating social dynamics but also for exploring from another perspective biases and shortcomings of LLMs, which may impact their interactions with humans.
Recent advancements in Large Language Models (LLMs) have substantially evolved Multi-Agent Systems (MASs) capabilities, enabling systems that not only automate tasks but also leverage near-human reasoning capabilities. To achieve this, LLM-based MASs need to be built around two critical principles: (i) a robust architecture that fully exploits LLM potential for specific tasks -- or related task sets -- and ($ii$) an effective methodology for equipping LLMs with the necessary capabilities to perform tasks and manage information efficiently. It goes without saying that a priori architectural designs can limit the scalability and domain adaptability of a given MAS. To address these challenges, in this paper we introduce Nexus: a lightweight Python framework designed to easily build and manage LLM-based MASs. Nexus introduces the following innovations: (i) a flexible multi-supervisor hierarchy, (ii) a simplified workflow design, and (iii) easy installation and open-source flexibility: Nexus can be installed via pip and is distributed under a permissive open-source license, allowing users to freely modify and extend its capabilities. Experimental results demonstrate that architectures built with Nexus exhibit state-of-the-art performance across diverse domains. In coding tasks, Nexus-driven MASs achieve a 99% pass rate on HumanEval and a flawless 100% on VerilogEval-Human, outperforming cutting-edge reasoning language models such as o3-mini and DeepSeek-R1. Moreover, these architectures display robust proficiency in complex reasoning and mathematical problem solving, achieving correct solutions for all randomly selected problems from the MATH dataset. In the realm of multi-objective optimization, Nexus-based architectures successfully address challenging timing closure tasks on designs from the VTR benchmark suite, while guaranteeing, on average, a power saving of nearly 30%.
Large Language Models (LLMs) have demonstrated remarkable capabilities in code generation from natural language prompts, revolutionizing software development workflows. As we advance towards agent-based development paradigms, these models form the cornerstone of next-generation software development lifecycles. However, current benchmarks for evaluating multilingual code generation capabilities are predominantly English-centric, limiting their applicability across the global developer community. To address this limitation, we present IndicEval-XL, a comprehensive benchmark for code generation that incorporates 6 major Indic languages, collectively spoken by approximately 14\% of the world's population. Our benchmark bridges these languages with 12 programming languages, creating a robust evaluation framework. This work is particularly significant given India's representation of one-eighth of the global population and the crucial role Indic languages play in Indian society. IndicEval-XL represents a significant step toward expanding the linguistic diversity in code generation systems and evaluation frameworks. By developing resources that support multiple languages, we aim to make AI-powered development tools more inclusive and accessible to developers of various linguistic backgrounds. To facilitate further research and development in this direction, we make our dataset and evaluation benchmark publicly available at https://github.com/telekom/IndicEval-XL
Young adults often encounter challenges in career exploration. Self-guided interventions, such as the letter-exchange exercise, where participants envision and adopt the perspective of their future selves by exchanging letters with their envisioned future selves, can support career development. However, the broader adoption of such interventions may be limited without structured guidance. To address this, we integrated Large Language Model (LLM)-based agents that simulate participants' future selves into the letter-exchange exercise and evaluated their effectiveness. A one-week experiment (N=36) compared three conditions: (1) participants manually writing replies to themselves from the perspective of their future selves (baseline), (2) future-self agents generating letters to participants, and (3) future-self agents engaging in chat conversations with participants. Results indicated that exchanging letters with future-self agents enhanced participants' engagement during the exercise, while overall benefits of the intervention on future orientation, career self-concept, and psychological support remained comparable across conditions. We discuss design implications for AI-augmented interventions for supporting young adults' career exploration.
Large language models (LLMs) often struggle with complex reasoning tasks due to their limitations in addressing the vast reasoning space and inherent ambiguities of natural language. We propose the Mixture-of-Search-Agents (MoSA) paradigm, a novel approach leveraging the collective expertise of multiple LLMs to enhance search-based reasoning. MoSA integrates diverse reasoning pathways by combining independent exploration with iterative refinement among LLMs, mitigating the limitations of single-model approaches. Using Monte Carlo Tree Search (MCTS) as a backbone, MoSA enables multiple agents to propose and aggregate reasoning steps, resulting in improved accuracy. Our comprehensive evaluation across four reasoning benchmarks demonstrates MoSA's consistent performance improvements over single-agent and other multi-agent baselines, particularly in complex mathematical and commonsense reasoning tasks.
This benchmark suite provides a comprehensive evaluation framework for assessing both individual LLMs and multi-agent systems in real-world planning scenarios. The suite encompasses eleven designed problems that progress from basic to highly complex, incorporating key aspects such as multi-agent coordination, inter-agent dependencies, and dynamic environmental disruptions. Each problem can be scaled along three dimensions: the number of parallel planning threads, the complexity of inter-dependencies, and the frequency of unexpected disruptions requiring real-time adaptation. The benchmark includes detailed specifications, evaluation metrics, and baseline implementations using contemporary frameworks like LangGraph, enabling rigorous testing of both single-agent and multi-agent planning capabilities. Through standardized evaluation criteria and scalable complexity, this benchmark aims to drive progress in developing more robust and adaptable AI planning systems for real-world applications.
In this project, our goal is to determine how to leverage the world-knowledge of pretrained large language models for efficient and robust learning in multiagent decision making. We examine this in a taxi routing and assignment problem where agents must decide how to best pick up passengers in order to minimize overall waiting time. While this problem is situated on a graphical road network, we show that with the proper prompting zero-shot performance is quite strong on this task. Furthermore, with limited fine-tuning along with the one-at-a-time rollout algorithm for look ahead, LLMs can out-compete existing approaches with 50 times fewer environmental interactions. We also explore the benefits of various linguistic prompting approaches and show that including certain easy-to-compute information in the prompt significantly improves performance. Finally, we highlight the LLM's built-in semantic understanding, showing its ability to adapt to environmental factors through simple prompts.
Reinforcement Learning from Human Feedback (RLHF) is essential for aligning large language models (LLMs) with human values. However, RLHF is susceptible to reward hacking, where the agent exploits flaws in the reward function rather than learning the intended behavior, thus degrading alignment. While reward shaping helps stabilize RLHF and partially mitigate reward hacking, a systematic investigation into shaping techniques and their underlying principles remains lacking. To bridge this gap, we present a comprehensive study of the prevalent reward shaping methods. Our analysis suggests three key design principles: (1) RL reward is ideally bounded, (2) RL benefits from rapid initial growth followed by gradual convergence, and (3) RL reward is best formulated as a function of centered reward. Guided by these insights, we propose Preference As Reward (PAR), a novel approach that leverages the latent preferences embedded within the reward model itself as the signal for reinforcement learning. We evaluated PAR on two base models, Gemma2-2B and Llama3-8B, using two datasets, Ultrafeedback-Binarized and HH-RLHF. Experimental results demonstrate PAR's superior performance over other reward shaping methods. On the AlpacaEval 2.0 benchmark, PAR achieves a win rate at least 5 percentage points higher than competing approaches. Furthermore, PAR exhibits remarkable data efficiency, requiring only a single reference reward for optimal performance, and maintains robustness against reward hacking even after two full epochs of training. Code is available at https://github.com/PorUna-byte/PAR.
The AgentSociety Challenge is the first competition in the Web Conference that aims to explore the potential of Large Language Model (LLM) agents in modeling user behavior and enhancing recommender systems on web platforms. The Challenge consists of two tracks: the User Modeling Track and the Recommendation Track. Participants are tasked to utilize a combined dataset from Yelp, Amazon, and Goodreads, along with an interactive environment simulator, to develop innovative LLM agents. The Challenge has attracted 295 teams across the globe and received over 1,400 submissions in total over the course of 37 official competition days. The participants have achieved 21.9% and 20.3% performance improvement for Track 1 and Track 2 in the Development Phase, and 9.1% and 15.9% in the Final Phase, representing a significant accomplishment. This paper discusses the detailed designs of the Challenge, analyzes the outcomes, and highlights the most successful LLM agent designs. To support further research and development, we have open-sourced the benchmark environment at https://tsinghua-fib-lab.github.io/AgentSocietyChallenge.
This work leverages Large Language Models (LLMs) to simulate human mobility, addressing challenges like high costs and privacy concerns in traditional models. Our hierarchical framework integrates persona generation, activity selection, and destination prediction, using real-world demographic and psychological data to create realistic movement patterns. Both physical models and language models are employed to explore and demonstrate different methodologies for human mobility simulation. By structuring data with summarization and weighted density metrics, the system ensures scalable memory management while retaining actionable insights. Preliminary results indicate that LLM-driven simulations align with observed real-world patterns, offering scalable, interpretable insights for social problems such as urban planning, traffic management, and public health. The framework's ability to dynamically generate personas and activities enables it to provide adaptable and realistic daily routines. This study demonstrates the transformative potential of LLMs in advancing mobility modeling for societal and urban applications. The source code and interactive demo for our framework are available at https://github.com/cju0/TrajLLM.
Zero-shot named entity recognition (NER) aims to develop entity recognition systems from unannotated text corpora. This task presents substantial challenges due to minimal human intervention. Recent work has adapted large language models (LLMs) for zero-shot NER by crafting specialized prompt templates. It advances model self-learning abilities by incorporating self-annotated demonstrations. However, two important challenges persist: (i) Correlations between contexts surrounding entities are overlooked, leading to wrong type predictions or entity omissions. (ii) The indiscriminate use of task demonstrations, retrieved through shallow similarity-based strategies, severely misleads LLMs during inference. In this paper, we introduce the cooperative multi-agent system (CMAS), a novel framework for zero-shot NER that uses the collective intelligence of multiple agents to address the challenges outlined above. CMAS has four main agents: (i) a self-annotator, (ii) a type-related feature (TRF) extractor, (iii) a demonstration discriminator, and (iv) an overall predictor. To explicitly capture correlations between contexts surrounding entities, CMAS reformulates NER into two subtasks: recognizing named entities and identifying entity type-related features within the target sentence. To enable controllable utilization of demonstrations, a demonstration discriminator is established to incorporate the self-reflection mechanism, automatically evaluating helpfulness scores for the target sentence. Experimental results show that CMAS significantly improves zero-shot NER performance across six benchmarks, including both domain-specific and general-domain scenarios. Furthermore, CMAS demonstrates its effectiveness in few-shot settings and with various LLM backbones.
In role-playing games (RPGs), the level of immersion is critical-especially when an in-game agent conveys tasks, hints, or ideas to the player. For an agent to accurately interpret the player's emotional state and contextual nuances, a foundational level of understanding is required, which can be achieved using a Large Language Model (LLM). Maintaining the LLM's focus across multiple context changes, however, necessitates a more robust approach, such as integrating the LLM with a dedicated task allocation model to guide its performance throughout gameplay. In response to this need, we introduce Voting-Based Task Assignment (VBTA), a framework inspired by human reasoning in task allocation and completion. VBTA assigns capability profiles to agents and task descriptions to tasks, then generates a suitability matrix that quantifies the alignment between an agent's abilities and a task's requirements. Leveraging six distinct voting methods, a pre-trained LLM, and integrating conflict-based search (CBS) for path planning, VBTA efficiently identifies and assigns the most suitable agent to each task. While existing approaches focus on generating individual aspects of gameplay, such as single quests, or combat encounters, our method shows promise when generating both unique combat encounters and narratives because of its generalizable nature.