Xinrun Wang

Longtao Zheng, Rundong Wang, Xinrun Wang et al.

Building agents with large language models (LLMs) for computer control is a burgeoning research area, where the agent receives computer states and performs actions to complete complex tasks. Previous computer agents have demonstrated the benefits of in-context learning (ICL); however, their performance is hindered by several issues. First, the limited context length of LLMs and complex computer states restrict the number of exemplars, as a single webpage can consume the entire context. Second, the exemplars in current methods, such as high-level plans and multi-choice questions, cannot represent complete trajectories, leading to suboptimal performance in long-horizon tasks. Third, existing computer agents rely on task-specific exemplars and overlook the similarity among tasks, resulting in poor generalization to novel tasks. To address these challenges, we introduce Synapse, a computer agent featuring three key components: i) state abstraction, which filters out task-irrelevant information from raw states, allowing more exemplars within the limited context, ii) trajectory-as-exemplar prompting, which prompts the LLM with complete trajectories of the abstracted states and actions to improve multi-step decision-making, and iii) exemplar memory, which stores the embeddings of exemplars and retrieves them via similarity search for generalization to novel tasks. We evaluate Synapse on MiniWoB++, a standard task suite, and Mind2Web, a real-world website benchmark. In MiniWoB++, Synapse achieves a 99.2% average success rate (a 10% relative improvement) across 64 tasks using demonstrations from only 48 tasks. Notably, Synapse is the first ICL method to solve the book-flight task in MiniWoB++. Synapse also exhibits a 56% relative improvement in average step success rate over the previous state-of-the-art prompting scheme in Mind2Web.

Chang Yang, Chuang Zhou, Yilin Xiao et al.

Memory emerges as the core module in the Large Language Model (LLM)-based agents for long-horizon complex tasks (e.g., multi-turn dialogue, game playing, scientific discovery), where memory can enable knowledge accumulation, iterative reasoning and self-evolution. Among diverse paradigms, graph stands out as a powerful structure for agent memory due to the intrinsic capabilities to model relational dependencies, organize hierarchical information, and support efficient retrieval. This survey presents a comprehensive review of agent memory from the graph-based perspective. First, we introduce a taxonomy of agent memory, including short-term vs. long-term memory, knowledge vs. experience memory, non-structural vs. structural memory, with an implementation view of graph-based memory. Second, according to the life cycle of agent memory, we systematically analyze the key techniques in graph-based agent memory, covering memory extraction for transforming the data into the contents, storage for organizing the data efficiently, retrieval for retrieving the relevant contents from memory to support reasoning, and evolution for updating the contents in the memory. Third, we summarize the open-sourced libraries and benchmarks that support the development and evaluation of self-evolving agent memory. We also explore diverse application scenarios. Finally, we identify critical challenges and future research directions. This survey aims to offer actionable insights to advance the development of more efficient and reliable graph-based agent memory systems. All the related resources, including research papers, open-source data, and projects, are collected for the community in https://github.com/DEEP-PolyU/Awesome-GraphMemory.

Haochong Xia, Shuo Sun, Xinrun Wang et al.

Financial simulators play an important role in enhancing forecasting accuracy, managing risks, and fostering strategic financial decision-making. Despite the development of financial market simulation methodologies, existing frameworks often struggle with adapting to specialized simulation context. We pinpoint the challenges as i) current financial datasets do not contain context labels; ii) current techniques are not designed to generate financial data with context as control, which demands greater precision compared to other modalities; iii) the inherent difficulties in generating context-aligned, high-fidelity data given the non-stationary, noisy nature of financial data. To address these challenges, our contributions are: i) we proposed the Contextual Market Dataset with market dynamics, stock ticker, and history state as context, leveraging a market dynamics modeling method that combines linear regression and Dynamic Time Warping clustering to extract market dynamics; ii) we present Market-GAN, a novel architecture incorporating a Generative Adversarial Networks (GAN) for the controllable generation with context, an autoencoder for learning low-dimension features, and supervisors for knowledge transfer; iii) we introduce a two-stage training scheme to ensure that Market-GAN captures the intrinsic market distribution with multiple objectives. In the pertaining stage, with the use of the autoencoder and supervisors, we prepare the generator with a better initialization for the adversarial training stage. We propose a set of holistic evaluation metrics that consider alignment, fidelity, data usability on downstream tasks, and market facts. We evaluate Market-GAN with the Dow Jones Industrial Average data from 2000 to 2023 and showcase superior performance in comparison to 4 state-of-the-art time-series generative models.

Shuo Sun, Molei Qin, Xinrun Wang et al.

The financial markets, which involve more than $90 trillion market capitals, attract the attention of innumerable investors around the world. Recently, reinforcement learning in financial markets (FinRL) has emerged as a promising direction to train agents for making profitable investment decisions. However, the evaluation of most FinRL methods only focuses on profit-related measures and ignores many critical axes, which are far from satisfactory for financial practitioners to deploy these methods into real-world financial markets. Therefore, we introduce PRUDEX-Compass, which has 6 axes, i.e., Profitability, Risk-control, Universality, Diversity, rEliability, and eXplainability, with a total of 17 measures for a systematic evaluation. Specifically, i) we propose AlphaMix+ as a strong FinRL baseline, which leverages mixture-of-experts (MoE) and risk-sensitive approaches to make diversified risk-aware investment decisions, ii) we evaluate 8 FinRL methods in 4 long-term real-world datasets of influential financial markets to demonstrate the usage of our PRUDEX-Compass, iii) PRUDEX-Compass together with 4 real-world datasets, standard implementation of 8 FinRL methods and a portfolio management environment is released as public resources to facilitate the design and comparison of new FinRL methods. We hope that PRUDEX-Compass can not only shed light on future FinRL research to prevent untrustworthy results from stagnating FinRL into successful industry deployment but also provide a new challenging algorithm evaluation scenario for the reinforcement learning (RL) community.

Wentao Zhang, Yilei Zhao, Shuo Sun et al.

Portfolio management (PM) is a fundamental financial trading task, which explores the optimal periodical reallocation of capitals into different stocks to pursue long-term profits. Reinforcement learning (RL) has recently shown its potential to train profitable agents for PM through interacting with financial markets. However, existing work mostly focuses on fixed stock pools, which is inconsistent with investors' practical demand. Specifically, the target stock pool of different investors varies dramatically due to their discrepancy on market states and individual investors may temporally adjust stocks they desire to trade (e.g., adding one popular stocks), which lead to customizable stock pools (CSPs). Existing RL methods require to retrain RL agents even with a tiny change of the stock pool, which leads to high computational cost and unstable performance. To tackle this challenge, we propose EarnMore, a rEinforcement leARNing framework with Maskable stOck REpresentation to handle PM with CSPs through one-shot training in a global stock pool (GSP). Specifically, we first introduce a mechanism to mask out the representation of the stocks outside the target pool. Second, we learn meaningful stock representations through a self-supervised masking and reconstruction process. Third, a re-weighting mechanism is designed to make the portfolio concentrate on favorable stocks and neglect the stocks outside the target pool. Through extensive experiments on 8 subset stock pools of the US stock market, we demonstrate that EarnMore significantly outperforms 14 state-of-the-art baselines in terms of 6 popular financial metrics with over 40% improvement on profit.

Pengdeng Li, Xinrun Wang, Shuxin Li et al.

In this work, we attempt to bridge the two fields of finite-agent and infinite-agent games, by studying how the optimal policies of agents evolve with the number of agents (population size) in mean-field games, an agent-centric perspective in contrast to the existing works focusing typically on the convergence of the empirical distribution of the population. To this end, the premise is to obtain the optimal policies of a set of finite-agent games with different population sizes. However, either deriving the closed-form solution for each game is theoretically intractable, training a distinct policy for each game is computationally intensive, or directly applying the policy trained in a game to other games is sub-optimal. We address these challenges through the Population-size-Aware Policy Optimization (PAPO). Our contributions are three-fold. First, to efficiently generate efficient policies for games with different population sizes, we propose PAPO, which unifies two natural options (augmentation and hypernetwork) and achieves significantly better performance. PAPO consists of three components: i) the population-size encoding which transforms the original value of population size to an equivalent encoding to avoid training collapse, ii) a hypernetwork to generate a distinct policy for each game conditioned on the population size, and iii) the population size as an additional input to the generated policy. Next, we construct a multi-task-based training procedure to efficiently train the neural networks of PAPO by sampling data from multiple games with different population sizes. Finally, extensive experiments on multiple environments show the significant superiority of PAPO over baselines, and the analysis of the evolution of the generated policies further deepens our understanding of the two fields of finite-agent and infinite-agent games.

Shuxin Li, Xinrun Wang, Youzhi Zhang et al.

Offline reinforcement learning (offline RL) is an emerging field that has recently begun gaining attention across various application domains due to its ability to learn strategies from earlier collected datasets. Offline RL proved very successful, paving a path to solving previously intractable real-world problems, and we aim to generalize this paradigm to a multiplayer-game setting. To this end, we introduce a problem of offline equilibrium finding (OEF) and construct multiple types of datasets across a wide range of games using several established methods. To solve the OEF problem, we design a model-based framework that can directly apply any online equilibrium finding algorithm to the OEF setting while making minimal changes. The three most prominent contemporary online equilibrium finding algorithms are adapted to the context of OEF, creating three model-based variants: OEF-PSRO and OEF-CFR, which generalize the widely-used algorithms PSRO and Deep CFR to compute Nash equilibria (NEs), and OEF-JPSRO, which generalizes the JPSRO to calculate (Coarse) Correlated equilibria ((C)CEs). We also combine the behavior cloning policy with the model-based policy to further improve the performance and provide a theoretical guarantee of the solution quality. Extensive experimental results demonstrate the superiority of our approach over offline RL algorithms and the importance of using model-based methods for OEF problems. We hope our work will contribute to advancing research in large-scale equilibrium finding.

Chang Yang, Ruiyu Wang, Xinrun Wang et al.

Generalization in reinforcement learning (RL) is of importance for real deployment of RL algorithms. Various schemes are proposed to address the generalization issues, including transfer learning, multi-task learning and meta learning, as well as the robust and adversarial reinforcement learning. However, there is not a unified formulation of the various schemes, as well as the comprehensive comparisons of methods across different schemes. In this work, we propose a game-theoretic framework for the generalization in reinforcement learning, named GiRL, where an RL agent is trained against an adversary over a set of tasks, where the adversary can manipulate the distributions over tasks within a given threshold. With different configurations, GiRL can reduce the various schemes mentioned above. To solve GiRL, we adapt the widely-used method in game theory, policy space response oracle (PSRO) with the following three important modifications: i) we use model-agnostic meta learning (MAML) as the best-response oracle, ii) we propose a modified projected replicated dynamics, i.e., R-PRD, which ensures the computed meta-strategy of the adversary fall in the threshold, and iii) we also propose a protocol for the few-shot learning of the multiple strategies during testing. Extensive experiments on MuJoCo environments demonstrate that our proposed methods can outperform existing baselines, e.g., MAML.

Yewen Li, Chaojie Wang, Xiaobo Xia et al.

Unsupervised out-of-distribution (U-OOD) detection is to identify OOD data samples with a detector trained solely on unlabeled in-distribution (ID) data. The likelihood function estimated by a deep generative model (DGM) could be a natural detector, but its performance is limited in some popular "hard" benchmarks, such as FashionMNIST (ID) vs. MNIST (OOD). Recent studies have developed various detectors based on DGMs to move beyond likelihood. However, despite their success on "hard" benchmarks, most of them struggle to consistently surpass or match the performance of likelihood on some "non-hard" cases, such as SVHN (ID) vs. CIFAR10 (OOD) where likelihood could be a nearly perfect detector. Therefore, we appeal for more attention to incremental effectiveness on likelihood, i.e., whether a method could always surpass or at least match the performance of likelihood in U-OOD detection. We first investigate the likelihood of variational DGMs and find its detection performance could be improved in two directions: i) alleviating latent distribution mismatch, and ii) calibrating the dataset entropy-mutual integration. Then, we apply two techniques for each direction, specifically post-hoc prior and dataset entropy-mutual calibration. The final method, named Resultant, combines these two directions for better incremental effectiveness compared to either technique alone. Experimental results demonstrate that the Resultant could be a new state-of-the-art U-OOD detector while maintaining incremental effectiveness on likelihood in a wide range of tasks.

Yanchen Deng, Chendong Zhao, Yixuan Li et al.

The discovery of advanced metallic alloys is hindered by vast composition spaces, competing property objectives, and real-world constraints on manufacturability. Here we introduce MATAI, a generalist machine learning framework for property prediction and inverse design of as-cast alloys. MATAI integrates a curated alloy database, deep neural network-based property predictors, a constraint-aware optimization engine, and an iterative AI-experiment feedback loop. The framework estimates key mechanical propertie, sincluding density, yield strength, ultimate tensile strength, and elongation, directly from composition, using multi-task learning and physics-informed inductive biases. Alloy design is framed as a constrained optimization problem and solved using a bi-level approach that combines local search with symbolic constraint programming. We demonstrate MATAI's capabilities on the Ti-based alloy system, a canonical class of lightweight structural materials, where it rapidly identifies candidates that simultaneously achieve lower density (<4.45 g/cm3), higher strength (>1000 MPa) and appreciable ductility (>5%) through only seven iterations. Experimental validation confirms that MATAI-designed alloys outperform commercial references such as TC4, highlighting the framework's potential to accelerate the discovery of lightweight, high-performance materials under real-world design constraints.

Haochong Xia, Yao Long Teng, Regan Tan et al.

In quantitative finance, the gap between training and real-world performance-driven by concept drift and distributional non-stationarity-remains a critical obstacle for building reliable data-driven systems. Models trained on static historical data often overfit, resulting in poor generalization in dynamic markets. The mantra "History Is Not Enough" underscores the need for adaptive data generation that learns to evolve with the market rather than relying solely on past observations. We present a drift-aware dataflow system that integrates machine learning-based adaptive control into the data curation process. The system couples a parameterized data manipulation module comprising single-stock transformations, multi-stock mix-ups, and curation operations, with an adaptive planner-scheduler that employs gradient-based bi-level optimization to control the system. This design unifies data augmentation, curriculum learning, and data workflow management under a single differentiable framework, enabling provenance-aware replay and continuous data quality monitoring. Extensive experiments on forecasting and reinforcement learning trading tasks demonstrate that our framework enhances model robustness and improves risk-adjusted returns. The system provides a generalizable approach to adaptive data management and learning-guided workflow automation for financial data.

Molei Qin, Xinyu Cai, Yewen Li et al.

Futures are contracts obligating the exchange of an asset at a predetermined date and price, notable for their high leverage and liquidity and, therefore, thrive in the Crypto market. RL has been widely applied in various quantitative tasks. However, most methods focus on the spot and could not be directly applied to the futures market with high leverage because of 2 challenges. First, high leverage amplifies reward fluctuations, making training stochastic and difficult to converge. Second, prior works lacked self-awareness of capability boundaries, exposing them to the risk of significant loss when encountering new market state (e.g.,a black swan event like COVID-19). To tackle these challenges, we propose the Efficient and Risk-Aware Ensemble Reinforcement Learning for Futures Trading (FineFT), a novel three-stage ensemble RL framework with stable training and proper risk management. In stage I, ensemble Q learners are selectively updated by ensemble TD errors to improve convergence. In stage II, we filter the Q-learners based on their profitabilities and train VAEs on market states to identify the capability boundaries of the learners. In stage III, we choose from the filtered ensemble and a conservative policy, guided by trained VAEs, to maintain profitability and mitigate risk with new market states. Through extensive experiments on crypto futures in a high-frequency trading environment with high fidelity and 5x leverage, we demonstrate that FineFT outperforms 12 SOTA baselines in 6 financial metrics, reducing risk by more than 40% while achieving superior profitability compared to the runner-up. Visualization of the selective update mechanism shows that different agents specialize in distinct market dynamics, and ablation studies certify routing with VAEs reduces maximum drawdown effectively, and selective update improves convergence and performance.

Shuxin Li, Chang Yang, Youzhi Zhang et al.

Nash equilibrium (NE) is a widely adopted solution concept in game theory due to its stability property. However, we observe that the NE strategy might not always yield the best results, especially against opponents who do not adhere to NE strategies. Based on this observation, we pose a new game-solving question: Can we learn a model that can exploit any, even NE, opponent to maximize their own utility? In this work, we make the first attempt to investigate this problem through in-context learning. Specifically, we introduce a novel method, In-Context Exploiter (ICE), to train a single model that can act as any player in the game and adaptively exploit opponents entirely by in-context learning. Our ICE algorithm involves generating diverse opponent strategies, collecting interactive history training data by a reinforcement learning algorithm, and training a transformer-based agent within a well-designed curriculum learning framework. Finally, comprehensive experimental results validate the effectiveness of our ICE algorithm, showcasing its in-context learning ability to exploit any unknown opponent, thereby positively answering our initial game-solving question.

Qinggang Zhang, Zhishang Xiang, Yilin Xiao et al.

Large language models (LLMs) augmented with retrieval systems have demonstrated significant potential in handling knowledge-intensive tasks. However, these models often struggle with unfaithfulness issues, generating outputs that either ignore the retrieved context or inconsistently blend it with the LLM`s parametric knowledge. This issue is particularly severe in cases of knowledge conflict, where the retrieved context conflicts with the model`s parametric knowledge. While existing faithful RAG approaches enforce strict context adherence through well-designed prompts or modified decoding strategies, our analysis reveals a critical limitation: they achieve faithfulness by forcibly suppressing the model`s parametric knowledge, which undermines the model`s internal knowledge structure and increases the risk of misinterpreting the context. To this end, this paper proposes FaithfulRAG, a novel framework that resolves knowledge conflicts by explicitly modeling discrepancies between the model`s parametric knowledge and retrieved context. Specifically, FaithfulRAG identifies conflicting knowledge at the fact level and designs a self-thinking process, allowing LLMs to reason about and integrate conflicting facts before generating responses. Extensive experiments demonstrate that our method outperforms state-of-the-art methods. The code is available at https://github.com/DeepLearnXMU/Faithful-RAG

Wentao Zhang, Yilei Zhao, Chuqiao Zong et al.

Financial AI holds great promise for transforming modern finance, with the potential to support a wide range of tasks such as market forecasting, portfolio management, quantitative trading, and automated analysis. However, existing platforms remain limited in task coverage, lack robust multimodal data integration, and offer insufficient support for the training and deployment of large language models (LLMs). In response to these limitations, we present FinWorld, an all-in-one open-source platform that provides end-to-end support for the entire financial AI workflow, from data acquisition to experimentation and deployment. FinWorld distinguishes itself through native integration of heterogeneous financial data, unified support for diverse AI paradigms, and advanced agent automation, enabling seamless development and deployment. Leveraging data from 2 representative markets, 4 stock pools, and over 800 million financial data points, we conduct comprehensive experiments on 4 key financial AI tasks. These experiments systematically evaluate deep learning and reinforcement learning algorithms, with particular emphasis on RL-based finetuning for LLMs and LLM Agents. The empirical results demonstrate that FinWorld significantly enhances reproducibility, supports transparent benchmarking, and streamlines deployment, thereby providing a strong foundation for future research and real-world applications. Code is available at Github~\footnote{https://github.com/DVampire/FinWorld}.

Xinrun Wang, Chang Yang, He Zhao et al.

LLM-based foundation agents that perceive, reason, and act across thousands of reasoning steps are rapidly becoming the dominant paradigm for deploying artificial intelligence in open-ended, long-horizon complex tasks. Despite this significance, the field remains overwhelmingly engineering-driven. Engineering practice has converged on useful primitives (tool loops, memory banks, harnesses, reflection steps), yet these are assembled by empirical trial and error rather than from first principles. Fundamental questions remain open: under what conditions does a long-running agent remain on-task? How should an agent respond when its environment exceeds its representational capacity? What architectural properties are necessary for safe self-improvement? We argue that cybernetics, the mid-twentieth-century science of control and communication in complex systems, provides the missing theoretical scaffold for foundation agents. By mapping six canonical laws of classical cybernetics onto six agent design principles, and synthesizing those principles into three engineering desiderata (reliability, lifelong running, and self-Improvement), we arrive at a framework termed Agent Cybernetics. Three application domains, code generation, computer use and automated research, exemplify the analytical framework of agent cybernetics by identifying failure modes and concrete engineering recommendations. We hope that agent cybernetics opens a new research venue and establishes the scientific foundation that foundation agents need for principled, reliable real-world deployment.

Caijun Xu, Changyi Xiao, Zhongyuan Peng et al.

Reinforcement learning (RL) offers a principled way to enhance the reasoning capabilities of large language models, yet its effectiveness hinges on training signals that remain informative as models evolve. In practice, RL progress often slows when task difficulty becomes poorly aligned with model capability, or when training is dominated by a narrow set of recurring problem patterns. To jointly address these issues, we propose SCALER (Synthetic sCalable Adaptive Learning Environment for Reasoning), a framework that sustains effective learning signals through adaptive environment design. SCALER introduces a scalable synthesis pipeline that converts real-world programming problems into verifiable reasoning environments with controllable difficulty and unbounded instance generation, enabling RL training beyond finite datasets while preserving strong correctness guarantees. Building on this, SCALER further employs an adaptive multi-environment RL strategy that dynamically adjusts instance difficulty and curates the active set of environments to track the model's capability frontier and maintain distributional diversity. This co-adaptation prevents reward sparsity, mitigates overfitting to narrow task patterns, and supports sustained improvement throughout training. Extensive experiments show that SCALER consistently outperforms dataset-based RL baselines across diverse reasoning benchmarks and exhibits more stable, long-horizon training dynamics.

Wentao Zhang, Lingxuan Zhao, Haochong Xia et al.

Financial trading is a crucial component of the markets, informed by a multimodal information landscape encompassing news, prices, and Kline charts, and encompasses diverse tasks such as quantitative trading and high-frequency trading with various assets. While advanced AI techniques like deep learning and reinforcement learning are extensively utilized in finance, their application in financial trading tasks often faces challenges due to inadequate handling of multimodal data and limited generalizability across various tasks. To address these challenges, we present FinAgent, a multimodal foundational agent with tool augmentation for financial trading. FinAgent's market intelligence module processes a diverse range of data-numerical, textual, and visual-to accurately analyze the financial market. Its unique dual-level reflection module not only enables rapid adaptation to market dynamics but also incorporates a diversified memory retrieval system, enhancing the agent's ability to learn from historical data and improve decision-making processes. The agent's emphasis on reasoning for actions fosters trust in its financial decisions. Moreover, FinAgent integrates established trading strategies and expert insights, ensuring that its trading approaches are both data-driven and rooted in sound financial principles. With comprehensive experiments on 6 financial datasets, including stocks and Crypto, FinAgent significantly outperforms 9 state-of-the-art baselines in terms of 6 financial metrics with over 36% average improvement on profit. Specifically, a 92.27% return (a 84.39% relative improvement) is achieved on one dataset. Notably, FinAgent is the first advanced multimodal foundation agent designed for financial trading tasks.

Weihao Tan, Wentao Zhang, Xinrun Xu et al.

Despite the success in specific scenarios, existing foundation agents still struggle to generalize across various virtual scenarios, mainly due to the dramatically different encapsulations of environments with manually designed observation and action spaces. To handle this issue, we propose the General Computer Control (GCC) setting to restrict foundation agents to interact with software through the most unified and standardized interface, i.e., using screenshots as input and keyboard and mouse actions as output. We introduce Cradle, a modular and flexible LMM-powered framework, as a preliminary attempt towards GCC. Enhanced by six key modules, Cradle can understand input screenshots and output executable code for low-level keyboard and mouse control after high-level planning, so that Cradle can interact with any software and complete long-horizon complex tasks without relying on any built-in APIs. Experimental results show that Cradle exhibits remarkable generalizability and impressive performance across four previously unexplored commercial video games, five software applications, and a comprehensive benchmark, OSWorld. Cradle is the first to enable foundation agents to follow the main storyline and complete 40-minute-long real missions in the complex AAA game Red Dead Redemption 2 (RDR2). Cradle can also create a city of a thousand people in Cities: Skylines, farm and harvest parsnips in Stardew Valley, and trade and bargain with a maximal weekly total profit of 87% in Dealer's Life 2. Cradle can not only operate daily software, like Chrome, Outlook, and Feishu, but also edit images and videos using Meitu and CapCut. Cradle greatly extends the reach of foundation agents by enabling the easy conversion of any software, especially complex games, into benchmarks to evaluate agents' various abilities and facilitate further data collection, thus paving the way for generalist agents.

Longtao Zheng, Zhiyuan Huang, Zhenghai Xue et al.

General virtual agents need to handle multimodal observations, master complex action spaces, and self-improve in dynamic, open-domain environments. However, existing environments are often domain-specific and require complex setups, which limits agent development and evaluation in real-world settings. As a result, current evaluations lack in-depth analyses that decompose fundamental agent capabilities. We introduce AgentStudio, a trinity of environments, tools, and benchmarks to address these issues. AgentStudio provides a lightweight, interactive environment with highly generic observation and action spaces, e.g., video observations and GUI/API actions. It integrates tools for creating online benchmark tasks, annotating GUI elements, and labeling actions in videos. Based on our environment and tools, we curate an online task suite that benchmarks both GUI interactions and function calling with efficient auto-evaluation. We also reorganize existing datasets and collect new ones using our tools to establish three datasets: GroundUI, IDMBench, and CriticBench. These datasets evaluate fundamental agent abilities, including GUI grounding, learning from videos, and success detection, pointing to the desiderata for robust, general, and open-ended virtual agents.

Chaojie Wang, Yishi Xu, Zhong Peng et al.

Large language models (LLMs) have exhibited remarkable performance on various natural language processing (NLP) tasks, especially for question answering. However, in the face of problems beyond the scope of knowledge, these LLMs tend to talk nonsense with a straight face, where the potential solution could be incorporating an Information Retrieval (IR) module and generating response based on these retrieved knowledge. In this paper, we present a novel framework to assist LLMs, such as ChatGPT, to retrieve question-related structured information on the knowledge graph, and demonstrate that Knowledge-based question answering (Keqing) could be a nature Chain-of-Thought (CoT) mentor to guide the LLM to sequentially find the answer entities of a complex question through interpretable logical chains. Specifically, the workflow of Keqing will execute decomposing a complex question according to predefined templates, retrieving candidate entities on knowledge graph, reasoning answers of sub-questions, and finally generating response with reasoning paths, which greatly improves the reliability of LLM's response. The experimental results on KBQA datasets show that Keqing can achieve competitive performance and illustrate the logic of answering each question.

Hangfan Zhang, Zhiyao Cui, Jianhao Chen et al.

Multi-agent debate (MAD) has gained significant attention as a promising line of research to improve the factual accuracy and reasoning capabilities of large language models (LLMs). Despite its conceptual appeal, current MAD research suffers from critical limitations in evaluation practices, including limited benchmark coverage, weak baseline comparisons, and inconsistent setups. This paper presents a systematic evaluation of 5 representative MAD methods across 9 benchmarks using 4 foundational models. Surprisingly, our findings reveal that MAD often fail to outperform simple single-agent baselines such as Chain-of-Thought and Self-Consistency, even when consuming significantly more inference-time computation. To advance MAD research, we further explore the role of model heterogeneity and find it as a universal antidote to consistently improve current MAD frameworks. Based on our findings, we argue that the field must stop overvaluing MAD in its current form; for true advancement, we must critically rethink evaluation paradigms and actively embrace model heterogeneity as a core design principle.

Xinrun Wang, Deshun Xia, Ke Xu et al.

Accurate trajectory prediction is fundamentally challenging due to high scene heterogeneity - the severe variance in motion velocity, spatial density, and interaction patterns across different real-world environments. However, most existing approaches typically train a single unified model, expecting a fixed-capacity architecture to generalize universally across all possible scenarios. This conventional model-centric paradigm is fundamentally flawed when confronting such extreme heterogeneity, inevitably leading to a severe generalization gap, degraded accuracy, and massive computational waste. To overcome this bottleneck, rather than refining restricted model-centric architectures, we propose selective learning, a novel scene-centric paradigm. It explicitly analyzes the characteristics of the underlying scene to dynamically route inputs to the most appropriate expert models. As a concrete implementation of this paradigm, we introduce SceneSelect. Specifically, SceneSelect utilizes unsupervised clustering on interpretable geometric and kinematic features to discover a latent scene taxonomy. A highly decoupled classification module is then trained to assign real-time inputs to these scene categories, and a highly extensible, plug-and-play scheduling policy automatically dispatches the trajectory sequence to the optimal expert predictor. Crucially, this decoupled design ensures excellent generalization capabilities, allowing seamless integration with different off-the-shelf models and robust adaptation across new datasets without requiring computationally expensive joint retraining. Extensive experiments on three public benchmarks (ETH-UCY, SDD, and NBA) demonstrate that our method consistently outperforms strong single-model and ensemble baselines, achieving an average improvement of 10.5%, showcasing the effectiveness of scene-aware selective learning.

Chang Yang, Xinrun Wang, Junzhe Jiang et al.

World model emerges as a key module in decision making, where MuZero and Dreamer achieve remarkable successes in complex tasks. Recent work leverages Large Language Models (LLMs) as general world simulators to simulate the dynamics of the world due to their generalizability. LLMs also serve as the world model for deliberative reasoning in Reasoning via Planning (RAP) and Tree of Thought (ToT). However, the world models are either evaluated as a general world simulator, or as a functional module of the agent, i.e., predicting the transitions to assist the planning. In this work, we propose a comprehensive evaluation of the world models with LLMs from the decision making perspective. Specifically, we leverage the 31 diverse environments from (Wang et al., 2023;2024) and curate the rule-based policy of each environment for the diverse evaluation. Then, we design three main tasks, i.e., policy verification, action proposal, and policy planning, where the world models can be used for decision making solely. Finally, we conduct the comprehensive evaluation of the advanced LLMs, i.e., GPT-4o and GPT-4o-mini, on the environments for the three main tasks under various settings. The key observations include: i) GPT-4o significantly outperforms GPT-4o-mini on the three main tasks, especially for the tasks which require the domain knowledge, ii) the performance of the world model with LLM will be decreased for long-term decision-making tasks, and iii) the combination of different functionalities of the world model will brings additional unstabilities of the performance.

Pengdeng Li, Shuxin Li, Xinrun Wang et al.

Pursuit-evasion games (PEGs) model interactions between a team of pursuers and an evader in graph-based environments such as urban street networks. Recent advancements have demonstrated the effectiveness of the pre-training and fine-tuning paradigm in PSRO to improve scalability in solving large-scale PEGs. However, these methods primarily focus on specific PEGs with fixed initial conditions that may vary substantially in real-world scenarios, which significantly hinders the applicability of the traditional methods. To address this issue, we introduce Grasper, a GeneRAlist purSuer for Pursuit-Evasion pRoblems, capable of efficiently generating pursuer policies tailored to specific PEGs. Our contributions are threefold: First, we present a novel architecture that offers high-quality solutions for diverse PEGs, comprising critical components such as (i) a graph neural network (GNN) to encode PEGs into hidden vectors, and (ii) a hypernetwork to generate pursuer policies based on these hidden vectors. As a second contribution, we develop an efficient three-stage training method involving (i) a pre-pretraining stage for learning robust PEG representations through self-supervised graph learning techniques like GraphMAE, (ii) a pre-training stage utilizing heuristic-guided multi-task pre-training (HMP) where heuristic-derived reference policies (e.g., through Dijkstra's algorithm) regularize pursuer policies, and (iii) a fine-tuning stage that employs PSRO to generate pursuer policies on designated PEGs. Finally, we perform extensive experiments on synthetic and real-world maps, showcasing Grasper's significant superiority over baselines in terms of solution quality and generalizability. We demonstrate that Grasper provides a versatile approach for solving pursuit-evasion problems across a broad range of scenarios, enabling practical deployment in real-world situations.

Junnan Dong, Zijin Hong, Yuanchen Bei et al.

Large language models (LLMs) have demonstrated their remarkable performance across various language understanding tasks. While emerging benchmarks have been proposed to evaluate LLMs in various domains such as mathematics and computer science, they merely measure the accuracy in terms of the final prediction on multi-choice questions. However, it remains insufficient to verify the essential understanding of LLMs given a chosen choice. To fill this gap, we present CLR-Bench to comprehensively evaluate the LLMs in complex college-level reasoning. Specifically, (i) we prioritize 16 challenging college disciplines in computer science and artificial intelligence. The dataset contains 5 types of questions, while each question is associated with detailed explanations from experts. (ii) To quantify a fair evaluation of LLMs' reasoning ability, we formalize the criteria with two novel metrics. Q$\rightarrow$A is utilized to measure the performance of direct answer prediction, and Q$\rightarrow$AR effectively considers the joint ability to answer the question and provide rationale simultaneously. Extensive experiments are conducted with 40 LLMs over 1,018 discipline-specific questions. The results demonstrate the key insights that LLMs, even the best closed-source LLM, i.e., GPT-4 turbo, tend to `guess' the college-level answers. It shows a dramatic decrease in accuracy from 63.31% Q$\rightarrow$A to 39.00% Q$\rightarrow$AR, indicating an unsatisfactory reasoning ability.

Chang Yang, Ruiyu Wang, Junzhe Jiang et al.

Reasoning is the fundamental capability of large language models (LLMs). Due to the rapid progress of LLMs, there are two main issues of current benchmarks: i) these benchmarks can be crushed in a short time (less than 1 year), and ii) these benchmarks may be easily hacked. To handle these issues, we propose the ever-scalingness for building the benchmarks which are uncrushable, unhackable, auto-verifiable and general. This paper presents Nondeterministic Polynomial-time Problem Challenge (NPPC), an ever-scaling reasoning benchmark for LLMs. Specifically, the NPPC has three main modules: i) npgym, which provides a unified interface of 25 well-known NP-complete problems and can generate any number of instances with any levels of complexities, ii) npsolver: which provides a unified interface to evaluate the problem instances with both online and offline models via APIs and local deployments, respectively, and iii) npeval: which provides the comprehensive and ready-to-use tools to analyze the performances of LLMs over different problems, the number of tokens, the aha moments, the reasoning errors and the solution errors. Extensive experiments over widely-used LLMs demonstrate: i) NPPC can successfully decrease the performances of advanced LLMs' performances to below 10%, demonstrating that NPPC is uncrushable, ii) DeepSeek-R1, Claude-3.7-Sonnet, and o1/o3-mini are the most powerful LLMs, where DeepSeek-R1 outperforms Claude-3.7-Sonnet and o1/o3-mini in most NP-complete problems considered, and iii) the numbers of tokens, aha moments in the advanced LLMs, e.g., Claude-3.7-Sonnet and DeepSeek-R1, are observed first to increase and then decrease when the problem instances become more and more difficult. We believe that NPPC is the first ever-scaling reasoning benchmark, serving as the uncrushable and unhackable testbed for LLMs toward artificial general intelligence (AGI).

Pengdeng Li, Shuxin Li, Chang Yang et al.

Policy-Space Response Oracles (PSRO) as a general algorithmic framework has achieved state-of-the-art performance in learning equilibrium policies of two-player zero-sum games. However, the hand-crafted hyperparameter value selection in most of the existing works requires extensive domain knowledge, forming the main barrier to applying PSRO to different games. In this work, we make the first attempt to investigate the possibility of self-adaptively determining the optimal hyperparameter values in the PSRO framework. Our contributions are three-fold: (1) Using several hyperparameters, we propose a parametric PSRO that unifies the gradient descent ascent (GDA) and different PSRO variants. (2) We propose the self-adaptive PSRO (SPSRO) by casting the hyperparameter value selection of the parametric PSRO as a hyperparameter optimization (HPO) problem where our objective is to learn an HPO policy that can self-adaptively determine the optimal hyperparameter values during the running of the parametric PSRO. (3) To overcome the poor performance of online HPO methods, we propose a novel offline HPO approach to optimize the HPO policy based on the Transformer architecture. Experiments on various two-player zero-sum games demonstrate the superiority of SPSRO over different baselines.

Weihao Tan, Changjiu Jiang, Yu Duan et al.

Autonomous agents navigating human society must master both production activities and social interactions, yet existing benchmarks rarely evaluate these skills simultaneously. To bridge this gap, we introduce StarDojo, a novel benchmark based on Stardew Valley, designed to assess AI agents in open-ended production-living simulations. In StarDojo, agents are tasked to perform essential livelihood activities such as farming and crafting, while simultaneously engaging in social interactions to establish relationships within a vibrant community. StarDojo features 1,000 meticulously curated tasks across five key domains: farming, crafting, exploration, combat, and social interactions. Additionally, we provide a compact subset of 100 representative tasks for efficient model evaluation. The benchmark offers a unified, user-friendly interface that eliminates the need for keyboard and mouse control, supports all major operating systems, and enables the parallel execution of multiple environment instances, making it particularly well-suited for evaluating the most capable foundation agents, powered by multimodal large language models (MLLMs). Extensive evaluations of state-of-the-art MLLMs agents demonstrate substantial limitations, with the best-performing model, GPT-4.1, achieving only a 12.7% success rate, primarily due to challenges in visual understanding, multimodal reasoning and low-level manipulation. As a user-friendly environment and benchmark, StarDojo aims to facilitate further research towards robust, open-ended agents in complex production-living environments.

Pengdeng Li, Shuxin Li, Chang Yang et al.

Decision-making problems, categorized as single-agent, e.g., Atari, cooperative multi-agent, e.g., Hanabi, competitive multi-agent, e.g., Hold'em poker, and mixed cooperative and competitive, e.g., football, are ubiquitous in the real world. Various methods are proposed to address the specific decision-making problems. Despite the successes in specific categories, these methods typically evolve independently and cannot generalize to other categories. Therefore, a fundamental question for decision-making is: \emph{Can we develop \textbf{a single algorithm} to tackle \textbf{ALL} categories of decision-making problems?} There are several main challenges to address this question: i) different decision-making categories involve different numbers of agents and different relationships between agents, ii) different categories have different solution concepts and evaluation measures, and iii) there lacks a comprehensive benchmark covering all the categories. This work presents a preliminary attempt to address the question with three main contributions. i) We propose the generalized mirror descent (GMD), a generalization of MD variants, which considers multiple historical policies and works with a broader class of Bregman divergences. ii) We propose the configurable mirror descent (CMD) where a meta-controller is introduced to dynamically adjust the hyper-parameters in GMD conditional on the evaluation measures. iii) We construct the \textsc{GameBench} with 15 academic-friendly games across different decision-making categories. Extensive experiments demonstrate that CMD achieves empirically competitive or better outcomes compared to baselines while providing the capability of exploring diverse dimensions of decision making.

Hangfan Zhang, Siyuan Xu, Zhimeng Guo et al.

Reinforcement learning (RL) has demonstrated potential in enhancing the reasoning capabilities of large language models (LLMs), but such training typically demands substantial efforts in creating and annotating data. In this work, we explore improving LLMs through RL with minimal data. Our approach alternates between the LLM proposing a task and then attempting to solve it. To minimize data dependency, we introduce two novel mechanisms grounded in self-awareness: (1) self-aware difficulty prediction, where the model learns to assess task difficulty relative to its own abilities and prioritize challenging yet solvable tasks, and (2) self-aware limit breaking, where the model recognizes when a task is beyond its capability boundary and proactively requests external data to break through that limit. Extensive experiments on nine benchmarks showing a 53.8% relative improvement with less than 1.2% extra data demonstrate the efficacy of self-aware RL and underscore the promise of self-evolving agent training.

Yanchen Deng, Xinrun Wang, Bo An

Local search is an important class of incomplete algorithms for solving Distributed Constraint Optimization Problems (DCOPs) but it often converges to poor local optima. While GDBA provides a comprehensive rule set to escape premature convergence, its empirical benefits remain marginal on general-valued problems. In this work, we systematically examine GDBA and identify three factors that potentially lead to its inferior performance, i.e., over-aggressive constraint violation conditions, unbounded penalty accumulation, and uncoordinated penalty updates. To address these issues, we propose Distributed Guided Local Search (DGLS), a novel GLS framework for DCOPs that incorporates an adaptive violation condition to selectively penalize constraints with high cost, a penalty evaporation mechanism to control the magnitude of penalization, and a synchronization scheme for coordinated penalty updates. We theoretically show that the penalty values are bounded, and agents play a potential game in our DGLS. Our extensive empirical results on various standard benchmarks demonstrate the great superiority of DGLS over state-of-the-art baselines. Particularly, compared to Damped Max-sum with high damping factors (e.g., 0.7 or 0.9), our DGLS achieves competitive performance on general-valued problems, and outperforms it by significant margins (\textbf{3.77\%--66.3\%}) on structured problems in terms of anytime results.

Penghui Yang, Chendong Zhao, Bijun Tang et al.

Alloy discovery is central to advancing modern industry but remains hindered by the vastness of compositional design space and the costly validation. Here, we present AutoMAT, a hierarchical and autonomous framework grounded in and validated by experiments, which integrates large language models, automated CALPHAD-based simulations, and AI-driven search to accelerate alloy design. Spanning the entire pipeline from ideation to validation, AutoMAT achieves high efficiency, accuracy, and interpretability without the need for manually curated large datasets. In a case study targeting a lightweight, high-strength alloy, AutoMAT identifies a titanium alloy with 8.1% lower density and comparable yield strength relative to the state-of-the-art reference, achieving the highest specific strength among all comparisons. In a second case targeting high-yield-strength high-entropy alloys, AutoMAT achieves a 28.2% improvement in yield strength over the base alloy. In both cases, AutoMAT reduces the discovery timeline from years to weeks, illustrating its potential as a scalable and versatile platform for next-generation alloy design.

Junzhe Jiang, Chang Yang, Xinrun Wang et al.

Stock market indices serve as fundamental market measurement that quantify systematic market dynamics. However, accurate index price prediction remains challenging, primarily because existing approaches treat indices as isolated time series and frame the prediction as a simple regression task. These methods fail to capture indices' inherent nature as aggregations of constituent stocks with complex, time-varying interdependencies. To address these limitations, we propose Cubic, a novel end-to-end framework that explicitly models the adaptive fusion of constituent stocks for index price prediction. Our main contributions are threefold. i) Fusion in the latent space: we introduce the fusion mechanism over the latent embedding of the stocks to extract the information from the vast number of stocks. ii) Binary encoding classification: since regression tasks are challenging due to continuous value estimation, we reformulate the regression into the classification task, where the target value is converted to binary and we optimize the prediction of the value of each digit with cross-entropy loss. iii) Confidence-guided prediction and trading: we introduce the regularization loss to address market prediction uncertainty for the index prediction and design the rule-based trading policies based on the confidence. Extensive experiments across multiple stock markets and indices demonstrate that Cubic consistently outperforms state-of-the-art baselines in stock index prediction tasks, achieving superior performance on both forecasting accuracy metrics and downstream trading profitability.

Junzhe Jiang, Chang Yang, Xinrun Wang et al.

The latency of the exchanges in Market Making (MM) is inevitable due to hardware limitations, system processing times, delays in receiving data from exchanges, the time required for order transmission to reach the market, etc. Existing reinforcement learning (RL) methods for Market Making (MM) overlook the impact of these latency, which can lead to unintended order cancellations due to price discrepancies between decision and execution times and result in undesired inventory accumulation, exposing MM traders to increased market risk. Therefore, these methods cannot be applied in real MM scenarios. To address these issues, we first build a realistic MM environment with random delays of 30-100 milliseconds for order placement and market information reception, and implement a batch matching mechanism that collects orders within every 500 milliseconds before matching them all at once, simulating the batch auction mechanisms adopted by some exchanges. Then, we propose Relaver, an RL-based method for MM to tackle the latency and inventory risk issues. The three main contributions of Relaver are: i) we introduce an augmented state-action space that incorporates order hold time alongside price and volume, enabling Relaver to optimize execution strategies under latency constraints and time-priority matching mechanisms, ii) we leverage dynamic programming (DP) to guide the exploration of RL training for better policies, iii) we train a market trend predictor, which can guide the agent to intelligently adjust the inventory to reduce the risk. Extensive experiments and ablation studies on four real-world datasets demonstrate that \textsc{Relaver} significantly improves the performance of state-of-the-art RL-based MM strategies across multiple metrics.

Junzhe Jiang, Chang Yang, Aixin Cui et al.

Financial tasks are pivotal to global economic stability; however, their execution faces challenges including labor intensive processes, low error tolerance, data fragmentation, and tool limitations. Although large language models (LLMs) have succeeded in various natural language processing tasks and have shown potential in automating workflows through reasoning and contextual understanding, current benchmarks for evaluating LLMs in finance lack sufficient domain-specific data, have simplistic task design, and incomplete evaluation frameworks. To address these gaps, this article presents FinMaster, a comprehensive financial benchmark designed to systematically assess the capabilities of LLM in financial literacy, accounting, auditing, and consulting. Specifically, FinMaster comprises three main modules: i) FinSim, which builds simulators that generate synthetic, privacy-compliant financial data for companies to replicate market dynamics; ii) FinSuite, which provides tasks in core financial domains, spanning 183 tasks of various types and difficulty levels; and iii) FinEval, which develops a unified interface for evaluation. Extensive experiments over state-of-the-art LLMs reveal critical capability gaps in financial reasoning, with accuracy dropping from over 90% on basic tasks to merely 40% on complex scenarios requiring multi-step reasoning. This degradation exhibits the propagation of computational errors, where single-metric calculations initially demonstrating 58% accuracy decreased to 37% in multimetric scenarios. To the best of our knowledge, FinMaster is the first benchmark that covers full-pipeline financial workflows with challenging tasks. We hope that FinMaster can bridge the gap between research and industry practitioners, driving the adoption of LLMs in real-world financial practices to enhance efficiency and accuracy.

Chuqiao Zong, Chaojie Wang, Molei Qin et al.

High-frequency trading (HFT) that executes algorithmic trading in short time scales, has recently occupied the majority of cryptocurrency market. Besides traditional quantitative trading methods, reinforcement learning (RL) has become another appealing approach for HFT due to its terrific ability of handling high-dimensional financial data and solving sophisticated sequential decision-making problems, \emph{e.g.,} hierarchical reinforcement learning (HRL) has shown its promising performance on second-level HFT by training a router to select only one sub-agent from the agent pool to execute the current transaction. However, existing RL methods for HFT still have some defects: 1) standard RL-based trading agents suffer from the overfitting issue, preventing them from making effective policy adjustments based on financial context; 2) due to the rapid changes in market conditions, investment decisions made by an individual agent are usually one-sided and highly biased, which might lead to significant loss in extreme markets. To tackle these problems, we propose a novel Memory Augmented Context-aware Reinforcement learning method On HFT, \emph{a.k.a.} MacroHFT, which consists of two training phases: 1) we first train multiple types of sub-agents with the market data decomposed according to various financial indicators, specifically market trend and volatility, where each agent owns a conditional adapter to adjust its trading policy according to market conditions; 2) then we train a hyper-agent to mix the decisions from these sub-agents and output a consistently profitable meta-policy to handle rapid market fluctuations, equipped with a memory mechanism to enhance the capability of decision-making. Extensive experiments on various cryptocurrency markets demonstrate that MacroHFT can achieve state-of-the-art performance on minute-level trading tasks.

Weihao Tan, Wentao Zhang, Shanqi Liu et al.

Despite the impressive performance across numerous tasks, large language models (LLMs) often fail in solving simple decision-making tasks due to the misalignment of the knowledge in LLMs with environments. On the contrary, reinforcement learning (RL) agents learn policies from scratch, which makes them always align with environments but difficult to incorporate prior knowledge for efficient explorations. To narrow the gap, we propose TWOSOME, a novel general online framework that deploys LLMs as decision-making agents to efficiently interact and align with embodied environments via RL without requiring any prepared datasets or prior knowledge of the environments. Firstly, we query the joint probabilities of each valid action with LLMs to form behavior policies. Then, to enhance the stability and robustness of the policies, we propose two normalization methods and summarize four prompt design principles. Finally, we design a novel parameter-efficient training architecture where the actor and critic share one frozen LLM equipped with low-rank adapters (LoRA) updated by PPO. We conduct extensive experiments to evaluate TWOSOME. i) TWOSOME exhibits significantly better sample efficiency and performance compared to the conventional RL method, PPO, and prompt tuning method, SayCan, in both classical decision-making environment, Overcooked, and simulated household environment, VirtualHome. ii) Benefiting from LLMs' open-vocabulary feature, TWOSOME shows superior generalization ability to unseen tasks. iii) Under our framework, there is no significant loss of the LLMs' original ability during online PPO finetuning.

Wanqi Xue, Youzhi Zhang, Shuxin Li et al.

Securing networked infrastructures is important in the real world. The problem of deploying security resources to protect against an attacker in networked domains can be modeled as Network Security Games (NSGs). Unfortunately, existing approaches, including the deep learning-based approaches, are inefficient to solve large-scale extensive-form NSGs. In this paper, we propose a novel learning paradigm, NSG-NFSP, to solve large-scale extensive-form NSGs based on Neural Fictitious Self-Play (NFSP). Our main contributions include: i) reforming the best response (BR) policy network in NFSP to be a mapping from action-state pair to action-value, to make the calculation of BR possible in NSGs; ii) converting the average policy network of an NFSP agent into a metric-based classifier, helping the agent to assign distributions only on legal actions rather than all actions; iii) enabling NFSP with high-level actions, which can benefit training efficiency and stability in NSGs; and iv) leveraging information contained in graphs of NSGs by learning efficient graph node embeddings. Our algorithm significantly outperforms state-of-the-art algorithms in both scalability and solution quality.

Shuxin Li, Youzhi Zhang, Xinrun Wang et al.

In many real-world scenarios, a team of agents coordinate with each other to compete against an opponent. The challenge of solving this type of game is that the team's joint action space grows exponentially with the number of agents, which results in the inefficiency of the existing algorithms, e.g., Counterfactual Regret Minimization (CFR). To address this problem, we propose a new framework of CFR: CFR-MIX. Firstly, we propose a new strategy representation that represents a joint action strategy using individual strategies of all agents and a consistency relationship to maintain the cooperation between agents. To compute the equilibrium with individual strategies under the CFR framework, we transform the consistency relationship between strategies to the consistency relationship between the cumulative regret values. Furthermore, we propose a novel decomposition method over cumulative regret values to guarantee the consistency relationship between the cumulative regret values. Finally, we introduce our new algorithm CFR-MIX which employs a mixing layer to estimate cumulative regret values of joint actions as a non-linear combination of cumulative regret values of individual actions. Experimental results show that CFR-MIX outperforms existing algorithms on various games significantly.

Aye Phyu Phyu Aung, Xinrun Wang, Runsheng Yu et al.

In this paper, we propose a new approach to train Generative Adversarial Networks (GANs) where we deploy a double-oracle framework using the generator and discriminator oracles. GAN is essentially a two-player zero-sum game between the generator and the discriminator. Training GANs is challenging as a pure Nash equilibrium may not exist and even finding the mixed Nash equilibrium is difficult as GANs have a large-scale strategy space. In DO-GAN, we extend the double oracle framework to GANs. We first generalize the players' strategies as the trained models of generator and discriminator from the best response oracles. We then compute the meta-strategies using a linear program. For scalability of the framework where multiple generators and discriminator best responses are stored in the memory, we propose two solutions: 1) pruning the weakly-dominated players' strategies to keep the oracles from becoming intractable; 2) applying continual learning to retain the previous knowledge of the networks. We apply our framework to established GAN architectures such as vanilla GAN, Deep Convolutional GAN, Spectral Normalization GAN and Stacked GAN. Finally, we conduct experiments on MNIST, CIFAR-10 and CelebA datasets and show that DO-GAN variants have significant improvements in both subjective qualitative evaluation and quantitative metrics, compared with their respective GAN architectures.

Wei Qiu, Xinrun Wang, Runsheng Yu et al.

Current value-based multi-agent reinforcement learning methods optimize individual Q values to guide individuals' behaviours via centralized training with decentralized execution (CTDE). However, such expected, i.e., risk-neutral, Q value is not sufficient even with CTDE due to the randomness of rewards and the uncertainty in environments, which causes the failure of these methods to train coordinating agents in complex environments. To address these issues, we propose RMIX, a novel cooperative MARL method with the Conditional Value at Risk (CVaR) measure over the learned distributions of individuals' Q values. Specifically, we first learn the return distributions of individuals to analytically calculate CVaR for decentralized execution. Then, to handle the temporal nature of the stochastic outcomes during executions, we propose a dynamic risk level predictor for risk level tuning. Finally, we optimize the CVaR policies with CVaR values used to estimate the target in TD error during centralized training and the CVaR values are used as auxiliary local rewards to update the local distribution via Quantile Regression loss. Empirically, we show that our method significantly outperforms state-of-the-art methods on challenging StarCraft II tasks, demonstrating enhanced coordination and improved sample efficiency.

Xinrun Wang, Tarun Nair, Haoyang Li et al.

Dams impact downstream river dynamics through flow regulation and disruption of upstream-downstream linkages. However, current dam operation is far from satisfactory due to the inability to respond the complicated and uncertain dynamics of the upstream-downstream system and various usages of the reservoir. Even further, the unsatisfactory dam operation can cause floods in downstream areas. Therefore, we leverage reinforcement learning (RL) methods to compute efficient dam operation guidelines in this work. Specifically, we build offline simulators with real data and different mathematical models for the upstream inflow, i.e., generalized least square (GLS) and dynamic linear model (DLM), then use the simulator to train the state-of-the-art RL algorithms, including DDPG, TD3 and SAC. Experiments show that the simulator with DLM can efficiently model the inflow dynamics in the upstream and the dam operation policies trained by RL algorithms significantly outperform the human-generated policy.

Xu He, Bo An, Yanghua Li et al.

With the rise of online e-commerce platforms, more and more customers prefer to shop online. To sell more products, online platforms introduce various modules to recommend items with different properties such as huge discounts. A web page often consists of different independent modules. The ranking policies of these modules are decided by different teams and optimized individually without cooperation, which might result in competition between modules. Thus, the global policy of the whole page could be sub-optimal. In this paper, we propose a novel multi-agent cooperative reinforcement learning approach with the restriction that different modules cannot communicate. Our contributions are three-fold. Firstly, inspired by a solution concept in game theory named correlated equilibrium, we design a signal network to promote cooperation of all modules by generating signals (vectors) for different modules. Secondly, an entropy-regularized version of the signal network is proposed to coordinate agents' exploration of the optimal global policy. Furthermore, experiments based on real-world e-commerce data demonstrate that our algorithm obtains superior performance over baselines.

Runsheng Yu, Zhenyu Shi, Xinrun Wang et al.

Existing value-factorized based Multi-Agent deep Reinforce-ment Learning (MARL) approaches are well-performing invarious multi-agent cooperative environment under thecen-tralized training and decentralized execution(CTDE) scheme,where all agents are trained together by the centralized valuenetwork and each agent execute its policy independently. How-ever, an issue remains open: in the centralized training process,when the environment for the team is partially observable ornon-stationary, i.e., the observation and action informationof all the agents cannot represent the global states, existingmethods perform poorly and sample inefficiently. Regret Min-imization (RM) can be a promising approach as it performswell in partially observable and fully competitive settings.However, it tends to model others as opponents and thus can-not work well under the CTDE scheme. In this work, wepropose a novel team RM based Bayesian MARL with threekey contributions: (a) we design a novel RM method to traincooperative agents as a team and obtain a team regret-basedpolicy for that team; (b) we introduce a novel method to de-compose the team regret to generate the policy for each agentfor decentralized execution; (c) to further improve the perfor-mance, we leverage a differential particle filter (a SequentialMonte Carlo method) network to get an accurate estimation ofthe state for each agent. Experimental results on two-step ma-trix games (cooperative game) and battle games (large-scalemixed cooperative-competitive games) demonstrate that ouralgorithm significantly outperforms state-of-the-art methods.