arXiv subjects

Guangnan Ye

Publications and source records attributed to Guangnan Ye.

At least 19 recordsLinked to original sources

arXiv2026

From Correctness to Utility: Gain-Based Prefix Evaluation for LLM Reasoning

Reasoning prefixes shape the future trajectory of LLM problem solving, yet existing process reward models usually evaluate them through local step correctness. We argue that correctness is a useful but indirect proxy for the effect we ultimately care about: whether a prefix increases the probability of successful completion. We define this effect as prefix gain, the solve-rate improvement induced by conditioning lightweight student model group on a prefix, and use it to train a Prefix Utility Model (PUM) with a simple pairwise ranking objective. PUM learns outcome-grounded prefix utility and can score both complete trajectories and partial reasoning prefixes. Across Best-of-$N$ selection, beam search, and reinforcement learning on mathematical reasoning, PUM provides a strong prefix-level supervision signal, especially when candidate pools are large, search budgets increase, or rule-based rewards are sparse. We release all data, models, and code at https://zhiqix.github.io/pum-project-page.

cs.CL↗

arXiv2026

Mind the Tool Failures: Achieving Synergistic Tool Gains for Medical Agents

Medical AI agents increasingly use external tools for diagnosis, treatment recommendation, and evidence retrieval, yet most existing approaches assume that task-appropriate tools are reliable within their intended scope. This assumption is fragile in real clinical settings, where even relevant tools may fail on challenging instances and lead to unsafe downstream decisions. To address this issue, we study medical tool use under imperfect-tool settings to correct failure instances missed by individual tools. Instance-dependent failure patterns create a gap between the best fixed single tool and an ideal instance-wise selector, which we refer to as the Single-Oracle risk gap. The core challenge is that conventional task-level tool selection cannot realize this gap, as it is inherently bounded by the performance of the best single tool. Motivated by this observation, we therefore account for instance-level heterogeneity and formulate tool use as an instance-level selection problem. Particularly, we propose a GRPO-based reinforcement learning framework with rewards for probabilistic risk minimization and disagreement-aware synergy learning, which promotes instance-level correction of erroneous tool consensus. Furthermore, an entropy-guided sampling strategy is adopted to upweight high-disagreement instances, which provide stronger signals for learning instance-specific tool synergy. These two components complement each other in mitigating instance-level heterogeneity and improving tool synergy. Experiments on two tasks and seven medical benchmarks show that our method consistently achieves robust and stable improvements over a broad range of baselines, highlighting the importance of synergy-aware tool use for reliable medical agentic systems.

cs.AI↗

arXiv2026

TransXion: A High-Fidelity Graph Benchmark for Realistic Anti-Money Laundering

Money laundering poses severe risks to global financial systems, driving the widespread adoption of machine learning for transaction monitoring. However, progress remains stifled by the lack of realistic benchmarks. Existing transaction-graph datasets suffer from two pervasive limitations: (i) they provide sparse node-level semantics beyond anonymized identifiers, and (ii) they rely on template-driven anomaly injection, which biases benchmarks toward static structural motifs and yields overly optimistic assessments of model robustness. We propose TransXion, a benchmark ecosystem for Anti-Money Laundering (AML) research that integrates profile-aware simulation of normal activity with stochastic, non-template synthesis of illicit subgraphs.TransXion jointly models persistent entity profiles and conditional transaction behavior, enabling evaluation of "out-of-character" anomalies where observed activity contradicts an entity's socio-economic context. The resulting dataset comprises approximately 3 million transactions among 50,000 entities, each endowed with rich demographic and behavioral attributes. Empirical analyses show that TransXion reproduces key structural properties of payment networks, including heavy-tailed activity distributions and localized subgraph structure. Across a diverse array of detection models spanning multiple algorithmic paradigms, TransXion yields substantially lower detection performance than widely used benchmarks, demonstrating increased difficulty and realism. TransXion provides a more faithful testbed for developing context-aware and robust AML detection methods. The dataset and code are publicly available at https://github.com/chaos-max/TransXion.

cs.LG↗

arXiv2026

GAM-RAG: Gain-Adaptive Memory for Evolving Retrieval in Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) grounds large language models with external evidence, but many implementations rely on pre-built indices that remain static after construction. Related queries therefore repeat similar multi-hop traversal, increasing latency and compute. Motivated by schema-based learning in cognitive neuroscience, we propose GAM-RAG, a training-free framework that accumulates retrieval experience from recurring or related queries and updates retrieval memory over time. GAM-RAG builds a lightweight, relation-free hierarchical index whose links capture potential co-occurrence rather than fixed semantic relations. During inference, successful retrieval episodes provide sentence-level feedback, updating sentence memories so evidence useful for similar reasoning types becomes easier to activate later. To balance stability and adaptability under noisy feedback, we introduce an uncertainty-aware, Kalman-inspired gain rule that jointly updates memory states and perplexity-based uncertainty estimates. It applies fast updates for reliable novel signals and conservative refinement for stable or noisy memories. We provide a theoretical analysis of the update dynamics, and empirically show that GAM-RAG improves average performance by 3.95% over the strongest baseline and by 8.19% with 5-turn memory, while reducing inference cost by 61%. Our code and datasets are available at: https://anonymous.4open.science/r/GAM_RAG-2EF6.

cs.AI↗

arXiv2026

Rethinking the Role of Entropy in Optimizing Tool-Use Behaviors for Large Language Model Agents

Tool-using agents based on Large Language Models (LLMs) excel in tasks such as mathematical reasoning and multi-hop question answering. However, in long trajectories, agents often trigger excessive and low-quality tool calls, increasing latency and degrading inference performance, making managing tool-use behavior challenging. In this work, we conduct entropy-based pilot experiments and observe a strong positive correlation between entropy reduction and high-quality tool calls. Building on this finding, we propose using entropy reduction as a supervisory signal and design two reward strategies to address the differing needs of optimizing tool-use behavior. Sparse outcome rewards provide coarse, trajectory-level guidance to improve efficiency, while dense process rewards offer fine-grained supervision to enhance performance. Experiments across diverse domains show that both reward designs improve tool-use behavior: the former reduces tool calls by 72.07% compared to the average of baselines, while the latter improves performance by 22.27%. These results position entropy reduction as a key mechanism for enhancing tool-use behavior, enabling agents to be more adaptive in real-world applications.

cs.AI↗

arXiv2026

Behavioral Consistency Validation for LLM Agents: An Analysis of Trading-Style Switching through Stock-Market Simulation

Recent works have increasingly applied Large Language Models (LLMs) as agents in financial stock market simulations to test if micro-level behaviors aggregate into macro-level phenomena. However, a crucial question arises: Do LLM agents' behaviors align with real market participants? This alignment is key to the validity of simulation results. To explore this, we select a financial stock market scenario to test behavioral consistency. Investors are typically classified as fundamental or technical traders, but most simulations fix strategies at initialization, failing to reflect real-world trading dynamics. In this work, we assess whether agents' strategy switching aligns with financial theory, providing a framework for this evaluation. We operationalize four behavioral-finance drivers-loss aversion, herding, wealth differentiation, and price misalignment-as personality traits set via prompting and stored long-term. In year-long simulations, agents process daily price-volume data, trade under a designated style, and reassess their strategy every 10 trading days. We introduce four alignment metrics and use Mann-Whitney U tests to compare agents' style-switching behavior with financial theory. Our results show that recent LLMs' switching behavior is only partially consistent with behavioral-finance theories, highlighting the need for further refinement in aligning agent behavior with financial theory.

q-fin.TR↗

arXiv2025

Schr\"odinger's Navigator: Imagining an Ensemble of Futures for Zero-Shot Object Navigation

Zero-shot object navigation (ZSON) requires robots to find target objects in unseen environments without task-specific fine-tuning or pre-built maps, a key capability for general-purpose service robots. Yet methods that perform well in simulation often degrade in cluttered real-world scenes with severe occlusion and latent hazards, where large unseen regions make single-scene inference brittle and unsafe. We propose Schr\"odinger's Navigator, a belief-aware framework that reasons at inference time over multiple trajectory-conditioned imagined 3D futures. Given candidate paths, a trajectory-conditioned 3D world model predicts hypothetical observations and maintains a superposition of plausible scene realizations rather than committing to one map. An adaptive occluder-aware sampler directs imagination to uncertainty-critical regions, while a Future-Aware Value Map (FAVM) aggregates imagined futures for robust, proactive action selection. Experiments in simulation and on a physical Go2 quadruped show that Schr\"odinger's Navigator outperforms strong ZSON baselines, improving hidden-target discovery and risk-aware waypoint selection in occlusion-heavy navigation scenarios. These results highlight imagined 3D futures as a scalable and generalizable strategy for zero-shot navigation in uncertain real-world environments.

cs.RO↗

arXiv2025

Reconstruction as a Bridge for Event-Based Visual Question Answering

Integrating event cameras with Multimodal Large Language Models (MLLMs) promises general scene understanding in challenging visual conditions, yet requires navigating a trade-off between preserving the unique advantages of event data and ensuring compatibility with frame-based models. We address this challenge by using reconstruction as a bridge, proposing a straightforward Frame-based Reconstruction and Tokenization (FRT) method and designing an efficient Adaptive Reconstruction and Tokenization (ART) method that leverages event sparsity. For robust evaluation, we introduce EvQA, the first objective, real-world benchmark for event-based MLLMs, comprising 1,000 event-Q&A pairs from 22 public datasets. Our experiments demonstrate that our methods achieve state-of-the-art performance on EvQA, highlighting the significant potential of MLLMs in event-based vision.

cs.CV↗

arXiv2025

How AI Agents Follow the Herd of AI? Network Effects, History, and Machine Optimism

Understanding decision-making in multi-AI-agent frameworks is crucial for analyzing strategic interactions in network-effect-driven contexts. This study investigates how AI agents navigate network-effect games, where individual payoffs depend on peer participatio--a context underexplored in multi-agent systems despite its real-world prevalence. We introduce a novel workflow design using large language model (LLM)-based agents in repeated decision-making scenarios, systematically manipulating price trajectories (fixed, ascending, descending, random) and network-effect strength. Our key findings include: First, without historical data, agents fail to infer equilibrium. Second, ordered historical sequences (e.g., escalating prices) enable partial convergence under weak network effects but strong effects trigger persistent "AI optimism"--agents overestimate participation despite contradictory evidence. Third, randomized history disrupts convergence entirely, demonstrating that temporal coherence in data shapes LLMs' reasoning, unlike humans. These results highlight a paradigm shift: in AI-mediated systems, equilibrium outcomes depend not just on incentives, but on how history is curated, which is impossible for human.

cs.MA↗

arXiv2025

When Machines Meet Each Other: Network Effects and the Strategic Role of History in Multi-Agent AI

As artificial intelligence (AI) enters the agentic era, large language models (LLMs) are increasingly deployed as autonomous agents that interact with one another rather than operate in isolation. This shift raises a fundamental question: how do machine agents behave in interdependent environments where outcomes depend not only on their own choices but also on the coordinated expectations of peers? To address this question, we study LLM agents in a canonical network-effect game, where economic theory predicts convergence to a fulfilled expectation equilibrium (FEE). We design an experimental framework in which 50 heterogeneous GPT-5-based agents repeatedly interact under systematically varied network-effect strengths, price trajectories, and decision-history lengths. The results reveal that LLM agents systematically diverge from FEE: they underestimate participation at low prices, overestimate at high prices, and sustain persistent dispersion. Crucially, the way history is structured emerges as a design lever. Simple monotonic histories-where past outcomes follow a steady upward or downward trend-help stabilize coordination, whereas nonmonotonic histories amplify divergence and path dependence. Regression analyses at the individual level further show that price is the dominant driver of deviation, history moderates this effect, and network effects amplify contextual distortions. Together, these findings advance machine behavior research by providing the first systematic evidence on multi-agent AI systems under network effects and offer guidance for configuring such systems in practice.

econ.GN↗

arXiv2025

Limited Reference, Reliable Generation: A Two-Component Framework for Tabular Data Generation in Low-Data Regimes

Synthetic tabular data generation is increasingly essential in machine learning, supporting downstream applications when real-world, high-quality tabular data is insufficient. Existing tabular generation approaches, such as generative adversarial networks (GANs) and fine-tuned Large Language Models (LLMs), typically require sufficient reference data, limiting their effectiveness in domain-specific datasets with scarce records. While prompt-based LLMs offer flexibility without parameter tuning, they often generate distributionally drifted data with localized redundancy, leading to degradation in downstream task performance. To overcome these issues, we propose ReFine, a framework that (i) extracts symbolic if-then rules from interpretable models and embeds them into prompts to explicitly guide the generation process toward the domain-specific distribution, and (ii) applies dual-granularity filtering that mitigates over-sampling patterns while preserving rare but informative samples to reduce localized redundancy. Extensive experiments on diverse benchmarks demonstrate that ReFine provides robust downstream utility, achieving a top-tier average rank across datasets and data regimes, with an average relative improvement of 7.48% in extreme low-data regimes.

cs.LG↗

arXiv2025

Teach2Eval: An Indirect Evaluation Method for LLM by Judging How It Teaches

Recent progress in large language models (LLMs) has outpaced the development of effective evaluation methods. Traditional benchmarks rely on task-specific metrics and static datasets, which often suffer from fairness issues, limited scalability, and contamination risks. In this paper, we introduce Teach2Eval, an indirect evaluation framework inspired by the Feynman Technique. Instead of directly testing LLMs on predefined tasks, our method evaluates a model's multiple abilities to teach weaker student models to perform tasks effectively. By converting open-ended tasks into standardized multiple-choice questions (MCQs) through teacher-generated feedback, Teach2Eval enables scalable, automated, and multi-dimensional assessment. Our approach not only avoids data leakage and memorization but also captures a broad range of cognitive abilities that are orthogonal to current benchmarks. Experimental results across 26 leading LLMs show strong alignment with existing human and model-based dynamic rankings, while offering additional interpretability for training guidance.

cs.CL↗

arXiv2025

FedEGG: Federated Learning with Explicit Global Guidance

Federated Learning (FL) holds great potential for diverse applications owing to its privacy-preserving nature. However, its convergence is often challenged by non-IID data distributions, limiting its effectiveness in real-world deployments. Existing methods help address these challenges via optimization-based client constraints, adaptive client selection, or the use of pre-trained models or synthetic data. In this work, we reinterpret these approaches as all introducing an \emph{implicit guiding task} to regularize and steer client learning. Following this insight, we propose to introduce an \emph{explicit global guiding task} into the current FL framework to improve convergence and performance. To this end, we present \textbf{FedEGG}, a new FL algorithm that constructs a global guiding task using a well-defined, easy-to-converge learning task based on a public dataset and Large Language Models (LLMs). This approach effectively combines the strengths of federated (the original FL task) and centralized (the global guiding task) learning. We provide a theoretical analysis of FedEGG's convergence, examining the impact of data heterogeneity between the guiding and FL tasks and the guiding strength. Our analysis derives an upper bound for the optimal guiding strength, offering practical insights for implementation. Empirically, FedEGG demonstrates superior performance over state-of-the-art FL methods under both IID and non-IID settings, and further improves their performances when combined.

cs.LG↗

arXiv2025

Towards Million-Scale Adversarial Robustness Evaluation With Stronger Individual Attacks

As deep learning models are increasingly deployed in safety-critical applications, evaluating their vulnerabilities to adversarial perturbations is essential for ensuring their reliability and trustworthiness. Over the past decade, a large number of white-box adversarial robustness evaluation methods (i.e., attacks) have been proposed, ranging from single-step to multi-step methods and from individual to ensemble methods. Despite these advances, challenges remain in conducting meaningful and comprehensive robustness evaluations, particularly when it comes to large-scale testing and ensuring evaluations reflect real-world adversarial risks. In this work, we focus on image classification models and propose a novel individual attack method, Probability Margin Attack (PMA), which defines the adversarial margin in the probability space rather than the logits space. We analyze the relationship between PMA and existing cross-entropy or logits-margin-based attacks, and show that PMA can outperform the current state-of-the-art individual methods. Building on PMA, we propose two types of ensemble attacks that balance effectiveness and efficiency. Furthermore, we create a million-scale dataset, CC1M, derived from the existing CC3M dataset, and use it to conduct the first million-scale white-box adversarial robustness evaluation of adversarially-trained ImageNet models. Our findings provide valuable insights into the robustness gaps between individual versus ensemble attacks and small-scale versus million-scale evaluations.

cs.LG↗

arXiv2025

StageDesigner: Artistic Stage Generation for Scenography via Theater Scripts

In this work, we introduce StageDesigner, the first comprehensive framework for artistic stage generation using large language models combined with layout-controlled diffusion models. Given the professional requirements of stage scenography, StageDesigner simulates the workflows of seasoned artists to generate immersive 3D stage scenes. Specifically, our approach is divided into three primary modules: Script Analysis, which extracts thematic and spatial cues from input scripts; Foreground Generation, which constructs and arranges essential 3D objects; and Background Generation, which produces a harmonious background aligned with the narrative atmosphere and maintains spatial coherence by managing occlusions between foreground and background elements. Furthermore, we introduce the StagePro-V1 dataset, a dedicated dataset with 276 unique stage scenes spanning different historical styles and annotated with scripts, images, and detailed 3D layouts, specifically tailored for this task. Finally, evaluations using both standard and newly proposed metrics, along with extensive user studies, demonstrate the effectiveness of StageDesigner. Project can be found at: https://deadsmither5.github.io/2025/01/03/StageDesigner/

cs.CV↗

arXiv2025

RAGFormer: Learning Semantic Attributes and Topological Structure for Fraud Detection

Fraud detection remains a challenging task due to the complex and deceptive nature of fraudulent activities. Current approaches primarily concentrate on learning only one perspective of the graph: either the topological structure of the graph or the attributes of individual nodes. However, we conduct empirical studies to reveal that these two types of features, while nearly orthogonal, are each independently effective. As a result, previous methods can not fully capture the comprehensive characteristics of the fraud graph. To address this dilemma, we present a novel framework called Relation-Aware GNN with transFormer~(RAGFormer) which simultaneously embeds both semantic and topological features into a target node. The simple yet effective network consists of a semantic encoder, a topology encoder, and an attention fusion module. The semantic encoder utilizes Transformer to learn semantic features and node interactions across different relations. We introduce Relation-Aware GNN as the topology encoder to learn topological features and node interactions within each relation. These two complementary features are interleaved through an attention fusion module to support prediction by both orthogonal features. Extensive experiments on two popular public datasets demonstrate that RAGFormer achieves state-of-the-art performance. The significant improvement of RAGFormer in an industrial credit card fraud detection dataset further validates the applicability of our method in real-world business scenarios.

cs.LG↗

arXiv2024

ArtNVG: Content-Style Separated Artistic Neighboring-View Gaussian Stylization

As demand from the film and gaming industries for 3D scenes with target styles grows, the importance of advanced 3D stylization techniques increases. However, recent methods often struggle to maintain local consistency in color and texture throughout stylized scenes, which is essential for maintaining aesthetic coherence. To solve this problem, this paper introduces ArtNVG, an innovative 3D stylization framework that efficiently generates stylized 3D scenes by leveraging reference style images. Built on 3D Gaussian Splatting (3DGS), ArtNVG achieves rapid optimization and rendering while upholding high reconstruction quality. Our framework realizes high-quality 3D stylization by incorporating two pivotal techniques: Content-Style Separated Control and Attention-based Neighboring-View Alignment. Content-Style Separated Control uses the CSGO model and the Tile ControlNet to decouple the content and style control, reducing risks of information leakage. Concurrently, Attention-based Neighboring-View Alignment ensures consistency of local colors and textures across neighboring views, significantly improving visual quality. Extensive experiments validate that ArtNVG surpasses existing methods, delivering superior results in content preservation, style alignment, and local consistency.

cs.CV↗

arXiv2024

DogLayout: Denoising Diffusion GAN for Discrete and Continuous Layout Generation

Layout Generation aims to synthesize plausible arrangements from given elements. Currently, the predominant methods in layout generation are Generative Adversarial Networks (GANs) and diffusion models, each presenting its own set of challenges. GANs typically struggle with handling discrete data due to their requirement for differentiable generated samples and have historically circumvented the direct generation of discrete labels by treating them as fixed conditions. Conversely, diffusion-based models, despite achieving state-of-the-art performance across several metrics, require extensive sampling steps which lead to significant time costs. To address these limitations, we propose \textbf{DogLayout} (\textbf{D}en\textbf{o}ising Diffusion \textbf{G}AN \textbf{Layout} model), which integrates a diffusion process into GANs to enable the generation of discrete label data and significantly reduce diffusion's sampling time. Experiments demonstrate that DogLayout considerably reduces sampling costs by up to 175 times and cuts overlap from 16.43 to 9.59 compared to existing diffusion models, while also surpassing GAN based and other layout methods. Code is available at https://github.com/deadsmither5/DogLayout.

cs.CV↗