En Zhang

VFEAgent is a novel multi-agent framework that automates the entire Finite Element Analysis (FEA) workflow, achieving high success rates in generating complete and physically valid simulations directly from multimodal inputs.

The paper introduces OmniVerifier-M1, a multimodal meta-verifier that uses symbolic outputs and decoupled reinforcement learning to provide robust, fine-grained verification and error localization for large multimodal models.

The paper introduces polynomial representations as a quantitative, distribution-aware metric for measuring model simplicity, demonstrating that the effective degree of this representation is a superior predictor of generalization compared to existing proxies.

ExpGraph is a model-agnostic framework that uses a self-evolving experience graph to enable LLM agents to reuse past successful strategies and failure lessons, significantly improving performance across diverse tasks.

ElasticMem introduces a novel framework that treats memory as an elastic latent resource, allowing LLM agents to adaptively manage and inject variable-budget memories for improved performance in long-term reasoning tasks.

GIRL-DETR introduces Gradient-Isolated Reinforcement Learning to enhance temporal localization in lightweight Video Moment Retrieval models, achieving high accuracy by decoupling feature representation from metric optimization.

The paper introduces a higher-order network framework to compare observed and simulated human mobility data, demonstrating that while synthetic data is promising, current simulation models have specific limitations regarding path-based movement patterns.

CRAFTQA introduces a novel adaptive, code-driven framework that significantly enhances complex structured data reasoning by dynamically generating custom code functions beyond predefined operations.

IstGPT introduces a novel LLM-based framework for real-time, fine-grained anomaly detection in complex industrial cyber-physical systems, achieving state-of-the-art performance across multiple benchmarks.

This paper shows that a stem-connected MiLAC can realize every beamformer on the complex Stiefel manifold and achieves the same sum-rate as a fully-connected MiLAC for multiuser downlink beamforming under certain conditions.

The Autonomous Rectified Flow framework is proposed to improve generative speech enhancement by eliminating explicit time-step conditioning and inferring denoising directions from spatial relationships.

This paper investigates how the extremality of fixed-volume spanning trees changes based on product-grid boundary factors in two and arbitrary dimensions.

This paper analyzes direction-of-arrival estimation using a tunable receive-side lossless reciprocal MiLAC combiner for antenna arrays and shows it can achieve the digital Cramér-Rao bound with fewer components than a digital receiver.

This paper proposes SkyChain Intelligence, a framework that integrates agentic AI, consortium blockchain, and Multi-Agent Deep Reinforcement Learning to optimize autonomy, security, and efficiency in Low-Altitude Economy ecosystems.

This paper proposes Learning to Allocate (L2A), an end-to-end framework for resource-adaptive inference in Large Language Models (LLMs) using budget-conditioned and input-aware gating networks.

This paper proposes a lightweight underwater image enhancement framework with two components: MBRConv-DCT for injecting frequency priors during training and FGDPA for fusing spatial and spectral cues.

This paper introduces MECoBench, a multimodal embodied cooperation benchmark, and explores the benefits and limitations of collaboration in multimodal large language models through extensive experiments.

This paper presents a user-adjustable, room-scale mesh structure for low-fidelity prototyping, equipped with resistive length sensors to transmit configuration data to a central computer for later reproduction in software.

The paper introduces TikStance, a multimodal and context-aware dataset for stance detection in political discussions on TikTok.

This paper introduces UniRank, an open benchmark for comparing and studying unified ranking models that combine sequential modeling and feature interaction.