arXiv subjects

Yinuo Huang

Publications and source records attributed to Yinuo Huang.

5 recordsLinked to original sources

arXiv2025

Single Image to High-Quality 3D Object via Latent Features

3D assets are essential in the digital age. While automatic 3D generation, such as image-to-3d, has made significant strides in recent years, it often struggles to achieve fast, detailed, and high-fidelity generation simultaneously. In this work, we introduce LatentDreamer, a novel framework for generating 3D objects from single images. The key to our approach is a pre-trained variational autoencoder that maps 3D geometries to latent features, which greatly reducing the difficulty of 3D generation. Starting from latent features, the pipeline of LatentDreamer generates coarse geometries, refined geometries, and realistic textures sequentially. The 3D objects generated by LatentDreamer exhibit high fidelity to the input images, and the entire generation process can be completed within a short time (typically in 70 seconds). Extensive experiments show that with only a small amount of training, LatentDreamer demonstrates competitive performance compared to contemporary approachs.

cs.CV↗

arXiv2025

Blind MIMO Semantic Communication via Parallel Variational Diffusion: A Completely Pilot-Free Approach

In this paper, we propose a novel blind multi-input multi-output (MIMO) semantic communication (SC) framework named Blind-MIMOSC that consists of a deep joint source-channel coding (DJSCC) transmitter and a diffusion-based blind receiver. The DJSCC transmitter aims to compress and map the source data into the transmitted signal by exploiting the structural characteristics of the source data, while the diffusion-based blind receiver employs a parallel variational diffusion (PVD) model to simultaneously recover the channel and the source data from the received signal without using any pilots. The PVD model leverages two pre-trained score networks to characterize the prior information of the channel and the source data, operating in a plug-and-play manner during inference. This design allows only the affected network to be retrained when channel conditions or source datasets change, avoiding the complicated full-network retraining required by end-to-end methods. This work presents the first fully pilot-free solution for joint channel estimation and source recovery in block-fading MIMO systems. Extensive experiments show that Blind-MIMOSC with PVD achieves superior channel and source recovery accuracy compared to state-of-the-art approaches, with drastically reduced channel bandwidth ratio.

eess.SP↗

arXiv2025

Why Tree-Style Branching Matters for Thought Advantage Estimation in GRPO

Group Relative Policy Optimization (GRPO) trains Chain-of-Thought reasoning with verifiable rewards, but estimating thought-level advantages without value functions often suffers from high variance. Although tree-style branching is used in practice to reduce variance, it lacks a theoretical explanation of why it works and whether it is important or potentially necessary. We study thought-level advantage estimation in GRPO from a variance perspective under a minimal tree-style setting where multiple continuations are sampled for each thought. Using the multivariate delta method, we reveal a sampling-dimension asymmetry. Increasing sampled thoughts ($K$) leaves a strictly positive estimation-variance floor, whereas increasing continuations per thought ($M$) drives the leading-order estimation variance to zero at rate $1/M$. This implies that, within the fixed-temperature GRPO-style estimator without value models studied here, accurate thought-level advantage estimation cannot be achieved by scaling thought sampling alone, making continuation-level branching a principled and potentially necessary mechanism rather than a heuristic. Experiments further provide empirical evidence for its effectiveness and potential necessity, demonstrating improved optimization stability, training efficiency, and final performance not only in math but also across vision domains and under different model architectures and sizes.

cs.CL↗

arXiv2025

AGI-Driven Generative Semantic Communications: Principles and Practices

Semantic communications leverage artificial intelligence (AI) technologies to extract semantic information for efficient data delivery, thereby significantly reducing communication cost. With the evolution towards artificial general intelligence (AGI), the increasing demands for AGI services pose new challenges to semantic communications. In this context, an AGI application is typically defined on a general-sense task, covering a broad, even unforeseen, set of objectives, as well as driven by the need for a human-friendly interface in forms (e.g., videos, images, or text) easily understood by human users.In response, we introduce an AGI-driven communication paradigm for supporting AGI applications, called generative semantic communication (GSC). We first describe the basic concept of GSC and its difference from existing semantic communications, and then introduce a general framework of GSC based on advanced AI technologies including foundation models and generative models. Two case studies are presented to verify the advantages of GSC. Finally, open challenges and new research directions are discussed to stimulate this line of research and pave the way for practical applications.

cs.AI↗

arXiv2024

Personalized Route Recommendation Based on User Habits for Vehicle Navigation

Navigation route recommendation is one of the important functions of intelligent transportation. However, users frequently deviate from recommended routes for various reasons, with personalization being a key problem in the field of research. This paper introduces a personalized route recommendation method based on user historical navigation data. First, we formulate route sorting as a pointwise problem based on a large set of pertinent features. Second, we construct route features and user profiles to establish a comprehensive feature dataset. Furthermore, we propose a Deep-Cross-Recurrent (DCR) learning model aimed at learning route sorting scores and offering customized route recommendations. This approach effectively captures recommended navigation routes and user preferences by integrating DCN-v2 and LSTM. In offline evaluations, our method compared with the minimum ETA (estimated time of arrival), LightGBM, and DCN-v2 indicated 8.72%, 2.19%, and 0.9% reduction in the mean inconsistency rate respectively, demonstrating significant improvements in recommendation accuracy.

cs.RO↗