arXiv subjects

Zhuo Chen

Publications and source records attributed to Zhuo Chen.

At least 145 records · Page 8Linked to original sources

arXiv2025

DiSTAR: Diffusion over a Scalable Token Autoregressive Representation for Speech Generation

Recent attempts to interleave autoregressive (AR) sketchers with diffusion-based refiners over continuous speech representations have shown promise, but they remain brittle under distribution shift and offer limited levers for controllability. We introduce DISTAR, a zero-shot text-to-speech framework that operates entirely in a discrete residual vector quantization (RVQ) code space and tightly couples an AR language model with a masked diffusion model, without forced alignment or a duration predictor. Concretely, DISTAR drafts block-level RVQ tokens with an AR language model and then performs parallel masked-diffusion infilling conditioned on the draft to complete the next block, yielding long-form synthesis with blockwise parallelism while mitigating classic AR exposure bias. The discrete code space affords explicit control at inference: DISTAR produces high-quality audio under both greedy and sample-based decoding using classifier-free guidance, supports trade-offs between robustness and diversity, and enables variable bit-rate and controllable computation via RVQ layer pruning at test time. Extensive experiments and ablations demonstrate that DISTAR surpasses state-of-the-art zero-shot TTS systems in robustness, naturalness, and speaker/style consistency, while maintaining rich output diversity. Audio samples are provided on https://anonymous.4open.science/w/DiSTAR_demo.

eess.AS↗

arXiv2025

Heptapod: Language Modeling on Visual Signals

We introduce Heptapod, an image autoregressive model that adheres to the foundational principles of language modeling. Heptapod employs \textbf{causal attention}, \textbf{eliminates reliance on CFG}, and \textbf{eschews the trend of semantic tokenizers}. Our key innovation is \textit{next 2D distribution prediction}: a causal Transformer with reconstruction-focused visual tokenizer, learns to predict the distribution over the entire 2D spatial grid of images at each timestep. This learning objective unifies the sequential modeling of autoregressive framework with the holistic self-supervised learning of masked autoencoding, enabling the model to capture comprehensive image semantics via generative training. On the ImageNet generation benchmark, Heptapod achieves an FID of $2.70$, significantly outperforming previous causal autoregressive approaches. We hope our work inspires a principled rethinking of language modeling on visual signals and beyond.

cs.CV↗

arXiv2025

Training Tactile Sensors to Learn Force Sensing from Each Other

Humans achieve stable and dexterous object manipulation by coordinating grasp forces across multiple fingers and palms, facilitated by a unified tactile memory system in the somatosensory cortex. This system encodes and stores tactile experiences across skin regions, enabling the flexible reuse and transfer of touch information. Inspired by this biological capability, we present GenForce, the first framework that enables transferable force sensing across tactile sensors in robotic hands. GenForce unifies tactile signals into shared marker representations, analogous to cortical sensory encoding, allowing force prediction models trained on one sensor to be transferred to others without the need for exhaustive force data collection. We demonstrate that GenForce generalizes across both homogeneous sensors with varying configurations and heterogeneous sensors with distinct sensing modalities and material properties. This transferable force sensing is also demonstrated with high performance in robot force control including daily object grasping, slip detection and avoidance. Our results highlight a scalable paradigm for cross-sensor robotic tactile learning, offering new pathways toward adaptable and tactile memory-driven manipulation in unstructured environments.

cs.RO↗

arXiv2025

SeamCrafter: Enhancing Mesh Seam Generation for Artist UV Unwrapping via Reinforcement Learning

Mesh seams play a pivotal role in partitioning 3D surfaces for UV parametrization and texture mapping. Poorly placed seams often result in severe UV distortion or excessive fragmentation, thereby hindering texture synthesis and disrupting artist workflows. Existing methods frequently trade one failure mode for another-producing either high distortion or many scattered islands. To address this, we introduce SeamCrafter, an autoregressive GPT-style seam generator conditioned on point cloud inputs. SeamCrafter employs a dual-branch point-cloud encoder that disentangles and captures complementary topological and geometric cues during pretraining. To further enhance seam quality, we fine-tune the model using Direct Preference Optimization (DPO) on a preference dataset derived from a novel seam-evaluation framework. This framework assesses seams primarily by UV distortion and fragmentation, and provides pairwise preference labels to guide optimization. Extensive experiments demonstrate that SeamCrafter produces seams with substantially lower distortion and fragmentation than prior approaches, while preserving topological consistency and visual fidelity.

cs.GR↗

arXiv2025

P3-SAM: Native 3D Part Segmentation

Segmenting 3D assets into their constituent parts is crucial for enhancing 3D understanding, facilitating model reuse, and supporting various applications such as part generation. However, current methods face limitations such as poor robustness when dealing with complex objects and cannot fully automate the process. In this paper, we propose a native 3D point-promptable part segmentation model termed P$^3$-SAM, designed to fully automate the segmentation of any 3D objects into components. Inspired by SAM, P$^3$-SAM consists of a feature extractor, multiple segmentation heads, and an IoU predictor, enabling interactive segmentation for users. We also propose an algorithm to automatically select and merge masks predicted by our model for part instance segmentation. Our model is trained on a newly built dataset containing nearly 3.7 million models with reasonable segmentation labels. Comparisons show that our method achieves precise segmentation results and strong robustness on any complex objects, attaining state-of-the-art performance. Our project page is available at https://murcherful.github.io/P3-SAM/.

cs.CV↗

arXiv2025

ArtUV: Artist-style UV Unwrapping

UV unwrapping is an essential task in computer graphics, enabling various visual editing operations in rendering pipelines. However, existing UV unwrapping methods struggle with time-consuming, fragmentation, lack of semanticity, and irregular UV islands, limiting their practical use. An artist-style UV map must not only satisfy fundamental criteria, such as overlap-free mapping and minimal distortion, but also uphold higher-level standards, including clean boundaries, efficient space utilization, and semantic coherence. We introduce ArtUV, a fully automated, end-to-end method for generating artist-style UV unwrapping. We simulates the professional UV mapping process by dividing it into two stages: surface seam prediction and artist-style UV parameterization. In the seam prediction stage, SeamGPT is used to generate semantically meaningful cutting seams. Then, in the parameterization stage, a rough UV obtained from an optimization-based method, along with the mesh, is fed into an Auto-Encoder, which refines it into an artist-style UV map. Our method ensures semantic consistency and preserves topological structure, making the UV map ready for 2D editing. We evaluate ArtUV across multiple benchmarks and show that it serves as a versatile solution, functioning seamlessly as either a plug-in for professional rendering tools or as a standalone system for rapid, high-quality UV generation.

cs.GR↗

arXiv2025

X-Part: high fidelity and structure coherent shape decomposition

Generating 3D shapes at part level is pivotal for downstream applications such as mesh retopology, UV mapping, and 3D printing. However, existing part-based generation methods often lack sufficient controllability and suffer from poor semantically meaningful decomposition. To this end, we introduce X-Part, a controllable generative model designed to decompose a holistic 3D object into semantically meaningful and structurally coherent parts with high geometric fidelity. X-Part exploits the bounding box as prompts for the part generation and injects point-wise semantic features for meaningful decomposition. Furthermore, we design an editable pipeline for interactive part generation. Extensive experimental results show that X-Part achieves state-of-the-art performance in part-level shape generation. This work establishes a new paradigm for creating production-ready, editable, and structurally sound 3D assets. Codes will be released for public research.

cs.GR↗

arXiv2025

Direct Preference Optimization for Speech Autoregressive Diffusion Models

Autoregressive diffusion models (ARDMs) have recently been applied to speech generation, achieving state-of-the-art (SOTA) performance in zero-shot text-to-speech. By autoregressively generating continuous speech tokens with next-token diffusion, these models offer a promising alternative to next-token prediction, avoiding the technical complexities associated with discrete speech tokenization. As a relatively new paradigm, research on reinforcement learning (RL)-based fine-tuning of speech ARDMs remains limited. In this paper, we propose Autoregressive Diffusion-Direct Preference Optimization (ARDM-DPO) to advance this research. By fine-tuning the recently proposed zero-shot text-to-speech model DiTAR with DPO, we achieve significant improvements in terms of speech expressiveness and robustness for long texts.

eess.AS↗

arXiv2025

KBM: Delineating Knowledge Boundary for Adaptive Retrieval in Large Language Models

Large Language Models (LLMs) often struggle with dynamically changing knowledge and handling unknown static information. Retrieval-Augmented Generation (RAG) is employed to tackle these challenges and has a significant impact on improving LLM performance. In fact, we find that not all questions need to trigger RAG. By retrieving parts of knowledge unknown to the LLM and allowing the LLM to answer the rest, we can effectively reduce both time and computational costs. In our work, we propose a Knowledge Boundary Model (KBM) to express the known/unknown of a given question, and to determine whether a RAG needs to be triggered. Experiments conducted on 11 English and Chinese datasets illustrate that the KBM effectively delineates the knowledge boundary, significantly decreasing the proportion of retrievals required for optimal end-to-end performance. Furthermore, we evaluate the effectiveness of KBM in three complex scenarios: dynamic knowledge, long-tail static knowledge, and multi-hop problems, as well as its functionality as an external LLM plug-in.

cs.CL↗

arXiv2025

FlowDrive: Energy Flow Field for End-to-End Autonomous Driving

Recent advances in end-to-end autonomous driving leverage multi-view images to construct BEV representations for motion planning. In motion planning, autonomous vehicles need considering both hard constraints imposed by geometrically occupied obstacles (e.g., vehicles, pedestrians) and soft, rule-based semantics with no explicit geometry (e.g., lane boundaries, traffic priors). However, existing end-to-end frameworks typically rely on BEV features learned in an implicit manner, lacking explicit modeling of risk and guidance priors for safe and interpretable planning. To address this, we propose FlowDrive, a novel framework that introduces physically interpretable energy-based flow fields-including risk potential and lane attraction fields-to encode semantic priors and safety cues into the BEV space. These flow-aware features enable adaptive refinement of anchor trajectories and serve as interpretable guidance for trajectory generation. Moreover, FlowDrive decouples motion intent prediction from trajectory denoising via a conditional diffusion planner with feature-level gating, alleviating task interference and enhancing multimodal diversity. Experiments on the NAVSIM v2 benchmark demonstrate that FlowDrive achieves state-of-the-art performance with an EPDMS of 86.3, surpassing prior baselines in both safety and planning quality. The project is available at https://astrixdrive.github.io/FlowDrive.github.io/.

cs.RO↗

arXiv2025

Hunyuan3D Studio: End-to-End AI Pipeline for Game-Ready 3D Asset Generation

The creation of high-quality 3D assets, a cornerstone of modern game development, has long been characterized by labor-intensive and specialized workflows. This paper presents Hunyuan3D Studio, an end-to-end AI-powered content creation platform designed to revolutionize the game production pipeline by automating and streamlining the generation of game-ready 3D assets. At its core, Hunyuan3D Studio integrates a suite of advanced neural modules (such as Part-level 3D Generation, Polygon Generation, Semantic UV, etc.) into a cohesive and user-friendly system. This unified framework allows for the rapid transformation of a single concept image or textual description into a fully-realized, production-quality 3D model complete with optimized geometry and high-fidelity PBR textures. We demonstrate that assets generated by Hunyuan3D Studio are not only visually compelling but also adhere to the stringent technical requirements of contemporary game engines, significantly reducing iteration time and lowering the barrier to entry for 3D content creation. By providing a seamless bridge from creative intent to technical asset, Hunyuan3D Studio represents a significant leap forward for AI-assisted workflows in game development and interactive media.

cs.CV↗

arXiv2025

Scaling Agents via Continual Pre-training

Large language models (LLMs) have evolved into agentic systems capable of autonomous tool use and multi-step reasoning for complex problem-solving. However, post-training approaches building upon general-purpose foundation models consistently underperform in agentic tasks, particularly in open-source implementations. We identify the root cause: the absence of robust agentic foundation models forces models during post-training to simultaneously learn diverse agentic behaviors while aligning them to expert demonstrations, thereby creating fundamental optimization tensions. To this end, we are the first to propose incorporating Agentic Continual Pre-training (Agentic CPT) into the deep research agents training pipeline to build powerful agentic foundational models. Based on this approach, we develop a deep research agent model named AgentFounder. We evaluate our AgentFounder-30B on 10 benchmarks and achieve state-of-the-art performance while retains strong tool-use ability, notably 39.9% on BrowseComp-en, 43.3% on BrowseComp-zh, and 31.5% Pass@1 on HLE.

cs.CL↗

arXiv2025

RoboBrain 2.0 Technical Report

We introduce RoboBrain 2.0, our latest generation of embodied vision-language foundation models, designed to unify perception, reasoning, and planning for complex embodied tasks in physical environments. It comes in two variants: a lightweight 7B model and a full-scale 32B model, featuring a heterogeneous architecture with a vision encoder and a language model. Despite its compact size, RoboBrain 2.0 achieves strong performance across a wide spectrum of embodied reasoning tasks. On both spatial and temporal benchmarks, the 32B variant achieves leading results, surpassing prior open-source and proprietary models. In particular, it supports key real-world embodied AI capabilities, including spatial understanding (e.g., affordance prediction, spatial referring, trajectory forecasting) and temporal decision-making (e.g., closed-loop interaction, multi-agent long-horizon planning, and scene graph updating). This report details the model architecture, data construction, multi-stage training strategies, infrastructure and practical applications. We hope RoboBrain 2.0 advances embodied AI research and serves as a practical step toward building generalist embodied agents. The code, checkpoint and benchmark are available at https://superrobobrain.github.io.

cs.RO↗

arXiv2025

Detecting Knowledge Boundary of Vision Large Language Models by Sampling-Based Inference

Despite the advancements made in Vision Large Language Models (VLLMs), like text Large Language Models (LLMs), they have limitations in addressing questions that require real-time information or are knowledge-intensive. Indiscriminately adopting Retrieval Augmented Generation (RAG) techniques is an effective yet expensive way to enable models to answer queries beyond their knowledge scopes. To mitigate the dependence on retrieval and simultaneously maintain, or even improve, the performance benefits provided by retrieval, we propose a method to detect the knowledge boundary of VLLMs, allowing for more efficient use of techniques like RAG. Specifically, we propose a method with two variants that fine-tune a VLLM on an automatically constructed dataset for boundary identification. Experimental results on various types of Visual Question Answering datasets show that our method successfully depicts a VLLM's knowledge boundary, based on which we are able to reduce indiscriminate retrieval while maintaining or improving the performance. In addition, we show that the knowledge boundary identified by our method for one VLLM can be used as a surrogate boundary for other VLLMs. Code will be released at https://github.com/Chord-Chen-30/VLLM-KnowledgeBoundary

cs.CL↗

arXiv2025

Advances in Speech Separation: Techniques, Challenges, and Future Trends

The field of speech separation, addressing the "cocktail party problem", has seen revolutionary advances with DNNs. Speech separation enhances clarity in complex acoustic environments and serves as crucial pre-processing for speech recognition and speaker recognition. However, current literature focuses narrowly on specific architectures or isolated approaches, creating fragmented understanding. This survey addresses this gap by providing systematic examination of DNN-based speech separation techniques. Our work differentiates itself through: (I) Comprehensive perspective: We systematically investigate learning paradigms, separation scenarios with known/unknown speakers, comparative analysis of supervised/self-supervised/unsupervised frameworks, and architectural components from encoders to estimation strategies. (II) Timeliness: Coverage of cutting-edge developments ensures access to current innovations and benchmarks. (III) Unique insights: Beyond summarization, we evaluate technological trajectories, identify emerging patterns, and highlight promising directions including domain-robust frameworks, efficient architectures, multimodal integration, and novel self-supervised paradigms. (IV) Fair evaluation: We provide quantitative evaluations on standard datasets, revealing true capabilities and limitations of different methods. This comprehensive survey serves as an accessible reference for experienced researchers and newcomers navigating speech separation's complex landscape.

cs.SD↗

arXiv2025

Brace $B_{\infty}$ algebras associated with Hopf algebroids

We apply the operadic modeling of brace $B_{\infty}$ algebras, as developed by Gerstenhaber and Voronov, to the context of Hopf algebroids in the sense of Xu. Specifically, we construct a strict $B_{\infty}$ isomorphism between the type I and type II twisted brace $B_{\infty}$ algebras arising from any twistor of a Hopf algebroid. As an application of this framework, we examine two specific brace $B_{\infty}$ algebras derived from Lie algebra pairs, and reveal previously obscured relationships between them. One of these is the dg Lie algebra governing deformations of algebraic dynamical twists, while the other arises from the quantum groupoid comprised of particular invariant differential operators.

math.QA↗

arXiv2025

Nonautonomous Dynamical Systems II: Variational Principles

Let $\boldsymbol{X}=\{X_k\}_{k=0}^\infty$ be a sequence of compact metric spaces $X_{k}$ and $\boldsymbol{T}=\{T_k\}_{k=0}^\infty$ a sequence of continuous mappings $T_{k}: X_{k} \to X_{k+1}$. The pair $(\boldsymbol{X},\boldsymbol{T})$ is called a nonautonomous dynamical system. In this paper, we study measure-theoretic entropies and pressures, Bowen and packing topological entropies and pressures on $(\boldsymbol{X},\boldsymbol{T})$, and we prove that they are invariant under equiconjugacies of nonautonomous dynamical systems. By establishing Billingsley type theorems for Bowen and packing topological pressures, we obtain their variational principles, that is, given a non-empty compact subset $K \subset X_{0}$ and an equicontinuous sequence $\boldsymbol{f}= \{f_k\}_{k=0}^\infty$ of functions $f_k : X_k\to \mathbb{R}$, we have that $$ P^{\mathrm{B}}(\boldsymbol{T},\boldsymbol{f},K)=\sup\{\underline{P}_μ(\boldsymbol{T},\boldsymbol{f}): μ\in M(X_{0}), μ(K)=1\}, $$ and for $\|\boldsymbol{f}\|<+\infty$ and $P^{\mathrm{P}}(\boldsymbol{T},\boldsymbol{f},K)>\|\boldsymbol{f}\|$, $$ P^{\mathrm{P}}(\boldsymbol{T},\boldsymbol{f},K)=\sup\{\overline{P}_μ(\boldsymbol{T},\boldsymbol{f}): μ\in M(X_{0}), μ(K)=1\}, $$ where $\underline{P}_μ $ and $\overline{P}_μ $, $P^{\mathrm{B}}$ and $P^{\mathrm{P}}$ denote measure-theoretic lower and upper pressures, Bowen and packing topological pressure, respectively. The Billingsley type theorems and variational principles for Bowen and packing topological entropies are direct consequences of the ones for Bowen and packing topological pressures.

math.DS↗

arXiv2025

Nonautonomous Dynamical Systems I: Topological Pressures and Entropies

Let $\boldsymbol{X}=\{X_{k}\}_{k=0}^{\infty}$ be a sequence of compact metric spaces $X_{k}$ and $\boldsymbol{T}=\{T_{k}\}_{k=0}^{\infty}$ a sequence of continuous mappings $T_{k}:X_{k} \to X_{k+1}$. The pair $(\boldsymbol{X},\boldsymbol{T})$ is called a nonautonomous dynamical system. Our main object is to study the variational principles of topological pressures and entropies on nonautonomous dynamical systems. In this paper, we introduce a variety of topological pressures ($\underline{Q}$,$\overline{Q}$,$\underline{P}$,$\overline{P}$,$P^{\mathrm{B}}$ and $P^{\mathrm{P}}$) for potentials $\boldsymbol{f}=\{f_{k} \in C(X_{k},\mathbb{R})\}_{k=0}^{\infty}$ on subsets $Z \subset X_{0}$ analogous to fractal dimensions, and we provide various key properties which are crucial for the study of the variational principles on nonautonomous dynamical systems. Especially, we obtain the power rules and product rules of these pressures, and we also show they are invariants under equiconjugacies of nonautonomous dynamical systems and equicontinuity on $\boldsymbol{f}$. From a fractal dimension point of view, these pressures are kinds of 'dimensions' describing the nonautonomous dynamical systems, and we obtain various properties of pressures analogous to fractal dimensions.

math.DS↗