arXiv subjects

Hehe Fan

Publications and source records attributed to Hehe Fan.

At least 55 records · Page 3Linked to original sources

arXiv2025

ITS3D: Inference-Time Scaling for Text-Guided 3D Diffusion Models

We explore inference-time scaling in text-guided 3D diffusion models to enhance generative quality without additional training. To this end, we introduce ITS3D, a framework that formulates the task as an optimization problem to identify the most effective Gaussian noise input. The framework is driven by a verifier-guided search algorithm, where the search algorithm iteratively refines noise candidates based on verifier feedback. To address the inherent challenges of 3D generation, we introduce three techniques for improved stability, efficiency, and exploration capability. 1) Gaussian normalization is applied to stabilize the search process. It corrects distribution shifts when noise candidates deviate from a standard Gaussian distribution during iterative updates. 2) The high-dimensional nature of the 3D search space increases computational complexity. To mitigate this, a singular value decomposition-based compression technique is employed to reduce dimensionality while preserving effective search directions. 3) To further prevent convergence to suboptimal local minima, a singular space reset mechanism dynamically updates the search space based on diversity measures. Extensive experiments demonstrate that ITS3D enhances text-to-3D generation quality, which shows the potential of computationally efficient search methods in generative processes. The source code is available at https://github.com/ZhenglinZhou/ITS3D.

cs.CV↗

arXiv2025

SciEducator: Scientific Video Understanding and Educating via Deming-Cycle Multi-Agent System

Recent advancements in multimodal large language models (MLLMs) and video agent systems have significantly improved general video understanding. However, when applied to scientific video understanding and educating, a domain that demands external professional knowledge integration and rigorous step-wise reasoning, existing approaches often struggle. To bridge this gap, we propose SciEducator, the first iterative self-evolving multi-agent system for scientific video comprehension and education. Rooted in the classical Deming Cycle from management science, our design reformulates its Plan-Do-Study-Act philosophy into a self-evolving reasoning and feedback mechanism, which facilitates the interpretation of intricate scientific activities in videos. Moreover, SciEducator can produce multimodal educational content tailored to specific scientific processes, including textual instructions, visual guides, audio narrations, and interactive references. To support evaluation, we construct SciVBench, a benchmark consisting of 500 expert-verified and literature-grounded science QA pairs across five categories, covering physical, chemical, and everyday phenomena. Extensive experiments demonstrate that SciEducator substantially outperforms leading closed-source MLLMs (e.g., Gemini, GPT-4o) and state-of-the-art video agents on the benchmark, establishing a new paradigm for the community.

cs.CV↗

arXiv2025

Endowing GPT-4 with a Humanoid Body: Building the Bridge Between Off-the-Shelf VLMs and the Physical World

Humanoid agents often struggle to handle flexible and diverse interactions in open environments. A common solution is to collect massive datasets to train a highly capable model, but this approach can be prohibitively expensive. In this paper, we explore an alternative solution: empowering off-the-shelf Vision-Language Models (VLMs, such as GPT-4) to control humanoid agents, thereby leveraging their strong open-world generalization to mitigate the need for extensive data collection. To this end, we present \textbf{BiBo} (\textbf{B}uilding humano\textbf{I}d agent \textbf{B}y \textbf{O}ff-the-shelf VLMs). It consists of two key components: (1) an \textbf{embodied instruction compiler}, which enables the VLM to perceive the environment and precisely translate high-level user instructions (e.g., {\small\itshape ``have a rest''}) into low-level primitive commands with control parameters (e.g., {\small\itshape ``sit casually, location: (1, 2), facing: 90$^\circ$''}); and (2) a diffusion-based \textbf{motion executor}, which generates human-like motions from these commands, while dynamically adapting to physical feedback from the environment. In this way, BiBo is capable of handling not only basic interactions but also diverse and complex motions. Experiments demonstrate that BiBo achieves an interaction task success rate of 90.2\% in open environments, and improves the precision of text-guided motion execution by 16.3\% over prior methods. The code will be made publicly available.

cs.RO↗

arXiv2025

InfiniDreamer: Arbitrarily Long Human Motion Generation via Segment Score Distillation

We present InfiniDreamer, a novel framework for arbitrarily long human motion generation. InfiniDreamer addresses the limitations of current motion generation methods, which are typically restricted to short sequences due to the lack of long motion training data. To achieve this, we first generate sub-motions corresponding to each textual description and then assemble them into a coarse, extended sequence using randomly initialized transition segments. We then introduce an optimization-based method called Segment Score Distillation (SSD) to refine the entire long motion sequence. SSD is designed to utilize an existing motion prior, which is trained only on short clips, in a training-free manner. Specifically, SSD iteratively refines overlapping short segments sampled from the coarsely extended long motion sequence, progressively aligning them with the pre-trained motion diffusion prior. This process ensures local coherence within each segment, while the refined transitions between segments maintain global consistency across the entire sequence. Extensive qualitative and quantitative experiments validate the superiority of our framework, showcasing its ability to generate coherent, contextually aware motion sequences of arbitrary length.

cs.CV↗

arXiv2025

Translution: Unifying Self-attention and Convolution for Adaptive and Relative Modeling

When modeling a given type of data, we consider it to involve two key aspects: 1) identifying relevant elements (e.g., image pixels or textual words) to a central element, as in a convolutional receptive field, or to a query element, as in self-attention, and 2) encoding these tokens effectively. Self-attention can adaptively identify these elements but relies on absolute positional embedding for structural representation learning. In contrast, convolution encodes elements in a relative manner, yet their fixed kernel size limits their ability to adaptively select the relevant elements. In this paper, we introduce Translution, an operation that unifies the adaptive identification capability of self-attention and the relative encoding advantage of convolution. However, this integration leads to a substantial increase in the number of parameters, exceeding most currently available computational resources. Therefore, we propose a lightweight variant of Translution, named α-Translution. Experiments on computer vision and natural language processing tasks show that Translution (including α-Translution) achieves superior accuracy compared to self-attention. The code is available at https://github.com/hehefan/Translution.

cs.LG↗

arXiv2025

Text-Scene: A Scene-to-Language Parsing Framework for 3D Scene Understanding

Enabling agents to understand and interact with complex 3D scenes is a fundamental challenge for embodied artificial intelligence systems. While Multimodal Large Language Models (MLLMs) have achieved significant progress in 2D image understanding, extending such capabilities to 3D scenes remains difficult: 1) 3D environment involves richer concepts such as spatial relationships, affordances, physics, layout, and so on, 2) the absence of large-scale 3D vision-language datasets has posed a significant obstacle. In this paper, we introduce Text-Scene, a framework that automatically parses 3D scenes into textual descriptions for scene understanding. Given a 3D scene, our model identifies object attributes and spatial relationships, and then generates a coherent summary of the whole scene, bridging the gap between 3D observation and language without requiring human-in-the-loop intervention. By leveraging both geometric analysis and MLLMs, Text-Scene produces descriptions that are accurate, detailed, and human-interpretable, capturing object-level details and global-level context. Experimental results on benchmarks demonstrate that our textual parses can faithfully represent 3D scenes and benefit downstream tasks. To evaluate the reasoning capability of MLLMs, we present InPlan3D, a comprehensive benchmark for 3D task planning, consisting of 3174 long-term planning tasks across 636 indoor scenes. We emphasize clarity and accessibility in our approach, aiming to make 3D scene content understandable through language. Code and datasets will be released.

cs.CV↗

arXiv2025

Dropping Experts, Recombining Neurons: Retraining-Free Pruning for Sparse Mixture-of-Experts LLMs

Sparse Mixture-of-Experts (SMoE) architectures are widely used in large language models (LLMs) due to their computational efficiency. However, though only a few experts are activated for each token, SMoE still requires loading all expert parameters, leading to high memory usage and challenges in deployment. Previous work has tried to reduce the overhead by pruning and merging experts, but primarily focused on expert-level operations, leaving neuron-level structure underexplored. We propose DERN (Dropping Experts, Recombining Neurons), a task-agnostic and retraining-free framework for expert pruning and reconstruction. We observe that experts are often misaligned and contain semantic conflicts at the neuron level, which poses challenges for direct merging. To solve this, DERN works in three steps: it first prunes redundant experts using router statistics; then it decomposes them into neuron-level expert segments, assigning each segment to its most compatible retained expert; and finally, it merges segments within each retained expert to build a compact representation. Experiments on Mixtral, Qwen, and DeepSeek SMoE models show that DERN improves performance by more than 5% on commonsense reasoning and MMLU benchmarks under 50% expert sparsity, without extra training. It also greatly reduces the number of experts and memory usage, making SMoE LLMs easier to deploy in practice.

cs.CL↗

arXiv2025

Learning Robust Representations via Bidirectional Transition for Visual Reinforcement Learning

Visual reinforcement learning has proven effective in solving control tasks with high-dimensional observations. However, extracting reliable and generalizable representations from vision-based observations remains a central challenge. Inspired by the human thought process, when the representation extracted from the observation can predict the future and trace history, the representation is reliable and accurate in comprehending the environment. Based on this concept, we introduce a Bidirectional Transition (BiT) model, which leverages the ability to bidirectionally predict environmental transitions both forward and backward to extract reliable representations. Our model demonstrates competitive generalization performance and sample efficiency on two settings of the DeepMind Control suite. Additionally, we utilize robotic manipulation and CARLA simulators to demonstrate the wide applicability of our method.

cs.CV↗

arXiv2025

MMAD: Multi-label Micro-Action Detection in Videos

Human body actions are an important form of non-verbal communication in social interactions. This paper specifically focuses on a subset of body actions known as micro-actions, which are subtle, low-intensity body movements with promising applications in human emotion analysis. In real-world scenarios, human micro-actions often temporally co-occur, with multiple micro-actions overlapping in time, such as concurrent head and hand movements. However, current research primarily focuses on recognizing individual micro-actions while overlooking their co-occurring nature. To address this gap, we propose a new task named Multi-label Micro-Action Detection (MMAD), which involves identifying all micro-actions in a given short video, determining their start and end times, and categorizing them. Accomplishing this requires a model capable of accurately capturing both long-term and short-term action relationships to detect multiple overlapping micro-actions. To facilitate the MMAD task, we introduce a new dataset named Multi-label Micro-Action-52 (MMA-52) and propose a baseline method equipped with a dual-path spatial-temporal adapter to address the challenges of subtle visual change in MMAD. We hope that MMA-52 can stimulate research on micro-action analysis in videos and prompt the development of spatio-temporal modeling in human-centric video understanding. The proposed MMA-52 dataset is available at: https://github.com/VUT-HFUT/Micro-Action.

cs.CV↗

arXiv2025

TSGS: Improving Gaussian Splatting for Transparent Surface Reconstruction via Normal and De-lighting Priors

Reconstructing transparent surfaces is essential for tasks such as robotic manipulation in labs, yet it poses a significant challenge for 3D reconstruction techniques like 3D Gaussian Splatting (3DGS). These methods often encounter a transparency-depth dilemma, where the pursuit of photorealistic rendering through standard $α$-blending undermines geometric precision, resulting in considerable depth estimation errors for transparent materials. To address this issue, we introduce Transparent Surface Gaussian Splatting (TSGS), a new framework that separates geometry learning from appearance refinement. In the geometry learning stage, TSGS focuses on geometry by using specular-suppressed inputs to accurately represent surfaces. In the second stage, TSGS improves visual fidelity through anisotropic specular modeling, crucially maintaining the established opacity to ensure geometric accuracy. To enhance depth inference, TSGS employs a first-surface depth extraction method. This technique uses a sliding window over $α$-blending weights to pinpoint the most likely surface location and calculates a robust weighted average depth. To evaluate the transparent surface reconstruction task under realistic conditions, we collect a TransLab dataset that includes complex transparent laboratory glassware. Extensive experiments on TransLab show that TSGS achieves accurate geometric reconstruction and realistic rendering of transparent objects simultaneously within the efficient 3DGS framework. Specifically, TSGS significantly surpasses current leading methods, achieving a 37.3% reduction in chamfer distance and an 8.0% improvement in F1 score compared to the top baseline. The code and dataset are available at https://longxiang-ai.github.io/TSGS/.

cs.CV↗

arXiv2025

Enhancing Large Language Models through Structured Reasoning

Recent Large Language Models (LLMs) have significantly advanced natural language processing and automated decision-making. However, these models still encounter difficulties when performing complex reasoning tasks involving logical deduction and systematic planning, primarily due to their reliance on implicit statistical relationships without structured knowledge representation.Inspired by cognitive science and neurosymbolic AI, we introduce a novel approach to enhance LLMs through explicit structured reasoning. First, we convert unstructured data into structured formats by explicitly annotating reasoning steps. We then employ this structured dataset to train LLMs through Supervised Fine-Tuning (SFT). Additionally, we enhance the structured reasoning capabilities of LLMs using Group Relative Policy Optimization (GRPO), incorporating two innovative algorithms--MAX-Flow and Longest Common Subsequence (LCS)--which notably improve reasoning effectiveness and reduce computational complexity. Experimental results from fine-tuning a DeepSeek-R1-Distill-Qwen-1.5B model demonstrate concise reasoning, robust performance across various scenarios, and improved compatibility with optimization techniques, validating the efficacy of structured reasoning integration in LLMs.

cs.CL↗

arXiv2025

EnergyMoGen: Compositional Human Motion Generation with Energy-Based Diffusion Model in Latent Space

Diffusion models, particularly latent diffusion models, have demonstrated remarkable success in text-driven human motion generation. However, it remains challenging for latent diffusion models to effectively compose multiple semantic concepts into a single, coherent motion sequence. To address this issue, we propose EnergyMoGen, which includes two spectrums of Energy-Based Models: (1) We interpret the diffusion model as a latent-aware energy-based model that generates motions by composing a set of diffusion models in latent space; (2) We introduce a semantic-aware energy model based on cross-attention, which enables semantic composition and adaptive gradient descent for text embeddings. To overcome the challenges of semantic inconsistency and motion distortion across these two spectrums, we introduce Synergistic Energy Fusion. This design allows the motion latent diffusion model to synthesize high-quality, complex motions by combining multiple energy terms corresponding to textual descriptions. Experiments show that our approach outperforms existing state-of-the-art models on various motion generation tasks, including text-to-motion generation, compositional motion generation, and multi-concept motion generation. Additionally, we demonstrate that our method can be used to extend motion datasets and improve the text-to-motion task.

cs.CV↗

arXiv2025

Uni3D-MoE: Scalable Multimodal 3D Scene Understanding via Mixture of Experts

Recent advancements in multimodal large language models (MLLMs) have demonstrated considerable potential for comprehensive 3D scene understanding. However, existing approaches typically utilize only one or a limited subset of 3D modalities, resulting in incomplete representations of 3D scenes and reduced interpretive accuracy. Furthermore, different types of queries inherently depend on distinct modalities, indicating that uniform processing of all modality tokens may fail to effectively capture query-specific context. To address these challenges, we propose Uni3D-MoE, a sparse Mixture-of-Experts (MoE)-based 3D MLLM designed to enable adaptive 3D multimodal fusion. Specifically, Uni3D-MoE integrates a comprehensive set of 3D modalities, including multi-view RGB and depth images, bird's-eye-view (BEV) maps, point clouds, and voxel representations. At its core, our framework employs a learnable routing mechanism within the sparse MoE-based large language model, dynamically selecting appropriate experts at the token level. Each expert specializes in processing multimodal tokens based on learned modality preferences, thus facilitating flexible collaboration tailored to diverse task-specific requirements. Extensive evaluations on standard 3D scene understanding benchmarks and specialized datasets demonstrate the efficacy of Uni3D-MoE.

cs.CV↗

arXiv2025

Prototypical Calibrating Ambiguous Samples for Micro-Action Recognition

Micro-Action Recognition (MAR) has gained increasing attention due to its crucial role as a form of non-verbal communication in social interactions, with promising potential for applications in human communication and emotion analysis. However, current approaches often overlook the inherent ambiguity in micro-actions, which arises from the wide category range and subtle visual differences between categories. This oversight hampers the accuracy of micro-action recognition. In this paper, we propose a novel Prototypical Calibrating Ambiguous Network (PCAN) to unleash and mitigate the ambiguity of MAR. Firstly, we employ a hierarchical action-tree to identify the ambiguous sample, categorizing them into distinct sets of ambiguous samples of false negatives and false positives, considering both body- and action-level categories. Secondly, we implement an ambiguous contrastive refinement module to calibrate these ambiguous samples by regulating the distance between ambiguous samples and their corresponding prototypes. This calibration process aims to pull false negative (FN) samples closer to their respective prototypes and push false positive (FP) samples apart from their affiliated prototypes. In addition, we propose a new prototypical diversity amplification loss to strengthen the model's capacity by amplifying the differences between different prototypes. Finally, we propose a prototype-guided rectification to rectify prediction by incorporating the representability of prototypes. Extensive experiments conducted on the benchmark dataset demonstrate the superior performance of our method compared to existing approaches. The code is available at https://github.com/kunli-cs/PCAN.

cs.CV↗

arXiv2025

Zero-1-to-A: Zero-Shot One Image to Animatable Head Avatars Using Video Diffusion

Animatable head avatar generation typically requires extensive data for training. To reduce the data requirements, a natural solution is to leverage existing data-free static avatar generation methods, such as pre-trained diffusion models with score distillation sampling (SDS), which align avatars with pseudo ground-truth outputs from the diffusion model. However, directly distilling 4D avatars from video diffusion often leads to over-smooth results due to spatial and temporal inconsistencies in the generated video. To address this issue, we propose Zero-1-to-A, a robust method that synthesizes a spatial and temporal consistency dataset for 4D avatar reconstruction using the video diffusion model. Specifically, Zero-1-to-A iteratively constructs video datasets and optimizes animatable avatars in a progressive manner, ensuring that avatar quality increases smoothly and consistently throughout the learning process. This progressive learning involves two stages: (1) Spatial Consistency Learning fixes expressions and learns from front-to-side views, and (2) Temporal Consistency Learning fixes views and learns from relaxed to exaggerated expressions, generating 4D avatars in a simple-to-complex manner. Extensive experiments demonstrate that Zero-1-to-A improves fidelity, animation quality, and rendering speed compared to existing diffusion-based methods, providing a solution for lifelike avatar creation. Code is publicly available at: https://github.com/ZhenglinZhou/Zero-1-to-A.

cs.CV↗

arXiv2025

VideoGrain: Modulating Space-Time Attention for Multi-grained Video Editing

Recent advancements in diffusion models have significantly improved video generation and editing capabilities. However, multi-grained video editing, which encompasses class-level, instance-level, and part-level modifications, remains a formidable challenge. The major difficulties in multi-grained editing include semantic misalignment of text-to-region control and feature coupling within the diffusion model. To address these difficulties, we present VideoGrain, a zero-shot approach that modulates space-time (cross- and self-) attention mechanisms to achieve fine-grained control over video content. We enhance text-to-region control by amplifying each local prompt's attention to its corresponding spatial-disentangled region while minimizing interactions with irrelevant areas in cross-attention. Additionally, we improve feature separation by increasing intra-region awareness and reducing inter-region interference in self-attention. Extensive experiments demonstrate our method achieves state-of-the-art performance in real-world scenarios. Our code, data, and demos are available at https://knightyxp.github.io/VideoGrain_project_page/

cs.CV↗

arXiv2025

DreamDPO: Aligning Text-to-3D Generation with Human Preferences via Direct Preference Optimization

Text-to-3D generation automates 3D content creation from textual descriptions, which offers transformative potential across various fields. However, existing methods often struggle to align generated content with human preferences, limiting their applicability and flexibility. To address these limitations, in this paper, we propose DreamDPO, an optimization-based framework that integrates human preferences into the 3D generation process, through direct preference optimization. Practically, DreamDPO first constructs pairwise examples, then compare their alignment with human preferences using reward or large multimodal models, and lastly optimizes the 3D representation with a preference-driven loss function. By leveraging pairwise comparison to reflect preferences, DreamDPO reduces reliance on precise pointwise quality evaluations while enabling fine-grained controllability through preference-guided optimization. Experiments demonstrate that DreamDPO achieves competitive results, and provides higher-quality and more controllable 3D content compared to existing methods. The code and models will be open-sourced.

cs.CL↗

arXiv2025

Prompt-Aware Controllable Shadow Removal

Shadow removal aims to restore the image content in shadowed regions. While deep learning-based methods have shown promising results, they still face key challenges: 1) uncontrolled removal of all shadows, or 2) controllable removal but heavily relies on precise shadow region masks. To address these issues, we introduce a novel paradigm: prompt-aware controllable shadow removal. Unlike existing approaches, our paradigm allows for targeted shadow removal from specific subjects based on user prompts (e.g., dots, lines, or subject masks). This approach eliminates the need for shadow annotations and offers flexible, user-controlled shadow removal. Specifically, we propose an end-to-end learnable model, the Prompt-Aware Controllable Shadow Removal Network (PACSRNet). PACSRNet consists of two key modules: a prompt-aware module that generates shadow masks for the specified subject based on the user prompt, and a shadow removal module that uses the shadow prior from the first module to restore the content in the shadowed regions. Additionally, we enhance the shadow removal module by incorporating feature information from the prompt-aware module through a linear operation, providing prompt-guided support for shadow removal. Recognizing that existing shadow removal datasets lack diverse user prompts, we contribute a new dataset specifically designed for prompt-based controllable shadow removal. Extensive experimental results demonstrate the effectiveness and superiority of PACSRNet.

cs.CV↗