Lutao Jiang

Haotian Bai, Yuanhuiyi Lyu, Lutao Jiang et al.

Text-to-3D form plays a crucial role in creating editable 3D scenes for AR/VR. Recent advances have shown promise in merging neural radiance fields (NeRFs) with pre-trained diffusion models for text-to-3D object generation. However, one enduring challenge is their inadequate capability to accurately parse and regenerate consistent multi-object environments. Specifically, these models encounter difficulties in accurately representing quantity and style prompted by multi-object texts, often resulting in a collapse of the rendering fidelity that fails to match the semantic intricacies. Moreover, amalgamating these elements into a coherent 3D scene is a substantial challenge, stemming from generic distribution inherent in diffusion models. To tackle the issue of 'guidance collapse' and further enhance scene consistency, we propose a novel framework, dubbed CompoNeRF, by integrating an editable 3D scene layout with object-specific and scene-wide guidance mechanisms. It initiates by interpreting a complex text into the layout populated with multiple NeRFs, each paired with a corresponding subtext prompt for precise object depiction. Next, a tailored composition module seamlessly blends these NeRFs, promoting consistency, while the dual-level text guidance reduces ambiguity and boosts accuracy. Noticeably, our composition design permits decomposition. This enables flexible scene editing and recomposition into new scenes based on the edited layout or text prompts. Utilizing the open-source Stable Diffusion model, CompoNeRF generates multi-object scenes with high fidelity. Remarkably, our framework achieves up to a \textbf{54\%} improvement by the multi-view CLIP score metric. Our user study indicates that our method has significantly improved semantic accuracy, multi-view consistency, and individual recognizability for multi-object scene generation.

Libo Zhang, Lutao Jiang, Ruyi Ji et al.

Automatic security inspection relying on computer vision technology is a challenging task in real-world scenarios due to many factors, such as intra-class variance, class imbalance, and occlusion. Most previous methods rarely touch the cases where the prohibited items are deliberately hidden in messy objects because of the scarcity of large-scale datasets, hindering their applications. To address this issue and facilitate related research, we present a large-scale dataset, named PIDray, which covers various cases in real-world scenarios for prohibited item detection, especially for deliberately hidden items. In specific, PIDray collects 124,486 X-ray images for $12$ categories of prohibited items, and each image is manually annotated with careful inspection, which makes it, to our best knowledge, to largest prohibited items detection dataset to date. Meanwhile, we propose a general divide-and-conquer pipeline to develop baseline algorithms on PIDray. Specifically, we adopt the tree-like structure to suppress the influence of the long-tailed issue in the PIDray dataset, where the first course-grained node is tasked with the binary classification to alleviate the influence of head category, while the subsequent fine-grained node is dedicated to the specific tasks of the tail categories. Based on this simple yet effective scheme, we offer strong task-specific baselines across object detection, instance segmentation, and multi-label classification tasks and verify the generalization ability on common datasets (e.g., COCO and PASCAL VOC). Extensive experiments on PIDray demonstrate that the proposed method performs favorably against current state-of-the-art methods, especially for deliberately hidden items. Our benchmark and codes will be released at https://github.com/lutao2021/PIDray.

Lutao Jiang, Ruyi Ji, Libo Zhang

In this paper, we develop a new method, termed SDF-3DGAN, for 3D object generation and 3D-Aware image synthesis tasks, which introduce implicit Signed Distance Function (SDF) as the 3D object representation method in the generative field. We apply SDF for higher quality representation of 3D object in space and design a new SDF neural renderer, which has higher efficiency and higher accuracy. To train only on 2D images, we first generate the objects, which are represented by SDF, from Gaussian distribution. Then we render them to 2D images and use them to apply GAN training method together with 2D images in the dataset. In the new rendering method, we relieve all the potential of SDF mathematical property to alleviate computation pressure in the previous SDF neural renderer. In specific, our new SDF neural renderer can solve the problem of sampling ambiguity when the number of sampling point is not enough, \ie use the less points to finish higher quality sampling task in the rendering pipeline. And in this rendering pipeline, we can locate the surface easily. Therefore, we apply normal loss on it to control the smoothness of generated object surface, which can make our method enjoy the much higher generation quality. Quantitative and qualitative experiments conducted on public benchmarks demonstrate favorable performance against the state-of-the-art methods in 3D object generation task and 3D-Aware image synthesis task. Our codes will be released at https://github.com/lutao2021/SDF-3DGAN.

Ye Pan, Chi Kit Wong, Yuanhuiyi Lyu et al.

Multimodal Large Language Models (MLLMs) have demonstrated remarkable video reasoning capabilities across diverse tasks. However, their ability to understand human intent at a fine-grained level in egocentric videos remains largely unexplored. Existing benchmarks focus primarily on episode-level intent reasoning, overlooking the finer granularity of step-level intent understanding. Yet applications such as intelligent assistants, robotic imitation learning, and augmented reality guidance require understanding not only what a person is doing at each step, but also why and what comes next, in order to provide timely and context-aware support. To this end, we introduce EgoIntent, a step-level intent understanding benchmark for egocentric videos. It comprises 3,014 steps spanning 15 diverse indoor and outdoor daily-life scenarios, and evaluates models on three complementary dimensions: local intent (What), global intent (Why), and next-step plan (Next). Crucially, each clip is truncated immediately before the key outcome of the queried step (e.g., contact or grasp) occurs and contains no frames from subsequent steps, preventing future-frame leakage and enabling a clean evaluation of anticipatory step understanding and next-step planning. We evaluate 15 MLLMs, including both state-of-the-art closed-source and open-source models. Even the best-performing model achieves an average score of only 33.31 across the three intent dimensions, underscoring that step-level intent understanding in egocentric videos remains a highly challenging problem that calls for further investigation.

Tanghui Jia, Dongyu Yan, Dehao Hao et al.

In this report, we introduce UltraShape 1.0, a scalable 3D diffusion framework for high-fidelity 3D geometry generation. The proposed approach adopts a two-stage generation pipeline: a coarse global structure is first synthesized and then refined to produce detailed, high-quality geometry. To support reliable 3D generation, we develop a comprehensive data processing pipeline that includes a novel watertight processing method and high-quality data filtering. This pipeline improves the geometric quality of publicly available 3D datasets by removing low-quality samples, filling holes, and thickening thin structures, while preserving fine-grained geometric details. To enable fine-grained geometry refinement, we decouple spatial localization from geometric detail synthesis in the diffusion process. We achieve this by performing voxel-based refinement at fixed spatial locations, where voxel queries derived from coarse geometry provide explicit positional anchors encoded via RoPE, allowing the diffusion model to focus on synthesizing local geometric details within a reduced, structured solution space. Our model is trained exclusively on publicly available 3D datasets, achieving strong geometric quality despite limited training resources. Extensive evaluations demonstrate that UltraShape 1.0 performs competitively with existing open-source methods in both data processing quality and geometry generation. All code and trained models will be released to support future research.

Yuanhuiyi Lyu, Kaiyu Lei, Ziqiao Weng et al.

Reasoning-based text-to-image (T2I) generation requires models to interpret complex prompts accurately. Existing reasoning frameworks can be broadly categorized into two types: (1) Text-Only Reasoning, which is computationally efficient but lacks access to visual context, often resulting in the omission of critical spatial and visual elements; and (2) Text-Image Interleaved Reasoning, which leverages a T2I generator to provide visual references during the reasoning process. While this approach enhances visual grounding, it incurs substantial computational costs and constrains the reasoning capacity of MLLMs to the representational limitations of the generator. To this end, we propose StruVis, a novel framework that enhances T2I generation through Thinking with Structured Vision. Instead of relying on intermediate image generation, StruVis employs text-based structured visual representations as intermediate reasoning states, thereby enabling the MLLM to effectively "perceive" visual structure within a purely text-based reasoning process. Powered by this, the reasoning potential for T2I generation of the MLLM is unlocked through structured-vision-guided reasoning. Additionally, as a generator-agnostic reasoning framework, our proposed StruVis can be seamlessly integrated with diverse T2I generators and efficiently enhance their performance in reasoning-based T2I generation. Extensive experiments demonstrate that StruVis achieves significant performance improvements on reasoning-based T2I benchmarks, e.g., a 4.61% gain on T2I-ReasonBench and a 4% gain on WISE.

Lutao Jiang, Xu Zheng, Yuanhuiyi Lyu et al.

Text-to-3D synthesis has recently seen intriguing advances by combining the text-to-image priors with 3D representation methods, e.g., 3D Gaussian Splatting (3D GS), via Score Distillation Sampling (SDS). However, a hurdle of existing methods is the low efficiency, per-prompt optimization for a single 3D object. Therefore, it is imperative for a paradigm shift from per-prompt optimization to feed-forward generation for any unseen text prompts, which yet remains challenging. An obstacle is how to directly generate a set of millions of 3D Gaussians to represent a 3D object. This paper presents BrightDreamer, an end-to-end feed-forward approach that can achieve generalizable and fast (77 ms) text-to-3D generation. Our key idea is to formulate the generation process as estimating the 3D deformation from an anchor shape with predefined positions. For this, we first propose a Text-guided Shape Deformation (TSD) network to predict the deformed shape and its new positions, used as the centers (one attribute) of 3D Gaussians. To estimate the other four attributes (i.e., scaling, rotation, opacity, and SH), we then design a novel Text-guided Triplane Generator (TTG) to generate a triplane representation for a 3D object. The center of each Gaussian enables us to transform the spatial feature into the four attributes. The generated 3D Gaussians can be finally rendered at 705 frames per second. Extensive experiments demonstrate the superiority of our method over existing methods. Also, BrightDreamer possesses a strong semantic understanding capability even for complex text prompts. The code is available in the project page.

Lutao Jiang, Hangyu Li, Lin Wang

Text-to-3D content creation has recently received much attention, especially with the prevalence of 3D Gaussians Splatting. In general, GS-based methods comprise two key stages: initialization and rendering optimization. To achieve initialization, existing works directly apply random sphere initialization or 3D diffusion models, e.g., Point-E, to derive the initial shapes. However, such strategies suffer from two critical yet challenging problems: 1) the final shapes are still similar to the initial ones even after training; 2) shapes can be produced only from simple texts, e.g., "a dog", not for lexically richer texts, e.g., "a dog is sitting on the top of the airplane". To address these problems, this paper proposes a novel general framework to boost the 3D GS Initialization for text-to-3D generation upon the lexical richness. Our key idea is to aggregate 3D Gaussians into spatially uniform voxels to represent complex shapes while enabling the spatial interaction among the 3D Gaussians and semantic interaction between Gaussians and texts. Specifically, we first construct a voxelized representation, where each voxel holds a 3D Gaussian with its position, scale, and rotation fixed while setting opacity as the sole factor to determine a position's occupancy. We then design an initialization network mainly consisting of two novel components: 1) Global Information Perception (GIP) block and 2) Gaussians-Text Fusion (GTF) block. Such a design enables each 3D Gaussian to assimilate the spatial information from other areas and semantic information from texts. Extensive experiments show the superiority of our framework of high-quality 3D GS initialization against the existing methods, e.g., Shap-E, by taking lexically simple, medium, and hard texts. Also, our framework can be seamlessly plugged into SoTA training frameworks, e.g., LucidDreamer, for semantically consistent text-to-3D generation.

Jiazhou Zhou, Qing Jiang, Kanghao Chen et al.

Object detection methods have evolved from closed-set to open-set paradigms over the years. Current open-set object detectors, however, remain constrained by their exclusive reliance on positive indicators based on given prompts like text descriptions or visual exemplars. This positive-only paradigm experiences consistent vulnerability to visually similar but semantically different distractors. We propose T-Rex-Omni, a novel framework that addresses this limitation by incorporating negative visual prompts to negate hard negative distractors. Specifically, we first introduce a unified visual prompt encoder that jointly processes positive and negative visual prompts. Next, a training-free Negating Negative Computing (NNC) module is proposed to dynamically suppress negative responses during the probability computing stage. To further boost performance through fine-tuning, our Negating Negative Hinge (NNH) loss enforces discriminative margins between positive and negative embeddings. T-Rex-Omni supports flexible deployment in both positive-only and joint positive-negative inference modes, accommodating either user-specified or automatically generated negative examples. Extensive experiments demonstrate remarkable zero-shot detection performance, significantly narrowing the performance gap between visual-prompted and text-prompted methods while showing particular strength in long-tailed scenarios (51.2 AP_r on LVIS-minival). This work establishes negative prompts as a crucial new dimension for advancing open-set visual recognition systems.

Lutao Jiang, Lin Wang

Following the burgeoning interest in implicit neural representation, Neural Light Field (NeLF) has been introduced to predict the color of a ray directly. Unlike Neural Radiance Field (NeRF), NeLF does not create a point-wise representation by predicting color and volume density for each point in space. However, the current NeLF methods face a challenge as they need to train a NeRF model first and then synthesize over 10K views to train NeLF for improved performance. Additionally, the rendering quality of NeLF methods is lower compared to NeRF methods. In this paper, we propose G-NeLF, a versatile grid-based NeLF approach that utilizes spatial-aware features to unleash the potential of the neural network's inference capability, and consequently overcome the difficulties of NeLF training. Specifically, we employ a spatial-aware feature sequence derived from a meticulously crafted grid as the ray's representation. Drawing from our empirical studies on the adaptability of multi-resolution hash tables, we introduce a novel grid-based ray representation for NeLF that can represent the entire space with a very limited number of parameters. To better utilize the sequence feature, we design a lightweight ray color decoder that simulates the ray propagation process, enabling a more efficient inference of the ray's color. G-NeLF can be trained without necessitating significant storage overhead and with the model size of only 0.95 MB to surpass previous state-of-the-art NeLF. Moreover, compared with grid-based NeRF methods, e.g., Instant-NGP, we only utilize one-tenth of its parameters to achieve higher performance. Our code will be released upon acceptance.

Haosen Wang, Zhenyang Li, Yinqiang Zhang et al.

Zero-shot object navigation has advanced rapidly with open-vocabulary detectors, image--text models, and language-guided exploration. However, even after current methods detect a plausible target hypothesis, the agent may still oscillate between exploration and pursuit, or abandon the object near success. We identify this failure mode as an action consistency gap: semantic evidence is repeatedly reinterpreted at each step without persistent commitment across the episode. We introduce ConsistNav, a training-free zero-shot ObjectNav framework built around a semantic executive composed of three coordinated modules: Finite-State Executive Controller stages target pursuit through guarded semantic phases; Persistent Candidate Memory accumulates cross-frame target evidence into stable object hypotheses; and Stability-Aware Action Control suppresses rotational stagnation, ineffective pursuit, and unverified stopping. This design changes neither the detector nor the low-level planner; instead, it controls when semantic evidence should influence navigation and when it should be suppressed or revisited. We conduct extensive experiments on HM3D and MP3D, where ConsistNav achieves state-of-the-art results among compared zero-shot ObjectNav methods and improves SR by 11.4% and SPL by 7.9% over the controlled baseline on MP3D. Ablation studies and real-world deployment experiments further demonstrate the effectiveness and robustness of the proposed executive mechanism.

Xu Zheng, Zihao Dongfang, Lutao Jiang et al.

Humans possess spatial reasoning abilities that enable them to understand spaces through multimodal observations, such as vision and sound. Large multimodal reasoning models extend these abilities by learning to perceive and reason, showing promising performance across diverse spatial tasks. However, systematic reviews and publicly available benchmarks for these models remain limited. In this survey, we provide a comprehensive review of multimodal spatial reasoning tasks with large models, categorizing recent progress in multimodal large language models (MLLMs) and introducing open benchmarks for evaluation. We begin by outlining general spatial reasoning, focusing on post-training techniques, explainability, and architecture. Beyond classical 2D tasks, we examine spatial relationship reasoning, scene and layout understanding, as well as visual question answering and grounding in 3D space. We also review advances in embodied AI, including vision-language navigation and action models. Additionally, we consider emerging modalities such as audio and egocentric video, which contribute to novel spatial understanding through new sensors. We believe this survey establishes a solid foundation and offers insights into the growing field of multimodal spatial reasoning. Updated information about this survey, codes and implementation of the open benchmarks can be found at https://github.com/zhengxuJosh/Awesome-Spatial-Reasoning.

Zixin Zhang, Kanghao Chen, Xingwang Lin et al.

The ability to use, understand, and create tools is a hallmark of human intelligence, enabling sophisticated interaction with the physical world. For any general-purpose intelligent agent to achieve true versatility, it must also master these fundamental skills. While modern Multimodal Large Language Models (MLLMs) leverage their extensive common knowledge for high-level planning in embodied AI and in downstream Vision-Language-Action (VLA) models, the extent of their true understanding of physical tools remains unquantified. To bridge this gap, we present PhysToolBench, the first benchmark dedicated to evaluating the comprehension of physical tools by MLLMs. Our benchmark is structured as a Visual Question Answering (VQA) dataset comprising over 1,000 image-text pairs. It assesses capabilities across three distinct difficulty levels: (1) Tool Recognition: Requiring the recognition of a tool's primary function. (2) Tool Understanding: Testing the ability to grasp the underlying principles of a tool's operation. (3) Tool Creation: Challenging the model to fashion a new tool from surrounding objects when conventional options are unavailable. Our comprehensive evaluation of 32 MLLMs-spanning proprietary, open-source, specialized embodied, and backbones in VLAs-reveals a significant deficiency in tool understanding. Furthermore, we provide an in-depth analysis and propose preliminary solutions. Code and dataset are publicly available.

Chenfei Liao, Wensong Wang, Zichen Wen et al.

Recent endeavors to accelerate inference in Multimodal Large Language Models (MLLMs) have primarily focused on visual token compression. The effectiveness of these methods is typically assessed by measuring the accuracy drop on established benchmarks, comparing model performance before and after compression. However, these benchmarks are originally designed to assess the perception and reasoning capabilities of MLLMs, rather than to evaluate compression techniques. As a result, directly applying them to visual token compression introduces a task mismatch. Strikingly, our investigation reveals that simple image downsampling consistently outperforms many advanced compression methods across multiple widely used benchmarks. Through extensive experiments, we make the following observations: (i) Current benchmarks are noisy for the visual token compression task. (ii) Down-sampling is able to serve as a data filter to evaluate the difficulty of samples in the visual token compression task. Motivated by these findings, we introduce VTC-Bench, an evaluation framework that incorporates a data filtering mechanism to denoise existing benchmarks, thereby enabling fairer and more accurate assessment of visual token compression methods. All data and code are available at https://github.com/Chenfei-Liao/VTC-Bench.

Yuanhuiyi Lyu, Chi Kit Wong, Chenfei Liao et al.

Recent works have made notable advancements in enhancing unified models for text-to-image generation through the Chain-of-Thought (CoT). However, these reasoning methods separate the processes of understanding and generation, which limits their ability to guide the reasoning of unified models in addressing the deficiencies of their generative capabilities. To this end, we propose a novel reasoning framework for unified models, Understanding-in-Generation (UiG), which harnesses the robust understanding capabilities of unified models to reinforce their performance in image generation. The core insight of our UiG is to integrate generative guidance by the strong understanding capabilities during the reasoning process, thereby mitigating the limitations of generative abilities. To achieve this, we introduce "Image Editing" as a bridge to infuse understanding into the generation process. Initially, we verify the generated image and incorporate the understanding of unified models into the editing instructions. Subsequently, we enhance the generated image step by step, gradually infusing the understanding into the generation process. Our UiG framework demonstrates a significant performance improvement in text-to-image generation over existing text-to-image reasoning methods, e.g., a 3.92% gain on the long prompt setting of the TIIF benchmark. The project code: https://github.com/QC-LY/UiG

Xu Zheng, Ziqiao Weng, Yuanhuiyi Lyu et al.

Retrieval-augmented generation (RAG) has emerged as a pivotal technique in artificial intelligence (AI), particularly in enhancing the capabilities of large language models (LLMs) by enabling access to external, reliable, and up-to-date knowledge sources. In the context of AI-Generated Content (AIGC), RAG has proven invaluable by augmenting model outputs with supplementary, relevant information, thus improving their quality. Recently, the potential of RAG has extended beyond natural language processing, with emerging methods integrating retrieval-augmented strategies into the computer vision (CV) domain. These approaches aim to address the limitations of relying solely on internal model knowledge by incorporating authoritative external knowledge bases, thereby improving both the understanding and generation capabilities of vision models. This survey provides a comprehensive review of the current state of retrieval-augmented techniques in CV, focusing on two main areas: (I) visual understanding and (II) visual generation. In the realm of visual understanding, we systematically review tasks ranging from basic image recognition to complex applications such as medical report generation and multimodal question answering. For visual content generation, we examine the application of RAG in tasks related to image, video, and 3D generation. Furthermore, we explore recent advancements in RAG for embodied AI, with a particular focus on applications in planning, task execution, multimodal perception, interaction, and specialized domains. Given that the integration of retrieval-augmented techniques in CV is still in its early stages, we also highlight the key limitations of current approaches and propose future research directions to drive the development of this promising area.

Xu Zheng, Yuanhuiyi Lyu, Lutao Jiang et al.

In this paper, we address the challenging modality-agnostic semantic segmentation (MaSS), aiming at centering the value of every modality at every feature granularity. Training with all available visual modalities and effectively fusing an arbitrary combination of them is essential for robust multi-modal fusion in semantic segmentation, especially in real-world scenarios, yet remains less explored to date. Existing approaches often place RGB at the center, treating other modalities as secondary, resulting in an asymmetric architecture. However, RGB alone can be limiting in scenarios like nighttime, where modalities such as event data excel. Therefore, a resilient fusion model must dynamically adapt to each modality's strengths while compensating for weaker inputs.To this end, we introduce the MAGIC++ framework, which comprises two key plug-and-play modules for effective multi-modal fusion and hierarchical modality selection that can be equipped with various backbone models. Firstly, we introduce a multi-modal interaction module to efficiently process features from the input multi-modal batches and extract complementary scene information with channel-wise and spatial-wise guidance. On top, a unified multi-scale arbitrary-modal selection module is proposed to utilize the aggregated features as the benchmark to rank the multi-modal features based on the similarity scores at hierarchical feature spaces. This way, our method can eliminate the dependence on RGB modality at every feature granularity and better overcome sensor failures and environmental noises while ensuring the segmentation performance. Under the common multi-modal setting, our method achieves state-of-the-art performance on both real-world and synthetic benchmarks. Moreover, our method is superior in the novel modality-agnostic setting, where it outperforms prior arts by a large margin.

Yuanhuiyi Lyu, Xu Zheng, Lutao Jiang et al.

Recent text-to-image generative models, e.g., Stable Diffusion V3 and Flux, have achieved notable progress. However, these models are strongly restricted to their limited knowledge, a.k.a., their own fixed parameters, that are trained with closed datasets. This leads to significant hallucinations or distortions when facing fine-grained and unseen novel real-world objects, e.g., the appearance of the Tesla Cybertruck. To this end, we present the first real-object-based retrieval-augmented generation framework (RealRAG), which augments fine-grained and unseen novel object generation by learning and retrieving real-world images to overcome the knowledge gaps of generative models. Specifically, to integrate missing memory for unseen novel object generation, we train a reflective retriever by self-reflective contrastive learning, which injects the generator's knowledge into the sef-reflective negatives, ensuring that the retrieved augmented images compensate for the model's missing knowledge. Furthermore, the real-object-based framework integrates fine-grained visual knowledge for the generative models, tackling the distortion problem and improving the realism for fine-grained object generation. Our Real-RAG is superior in its modular application to all types of state-of-the-art text-to-image generative models and also delivers remarkable performance boosts with all of them, such as a gain of 16.18% FID score with the auto-regressive model on the Stanford Car benchmark.

Xu Zheng, Haiwei Xue, Jialei Chen et al.

Simultaneously using multimodal inputs from multiple sensors to train segmentors is intuitively advantageous but practically challenging. A key challenge is unimodal bias, where multimodal segmentors over rely on certain modalities, causing performance drops when others are missing, common in real world applications. To this end, we develop the first framework for learning robust segmentor that can handle any combinations of visual modalities. Specifically, we first introduce a parallel multimodal learning strategy for learning a strong teacher. The cross-modal and unimodal distillation is then achieved in the multi scale representation space by transferring the feature level knowledge from multimodal to anymodal segmentors, aiming at addressing the unimodal bias and avoiding over-reliance on specific modalities. Moreover, a prediction level modality agnostic semantic distillation is proposed to achieve semantic knowledge transferring for segmentation. Extensive experiments on both synthetic and real-world multi-sensor benchmarks demonstrate that our method achieves superior performance.

Xu Zheng, Chenfei Liao, Yuqian Fu et al.

Recent advances in Multimodal Large Language Models (MLLMs) have shown promising results in integrating diverse modalities such as texts and images. MLLMs are heavily influenced by modality bias, often relying on language while under-utilizing other modalities like visual inputs. This position paper argues that MLLMs are deeply affected by modality bias. Firstly, we diagnose the current state of modality bias, highlighting its manifestations across various tasks. Secondly, we propose a systematic research road-map related to modality bias in MLLMs. Thirdly, we identify key factors of modality bias in MLLMs and offer actionable suggestions for future research to mitigate it. To substantiate these findings, we conduct experiments that demonstrate the influence of each factor: 1. Data Characteristics: Language data is compact and abstract, while visual data is redundant and complex, creating an inherent imbalance in learning dynamics. 2. Imbalanced Backbone Capabilities: The dominance of pretrained language models in MLLMs leads to overreliance on language and neglect of visual information. 3. Training Objectives: Current objectives often fail to promote balanced cross-modal alignment, resulting in shortcut learning biased toward language. These findings highlight the need for balanced training strategies and model architectures to better integrate multiple modalities in MLLMs. We call for interdisciplinary efforts to tackle these challenges and drive innovation in MLLM research. Our work provides a fresh perspective on modality bias in MLLMs and offers insights for developing more robust and generalizable multimodal systems-advancing progress toward Artificial General Intelligence.

Xu Zheng, Yuanhuiyi Lyu, Lutao Jiang et al.

Fusing and balancing multi-modal inputs from novel sensors for dense prediction tasks, particularly semantic segmentation, is critically important yet remains a significant challenge. One major limitation is the tendency of multi-modal frameworks to over-rely on easily learnable modalities, a phenomenon referred to as unimodal dominance or bias. This issue becomes especially problematic in real-world scenarios where the dominant modality may be unavailable, resulting in severe performance degradation. To this end, we apply a simple but effective plug-and-play regularization term based on functional entropy, which introduces no additional parameters or modules. This term is designed to intuitively balance the contribution of each visual modality to the segmentation results. Specifically, we leverage the log-Sobolev inequality to bound functional entropy using functional-Fisher-information. By maximizing the information contributed by each visual modality, our approach mitigates unimodal dominance and establishes a more balanced and robust segmentation framework. A multi-scale regularization module is proposed to apply our proposed plug-and-play term on high-level features and also segmentation predictions for more balanced multi-modal learning. Extensive experiments on three datasets demonstrate that our proposed method achieves superior performance, i.e., +13.94%, +3.25%, and +3.64%, without introducing any additional parameters.

Lutao Jiang, Jiantao Lin, Kanghao Chen et al.

We propose DiMeR, a novel geometry-texture disentangled feed-forward model with 3D supervision for sparse-view mesh reconstruction. Existing methods confront two persistent obstacles: (i) textures can conceal geometric errors, i.e., visually plausible images can be rendered even with wrong geometry, producing multiple ambiguous optimization objectives in geometry-texture mixed solution space for similar objects; and (ii) prevailing mesh extraction methods are redundant, unstable, and lack 3D supervision. To solve these challenges, we rethink the inductive bias for mesh reconstruction. First, we disentangle the unified geometry-texture solution space, where a single input admits multiple feasible solutions, into geometry and texture spaces individually. Specifically, given that normal maps are strictly consistent with geometry and accurately capture surface variations, the normal maps serve as the sole input for geometry prediction in DiMeR, while the texture is estimated from RGB images. Second, we streamline the algorithm of mesh extraction by eliminating modules with low performance/cost ratios and redesigning regularization losses with 3D supervision. Notably, DiMeR still accepts raw RGB images as input by leveraging foundation models for normal prediction. Extensive experiments demonstrate that DiMeR generalises across sparse-view-, single-image-, and text-to-3D tasks, consistently outperforming baselines. On the GSO and OmniObject3D datasets, DiMeR significantly reduces Chamfer Distance by more than 30%.

Xu Zheng, Chenfei Liao, Ziqiao Weng et al.

Omnidirectional vision, using 360-degree vision to understand the environment, has become increasingly critical across domains like robotics, industrial inspection, and environmental monitoring. Compared to traditional pinhole vision, omnidirectional vision provides holistic environmental awareness, significantly enhancing the completeness of scene perception and the reliability of decision-making. However, foundational research in this area has historically lagged behind traditional pinhole vision. This talk presents an emerging trend in the embodied AI era: the rapid development of omnidirectional vision, driven by growing industrial demand and academic interest. We highlight recent breakthroughs in omnidirectional generation, omnidirectional perception, omnidirectional understanding, and related datasets. Drawing on insights from both academia and industry, we propose an ideal panoramic system architecture in the embodied AI era, PANORAMA, which consists of four key subsystems. Moreover, we offer in-depth opinions related to emerging trends and cross-community impacts at the intersection of panoramic vision and embodied AI, along with the future roadmap and open challenges. This overview synthesizes state-of-the-art advancements and outlines challenges and opportunities for future research in building robust, general-purpose omnidirectional AI systems in the embodied AI era.

Tongyan Hua, Lutao Jiang, Ying-Cong Chen et al.

Recent advancements in generative models have enabled 3D urban scene generation from satellite imagery, unlocking promising applications in gaming, digital twins, and beyond. However, most existing methods rely heavily on neural rendering techniques, which hinder their ability to produce detailed 3D structures on a broader scale, largely due to the inherent structural ambiguity derived from relatively limited 2D observations. To address this challenge, we propose Sat2City, a novel framework that synergizes the representational capacity of sparse voxel grids with latent diffusion models, tailored specifically for our novel 3D city dataset. Our approach is enabled by three key components: (1) A cascaded latent diffusion framework that progressively recovers 3D city structures from satellite imagery, (2) a Re-Hash operation at its Variational Autoencoder (VAE) bottleneck to compute multi-scale feature grids for stable appearance optimization and (3) an inverse sampling strategy enabling implicit supervision for smooth appearance transitioning.To overcome the challenge of collecting real-world city-scale 3D models with high-quality geometry and appearance, we introduce a dataset of synthesized large-scale 3D cities paired with satellite-view height maps. Validated on this dataset, our framework generates detailed 3D structures from a single satellite image, achieving superior fidelity compared to existing city generation models.

Lutao Jiang, Zidong Cao, Weikai Chen et al.

Promptable instance segmentation is widely adopted in embodied and AR systems, yet the performance of foundation models trained on perspective imagery often degrades on 360° panoramas. In this paper, we introduce Segment Any 4K Panorama (SAP), a foundation model for 4K high-resolution panoramic instance-level segmentation. We reformulate panoramic segmentation as fixed-trajectory perspective video segmentation, decomposing a panorama into overlapping perspective patches sampled along a continuous spherical traversal. This memory-aligned reformulation preserves native 4K resolution while restoring the smooth viewpoint transitions required for stable cross-view propagation. To enable large-scale supervision, we synthesize 183,440 4K-resolution panoramic images with instance segmentation labels using the InfiniGen engine. Trained under this trajectory-aligned paradigm, SAP generalizes effectively to real-world 360° images, achieving +17.2 zero-shot mIoU gain over vanilla SAM2 of different sizes on real-world 4K panorama benchmark.