arXiv subjects

Ray Zhang

Publications and source records attributed to Ray Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Video-MME-Logical: A Controlled Diagnostic Benchmark for Video Temporal-Logical Reasoning

Recent interest in multimodal large language models (MLLMs) raises a central question: can they reason over dynamic visual evidence rather than merely recognize objects or events in individual frames? This ability, which we refer to as video temporal-logical reasoning, requires models to maintain, update, and compose evidence as visual states evolve across frames. Existing video benchmarks often conflate this capability with scene complexity, static recognition, or uncontrolled temporal variation. To isolate this capability, we introduce Video-MME-Logical, a controlled benchmark organized around five temporal-logical operations: state tracking, sequential counting, temporal ordering, dynamic spatiality, and structural composition. The benchmark contains 25 fine-grained task categories generated with controlled object states, transitions, temporal dependencies, and logical compositions. It enables difficulty-controlled final-answer evaluation by varying temporal horizon and reasoning complexity, and supports intermediate-state diagnostics by verifying whether models recover the required logical reasoning trace before producing the final answer. Experiments with state-of-the-art MLLMs reveal a substantial human-model gap, especially as temporal-logical complexity increases. Supervised fine-tuning on up to 500K generated samples improves performance but remains insufficient to close the reasoning gap, positioning Video-MME-Logical as a scalable testbed for analyzing and improving temporal-logical reasoning in MLLMs.

cs.CV↗

arXiv2026

InterleaveThinker: Reinforcing Agentic Interleaved Generation

Recent image generators have demonstrated impressive photorealism and instruction-following capabilities in single-image generation and editing. However, constrained by their architectures, they cannot achieve interleaved generation (text-image sequence), which has crucial applications in visual narratives, guidance, and embodied manipulation. Even the latest open-source Unified Multimodal Models (UMMs) exhibit limited performance in this regard. In this paper, we introduce InterleaveThinker, the first multi-agent pipeline designed to endow any existing image generator with interleaved generation capabilities. Specifically, we employ a planner agent to organize the image-text input sequence, instructing the image generator on the required execution at each step. Subsequently, we introduce a critic agent to evaluate the generator's outputs, identify samples that deviate from the planned instructions, and refine the instructions for regeneration. To implement this pipeline, we construct the Interleave-Planner-SFT-80k and Interleave-Critic-SFT-112k to perform a format cold-start. Then we develop Interleave-Critic-RL-13k to reinforce the step-wise instruction correction capability within a generation trajectory using GRPO. Since a single interleaved generation trajectory may involve over 25 generator calls, optimizing the entire trajectory is computationally impractical. Therefore, we propose accuracy reward and step-wise reward, allowing single-step RL to effectively guide the entire generation trajectory. The results show that InterleaveThinker improves performance across various image generators. On interleaved generation benchmarks, it achieves performance comparable to Nano Banana and GPT-5. Surprisingly, it also significantly enhances the base model on reasoning-based benchmarks; for example, on 4-step FLUX.2-klein, we observe substantial gains on WISE and RISE.

cs.CV↗

arXiv2026

Generalized-CVO: Fast and Correspondence-Free Local Point Cloud Registration with Second Order Riemannian Optimization

We propose a fast and correspondence-free local point cloud registration method that leverages geometric surface structure and reproducing kernel Hilbert space (RKHS) embeddings. The method represents point clouds as continuous functions with point-wise anisotropic kernels that encode local geometry. This formulation improves alignment along surface normals while relaxing alignment along tangential directions. To solve the resulting registration problem, we propose a second-order on-manifold optimization scheme with approximate Riemannian Hessians, achieving a speedup of up to 10x over the first-order solvers used in prior correspondence-free RKHS-based methods. We demonstrate improved frame-to-frame LiDAR and RGB-D tracking accuracy across diverse indoor and outdoor datasets. On a LiDAR tracking registration task in the driving domain, we achieve a reduction of $>55\%$ in both translational and rotational drift in challenging feature-sparse environments. On object registration benchmarks, we show improved robustness over ICP-based methods and further gains when refining global initialization, particularly under moderate misalignment.

cs.CV↗

arXiv2026

Dynamic Association of Semantics and Parameter Estimates by Filtering

We propose a probabilistic semantic filtering framework in which parameters of a dynamical system are inferred and associated with a closed set of semantic classes in a map. We extend existing methods to a multi-parameter setting using a posterior that tightly couples semantics with the parameter likelihoods, and propose a filter to compute this posterior sequentially, subject to dynamics in the map's state. Using Bayesian moment matching, we show that the computational complexity of measurement updates scales linearly in the dimension of the parameter space. Finally, we demonstrate limitations of applying existing methods to a problem from the driving domain, and show that the proposed framework better captures time-varying parameter-to-semantic associations.

eess.SY↗

arXiv2026

Vision-Conditioned Variational Bayesian Last Layer Dynamics Models

Agile control of robotic systems often requires anticipating how the environment affects system behavior. For example, a driver must perceive the road ahead to anticipate available friction and plan actions accordingly. Achieving such proactive adaptation within autonomous frameworks remains a challenge, particularly under rapidly changing conditions. Traditional modeling approaches often struggle to capture abrupt variations in system behavior, while adaptive methods are inherently reactive and may adapt too late to ensure safety. We propose a vision-conditioned variational Bayesian last-layer dynamics model that leverages visual context to anticipate changes in the environment. The model first learns nominal vehicle dynamics and is then fine-tuned with feature-wise affine transformations of latent features, enabling context-aware dynamics prediction. The resulting model is integrated into an optimal controller for vehicle racing. We validate our method on a Lexus LC500 racing through water puddles. With vision-conditioning, the system completed all 12 attempted laps under varying conditions. In contrast, all baselines without visual context consistently lost control, demonstrating the importance of proactive dynamics adaptation in high-performance applications.

cs.RO↗

arXiv2025

Are We Ready for RL in Text-to-3D Generation? A Progressive Investigation

Reinforcement learning (RL), earlier proven to be effective in large language and multi-modal models, has been successfully extended to enhance 2D image generation recently. However, applying RL to 3D generation remains largely unexplored due to the higher spatial complexity of 3D objects, which require globally consistent geometry and fine-grained local textures. This makes 3D generation significantly sensitive to reward designs and RL algorithms. To address these challenges, we conduct the first systematic study of RL for text-to-3D autoregressive generation across several dimensions. (1) Reward designs: We evaluate reward dimensions and model choices, showing that alignment with human preference is crucial, and that general multi-modal models provide robust signal for 3D attributes. (2) RL algorithms: We study GRPO variants, highlighting the effectiveness of token-level optimization, and further investigate the scaling of training data and iterations. (3) Text-to-3D Benchmarks: Since existing benchmarks fail to measure implicit reasoning abilities in 3D generation models, we introduce MME-3DR. (4) Advanced RL paradigms: Motivated by the natural hierarchy of 3D generation, we propose Hi-GRPO, which optimizes the global-to-local hierarchical 3D generation through dedicated reward ensembles. Based on these insights, we develop AR3D-R1, the first RL-enhanced text-to-3D model, expert from coarse shape to texture refinement. We hope this study provides insights into RL-driven reasoning for 3D generation. Code is released at https://github.com/Ivan-Tang-3D/3DGen-R1.

cs.CV↗

arXiv2025

Semantic Property Maps for Driving Applications

We consider the problem of estimating the parameters of a vehicle dynamics model for predictive control in driving applications. Instead of solely using the instantaneous parameters estimated from the vehicle signals, we combine this with cameras and update a probabilistic map with parameter estimates and semantic information using Bayesian moment matching. Key to this approach is the map representation, which is constructed with conjugate priors to the measurement likelihoods and defined in the same path coordinates as the vehicle controller, such that the map can be externalized to provide a local representation of the parameter likelihoods that vary in space. The result is a spatial map of vehicle parameters adapted online to enhance the driving control system. We provide theoretical guarantees on the smoothness of relevant parameter likelihood statistics as a function of space, which is critical for their use in predictive control.

eess.SY↗

arXiv2025

What Types of Code Review Comments Do Developers Most Frequently Resolve?

Large language model (LLM)-powered code review automation tools have been introduced to generate code review comments. However, not all generated comments will drive code changes. Understanding what types of generated review comments are likely to trigger code changes is crucial for identifying those that are actionable. In this paper, we set out to investigate (1) the types of review comments written by humans and LLMs, and (2) the types of generated comments that are most frequently resolved by developers. To do so, we developed an LLM-as-a-Judge to automatically classify review comments based on our own taxonomy of five categories. Our empirical study confirms that (1) the LLM reviewer and human reviewers exhibit distinct strengths and weaknesses depending on the project context, and (2) readability, bugs, and maintainability-related comments had higher resolution rates than those focused on code design. These results suggest that a substantial proportion of LLM-generated comments are actionable and can be resolved by developers. Our work highlights the complementarity between LLM and human reviewers and offers suggestions to improve the practical effectiveness of LLM-powered code review tools.

cs.SE↗

arXiv2025

CrossLMM: Decoupling Long Video Sequences from LMMs via Dual Cross-Attention Mechanisms

The advent of Large Multimodal Models (LMMs) has significantly enhanced Large Language Models (LLMs) to process and interpret diverse data modalities (e.g., image and video). However, as input complexity increases, particularly with long video sequences, the number of required tokens has grown significantly, leading to quadratically computational costs. This has made the efficient compression of video tokens in LMMs, while maintaining performance integrity, a pressing research challenge. In this paper, we introduce CrossLMM, decoupling long video sequences from LMMs via a dual cross-attention mechanism, which substantially reduces visual token quantity with minimal performance degradation. Specifically, we first implement a significant token reduction from pretrained visual encoders through a pooling methodology. Then, within LLM layers, we employ a visual-to-visual cross-attention mechanism, wherein the pooled visual tokens function as queries against the original visual token set. This module enables more efficient token utilization while retaining fine-grained informational fidelity. In addition, we introduce a text-to-visual cross-attention mechanism, for which the text tokens are enhanced through interaction with the original visual tokens, enriching the visual comprehension of the text tokens. Comprehensive empirical evaluation demonstrates that our approach achieves comparable or superior performance across diverse video-based LMM benchmarks, despite utilizing substantially fewer computational resources.

cs.CV↗

arXiv2025

SciVerse: Unveiling the Knowledge Comprehension and Visual Reasoning of LMMs on Multi-modal Scientific Problems

The rapid advancement of Large Multi-modal Models (LMMs) has enabled their application in scientific problem-solving, yet their fine-grained capabilities remain under-explored. In this paper, we introduce SciVerse, a multi-modal scientific evaluation benchmark to thoroughly assess LMMs across 5,735 test instances in five distinct versions. We aim to investigate three key dimensions of LMMs: scientific knowledge comprehension, multi-modal content interpretation, and Chain-of-Thought (CoT) reasoning. To unveil whether LMMs possess sufficient scientific expertise, we first transform each problem into three versions containing different levels of knowledge required for solving, i.e., Knowledge-free, -lite, and -rich. Then, to explore how LMMs interpret multi-modal scientific content, we annotate another two versions, i.e., Vision-rich and -only, marking more question information from texts to diagrams. Comparing the results of different versions, SciVerse systematically examines the professional knowledge stock and visual perception skills of LMMs in scientific domains. In addition, to rigorously assess CoT reasoning, we propose a new scientific CoT evaluation strategy, conducting a step-wise assessment on knowledge and logical errors in model outputs. Our extensive evaluation of different LMMs on SciVerse reveals critical limitations in their scientific proficiency and provides new insights into future developments. Project page: https://sciverse-cuhk.github.io

cs.CV↗

arXiv2025

Exploring the Potential of Encoder-free Architectures in 3D LMMs

Encoder-free architectures have been preliminarily explored in the 2D Large Multimodal Models (LMMs), yet it remains an open question whether they can be effectively applied to 3D understanding scenarios. In this paper, we present the first comprehensive investigation into the potential of encoder-free architectures to alleviate the challenges of encoder-based 3D LMMs. These long-standing challenges include the failure to adapt to varying point cloud resolutions during inference and the point features from the encoder not meeting the semantic needs of Large Language Models (LLMs). We identify key aspects for 3D LMMs to remove the pre-trained encoder and enable the LLM to assume the role of the 3D encoder: 1) We propose the LLM-embedded Semantic Encoding strategy in the pre-training stage, exploring the effects of various point cloud self-supervised losses. And we present the Hybrid Semantic Loss to extract high-level semantics. 2) We introduce the Hierarchical Geometry Aggregation strategy in the instruction tuning stage. This incorporates inductive bias into the LLM layers to focus on the local details of the point clouds. To the end, we present the first Encoder-free 3D LMM, ENEL. Our 7B model rivals the state-of-the-art model, PointLLM-PiSA-13B, achieving 57.91%, 61.0%, and 55.20% on the classification, captioning, and VQA tasks, respectively. Our results show that the encoder-free architecture is highly promising for replacing encoder-based architectures in the field of 3D understanding. The code is released at https://github.com/Ivan-Tang-3D/ENEL

cs.CV↗

arXiv2024

SLAM assisted 3D tracking system for laparoscopic surgery

A major limitation of minimally invasive surgery is the difficulty in accurately locating the internal anatomical structures of the target organ due to the lack of tactile feedback and transparency. Augmented reality (AR) offers a promising solution to overcome this challenge. Numerous studies have shown that combining learning-based and geometric methods can achieve accurate preoperative and intraoperative data registration. This work proposes a real-time monocular 3D tracking algorithm for post-registration tasks. The ORB-SLAM2 framework is adopted and modified for prior-based 3D tracking. The primitive 3D shape is used for fast initialization of the monocular SLAM. A pseudo-segmentation strategy is employed to separate the target organ from the background for tracking purposes, and the geometric prior of the 3D shape is incorporated as an additional constraint in the pose graph. Experiments from in-vivo and ex-vivo tests demonstrate that the proposed 3D tracking system provides robust 3D tracking and effectively handles typical challenges such as fast motion, out-of-field-of-view scenarios, partial visibility, and "organ-background" relative motion.

cs.RO↗

arXiv2024

Correspondence-Free SE(3) Point Cloud Registration in RKHS via Unsupervised Equivariant Learning

This paper introduces a robust unsupervised SE(3) point cloud registration method that operates without requiring point correspondences. The method frames point clouds as functions in a reproducing kernel Hilbert space (RKHS), leveraging SE(3)-equivariant features for direct feature space registration. A novel RKHS distance metric is proposed, offering reliable performance amidst noise, outliers, and asymmetrical data. An unsupervised training approach is introduced to effectively handle limited ground truth data, facilitating adaptation to real datasets. The proposed method outperforms classical and supervised methods in terms of registration accuracy on both synthetic (ModelNet40) and real-world (ETH3D) noisy, outlier-rich datasets. To our best knowledge, this marks the first instance of successful real RGB-D odometry data registration using an equivariant method. The code is available at {https://sites.google.com/view/eccv24-equivalign}

cs.CV↗

arXiv2024

MR-MLLM: Mutual Reinforcement of Multimodal Comprehension and Vision Perception

In recent years, multimodal large language models (MLLMs) have shown remarkable capabilities in tasks like visual question answering and common sense reasoning, while visual perception models have made significant strides in perception tasks, such as detection and segmentation. However, MLLMs mainly focus on high-level image-text interpretations and struggle with fine-grained visual understanding, and vision perception models usually suffer from open-world distribution shifts due to their limited model capacity. To overcome these challenges, we propose the Mutually Reinforced Multimodal Large Language Model (MR-MLLM), a novel framework that synergistically enhances visual perception and multimodal comprehension. First, a shared query fusion mechanism is proposed to harmonize detailed visual inputs from vision models with the linguistic depth of language models, enhancing multimodal comprehension and vision perception synergistically. Second, we propose the perception-enhanced cross-modal integration method, incorporating novel modalities from vision perception outputs, like object detection bounding boxes, to capture subtle visual elements, thus enriching the understanding of both visual and textual data. In addition, an innovative perception-embedded prompt generation mechanism is proposed to embed perceptual information into the language model's prompts, aligning the responses contextually and perceptually for a more accurate multimodal interpretation. Extensive experiments demonstrate MR-MLLM's superior performance in various multimodal comprehension and vision perception tasks, particularly those requiring corner case vision perception and fine-grained language comprehension.

cs.CV↗

arXiv2024

Any2Point: Empowering Any-modality Large Models for Efficient 3D Understanding

Large foundation models have recently emerged as a prominent focus of interest, attaining superior performance in widespread scenarios. Due to the scarcity of 3D data, many efforts have been made to adapt pre-trained transformers from vision to 3D domains. However, such 2D-to-3D approaches are still limited, due to the potential loss of spatial geometries and high computation cost. More importantly, their frameworks are mainly designed for 2D models, lacking a general any-to-3D paradigm. In this paper, we introduce Any2Point, a parameter-efficient method to empower any-modality large models (vision, language, audio) for 3D understanding. Given a frozen transformer from any source modality, we propose a 3D-to-any (1D or 2D) virtual projection strategy that correlates the input 3D points to the original 1D or 2D positions within the source modality. This mechanism enables us to assign each 3D token with a positional encoding paired with the pre-trained model, which avoids 3D geometry loss caused by the true projection and better motivates the transformer for 3D learning with 1D/2D positional priors. Then, within each transformer block, we insert an any-to-3D guided adapter module for parameter-efficient fine-tuning. The adapter incorporates prior spatial knowledge from the source modality to guide the local feature aggregation of 3D tokens, compelling the semantic adaption of any-modality transformers. We conduct extensive experiments to showcase the effectiveness and efficiency of our method. Code and models are released at https://github.com/Ivan-Tang-3D/Any2Point.

cs.CV↗

arXiv2024

RKHS-BA: A Robust Correspondence-Free Multi-View Registration Framework with Semantic Point Clouds

This work reports a novel multi-frame Bundle Adjustment (BA) framework called RKHS-BA. It uses continuous landmark representations that encode RGB-D/LiDAR and semantic observations in a Reproducing Kernel Hilbert Space (RKHS). With a correspondence-free pose graph formulation, the proposed system constructs a loss function that achieves more generalized convergence than classical point-wise convergence. We demonstrate its applications in multi-view point cloud registration, sliding-window odometry, and global LiDAR mapping on simulated and real data. It shows highly robust pose estimations in extremely noisy scenes and exhibits strong generalization with various types of semantic inputs. The open source implementation is released in https://github.com/UMich-CURLY/RKHS_BA.

cs.RO↗

arXiv2023

Cloud-Device Collaborative Learning for Multimodal Large Language Models

The burgeoning field of Multimodal Large Language Models (MLLMs) has exhibited remarkable performance in diverse tasks such as captioning, commonsense reasoning, and visual scene understanding. However, the deployment of these large-scale MLLMs on client devices is hindered by their extensive model parameters, leading to a notable decline in generalization capabilities when these models are compressed for device deployment. Addressing this challenge, we introduce a Cloud-Device Collaborative Continual Adaptation framework, designed to enhance the performance of compressed, device-deployed MLLMs by leveraging the robust capabilities of cloud-based, larger-scale MLLMs. Our framework is structured into three key components: a device-to-cloud uplink for efficient data transmission, cloud-based knowledge adaptation, and an optimized cloud-to-device downlink for model deployment. In the uplink phase, we employ an Uncertainty-guided Token Sampling (UTS) strategy to effectively filter out-of-distribution tokens, thereby reducing transmission costs and improving training efficiency. On the cloud side, we propose Adapter-based Knowledge Distillation (AKD) method to transfer refined knowledge from large-scale to compressed, pocket-size MLLMs. Furthermore, we propose a Dynamic Weight update Compression (DWC) strategy for the downlink, which adaptively selects and quantizes updated weight parameters, enhancing transmission efficiency and reducing the representational disparity between cloud and device models. Extensive experiments on several multimodal benchmarks demonstrate the superiority of our proposed framework over prior Knowledge Distillation and device-cloud collaboration methods. Notably, we also validate the feasibility of our approach to real-world experiments.

cs.CV↗

arXiv2023

BDIS-SLAM: A lightweight CPU-based dense stereo SLAM for surgery

Purpose: Common dense stereo Simultaneous Localization and Mapping (SLAM) approaches in Minimally Invasive Surgery (MIS) require high-end parallel computational resources for real-time implementation. Yet, it is not always feasible since the computational resources should be allocated to other tasks like segmentation, detection, and tracking. To solve the problem of limited parallel computational power, this research aims at a lightweight dense stereo SLAM system that works on a single-core CPU and achieves real-time performance (more than 30 Hz in typical scenarios). Methods: A new dense stereo mapping module is integrated with the ORB-SLAM2 system and named BDIS-SLAM. Our new dense stereo mapping module includes stereo matching and 3D dense depth mosaic methods. Stereo matching is achieved with the recently proposed CPU-level real-time matching algorithm Bayesian Dense Inverse Searching (BDIS). A BDIS-based shape recovery and a depth mosaic strategy are integrated as a new thread and coupled with the backbone ORB-SLAM2 system for real-time stereo shape recovery. Results: Experiments on in-vivo data sets show that BDIS-SLAM runs at over 30 Hz speed on modern single-core CPU in typical endoscopy/colonoscopy scenarios. BDIS-SLAM only consumes around an additional 12% time compared with the backbone ORB-SLAM2. Although our lightweight BDIS-SLAM simplifies the process by ignoring deformation and fusion procedures, it can provide a usable dense mapping for modern MIS on computationally constrained devices. Conclusion: The proposed BDIS-SLAM is a lightweight stereo dense SLAM system for MIS. It achieves 30 Hz on a modern single-core CPU in typical endoscopy/colonoscopy scenarios (image size around 640*480). BDIS-SLAM provides a low-cost solution for dense mapping in MIS and has the potential to be applied in surgical robots and AR systems.

cs.RO↗