Haotian Xue

Yue Wang, Wenjie Deng, Haotian Xue et al.

Intraocular foreign body removal demands millimeter-level precision in confined intraocular spaces, yet existing robotic systems predominantly rely on manual teleoperation with steep learning curves. To address the challenges of autonomous manipulation, particularly kinematic uncertainties from variable motion scaling and Remote Center of Motion (RCM) point variation, we propose RCM-ACT, an imitation learning framework for autonomous intraocular foreign body ring manipulation. Our approach integrates RCM dynamic calibration to resolve coordinate system inconsistencies caused by intraocular instrument variation and introduces the RCM-ACT architecture, which combines action chunking transformers with episode-level kinematic realignment. Trained solely on stereo visual data and instrument kinematics from expert demonstrations in an artificial eye model, RCM-ACT successfully completes ring grasping and positioning tasks without explicit depth sensing. Experimental validation demonstrates the successful implementation of end-to-end autonomy under uncalibrated microscopy conditions, achieving a mean 3-D Euclidean grasp deviation of 0.686 mm and 11/20 full-task successes. The results provide a viable framework for developing intelligent eye surgical systems capable of complex intraocular procedures.

Haotian Xue, Chumeng Liang, Xiaoyu Wu et al. · gatech

While generative diffusion models excel in producing high-quality images, they can also be misused to mimic authorized images, posing a significant threat to AI systems. Efforts have been made to add calibrated perturbations to protect images from diffusion-based mimicry pipelines. However, most of the existing methods are too ineffective and even impractical to be used by individual users due to their high computation and memory requirements. In this work, we present novel findings on attacking latent diffusion models (LDM) and propose new plug-and-play strategies for more effective protection. In particular, we explore the bottleneck in attacking an LDM, discovering that the encoder module rather than the denoiser module is the vulnerable point. Based on this insight, we present our strategy using Score Distillation Sampling (SDS) to double the speed of protection and reduce memory occupation by half without compromising its strength. Additionally, we provide a robust protection strategy by counterintuitively minimizing the semantic loss, which can assist in generating more natural perturbations. Finally, we conduct extensive experiments to substantiate our findings and comprehensively evaluate our newly proposed strategies. We hope our insights and protective measures can contribute to better defense against malicious diffusion-based mimicry, advancing the development of secure AI systems. The code is available in https://github.com/xavihart/Diff-Protect

Haotian Xue, Antonio Torralba, Joshua B. Tenenbaum et al. · gatech

Given a visual scene, humans have strong intuitions about how a scene can evolve over time under given actions. The intuition, often termed visual intuitive physics, is a critical ability that allows us to make effective plans to manipulate the scene to achieve desired outcomes without relying on extensive trial and error. In this paper, we present a framework capable of learning 3D-grounded visual intuitive physics models from videos of complex scenes with fluids. Our method is composed of a conditional Neural Radiance Field (NeRF)-style visual frontend and a 3D point-based dynamics prediction backend, using which we can impose strong relational and structural inductive bias to capture the structure of the underlying environment. Unlike existing intuitive point-based dynamics works that rely on the supervision of dense point trajectory from simulators, we relax the requirements and only assume access to multi-view RGB images and (imperfect) instance masks acquired using color prior. This enables the proposed model to handle scenarios where accurate point estimation and tracking are hard or impossible. We generate datasets including three challenging scenarios involving fluid, granular materials, and rigid objects in the simulation. The datasets do not include any dense particle information so most previous 3D-based intuitive physics pipelines can barely deal with that. We show our model can make long-horizon future predictions by learning from raw images and significantly outperforms models that do not employ an explicit 3D representation space. We also show that once trained, our model can achieve strong generalization in complex scenarios under extrapolate settings.

Shengyuan Hou, Jushi Kai, Haotian Xue et al. · gatech, meta-ai

Recent works have revealed that Transformers are implicitly learning the syntactic information in its lower layers from data, albeit is highly dependent on the quality and scale of the training data. However, learning syntactic information from data is not necessary if we can leverage an external syntactic parser, which provides better parsing quality with well-defined syntactic structures. This could potentially improve Transformer's performance and sample efficiency. In this work, we propose a syntax-guided localized self-attention for Transformer that allows directly incorporating grammar structures from an external constituency parser. It prohibits the attention mechanism to overweight the grammatically distant tokens over close ones. Experimental results show that our model could consistently improve translation performance on a variety of machine translation datasets, ranging from small to large dataset sizes, and with different source languages.

Zeqian Long, Ozgur Kara, Haotian Xue et al. · gatech

Image-to-video (I2V) generation has the potential for societal harm because it enables the unauthorized animation of static images to create realistic deepfakes. While existing defenses effectively protect against static image manipulation, extending these to I2V generation remains underexplored and non-trivial. In this paper, we systematically analyze why modern I2V models are highly robust against naive image-level adversarial attacks (i.e., immunization). We observe that the video encoding process rapidly dilutes the adversarial noise across future frames, and the continuous text-conditioned guidance actively overrides the intended disruptive effect of the immunization. Building on these findings, we propose the Immune2V framework which enforces temporally balanced latent divergence at the encoder level to prevent signal dilution, and aligns intermediate generative representations with a precomputed collapse-inducing trajectory to counteract the text-guidance override. Extensive experiments demonstrate that Immune2V produces substantially stronger and more persistent degradation than adapted image-level baselines under the same imperceptibility budget.

Lama Moukheiber, Caleb M. Yeung, Haotian Xue et al. · gatech

Spatial reasoning and visual grounding are core capabilities for vision-language models (VLMs), yet most medical VLMs produce predictions without transparent reasoning or spatial evidence. Existing benchmarks also evaluate VLMs on isolated 2D images, overlooking the volumetric nature of clinical imaging, where findings can span multiple frames or appear on only a few slices. We introduce Spatially Grounded MRI Visual Question Answering (SGMRI-VQA), a 41,307-pair benchmark for multi-frame, spatially grounded reasoning on volumetric MRI. Built from expert radiologist annotations in the fastMRI+ dataset across brain and knee studies, each QA pair includes a clinician-aligned chain-of-thought trace with frame-indexed bounding box coordinates. Tasks are organized hierarchically across detection, localization, counting/classification, and captioning, requiring models to jointly reason about what is present, where it is, and across which frames it extends. We benchmark 10 VLMs and show that supervised fine-tuning of Qwen3-VL-8B with bounding box supervision consistently improves grounding performance over strong zero-shot baselines, indicating that targeted spatial supervision is an effective path toward grounded clinical reasoning.

Zelin Zhao, Petr Molodyk, Haotian Xue et al. · gatech

In this paper, we present Laplacian multiscale flow matching (LapFlow), a novel framework that enhances flow matching by leveraging multi-scale representations for image generative modeling. Our approach decomposes images into Laplacian pyramid residuals and processes different scales in parallel through a mixture-of-transformers (MoT) architecture with causal attention mechanisms. Unlike previous cascaded approaches that require explicit renoising between scales, our model generates multi-scale representations in parallel, eliminating the need for bridging processes. The proposed multi-scale architecture not only improves generation quality but also accelerates the sampling process and promotes scaling flow matching methods. Through extensive experimentation on CelebA-HQ and ImageNet, we demonstrate that our method achieves superior sample quality with fewer GFLOPs and faster inference compared to single-scale and multi-scale flow matching baselines. The proposed model scales effectively to high-resolution generation (up to 1024$\times$1024) while maintaining lower computational overhead.

Haotian Xue, Qi Chen, Zhonghao Wang et al.

Video diffusion models achieve strong frame-level fidelity but still struggle with motion coherence, dynamics and realism, often producing jitter, ghosting, or implausible dynamics. A key limitation is that the standard denoising MSE objective provides no direct supervision on temporal consistency, allowing models to achieve low loss while still generating poor motion. We propose MoGAN, a motion-centric post-training framework that improves motion realism without reward models or human preference data. Built atop a 3-step distilled video diffusion model, we train a DiT-based optical-flow discriminator to differentiate real from generated motion, combined with a distribution-matching regularizer to preserve visual fidelity. With experiments on Wan2.1-T2V-1.3B, MoGAN substantially improves motion quality across benchmarks. On VBench, MoGAN boosts motion score by +7.3% over the 50-step teacher and +13.3% over the 3-step DMD model. On VideoJAM-Bench, MoGAN improves motion score by +7.4% over the teacher and +8.8% over DMD, while maintaining comparable or even better aesthetic and image-quality scores. A human study further confirms that MoGAN is preferred for motion quality (52% vs. 38% for the teacher; 56% vs. 29% for DMD). Overall, MoGAN delivers significantly more realistic motion without sacrificing visual fidelity or efficiency, offering a practical path toward fast, high-quality video generation. Project webpage is: https://xavihart.github.io/mogan.

Haotian Xue, Yongxin Chen · gatech

Adversarial examples for diffusion models are widely used as solutions for safety concerns. By adding adversarial perturbations to personal images, attackers can not edit or imitate them easily. However, it is essential to note that all these protections target the latent diffusion model (LDMs), the adversarial examples for diffusion models in the pixel space (PDMs) are largely overlooked. This may mislead us to think that the diffusion models are vulnerable to adversarial attacks like most deep models. In this paper, we show novel findings that: even though gradient-based white-box attacks can be used to attack the LDMs, they fail to attack PDMs. This finding is supported by extensive experiments of almost a wide range of attacking methods on various PDMs and LDMs with different model structures, which means diffusion models are indeed much more robust against adversarial attacks. We also find that PDMs can be used as an off-the-shelf purifier to effectively remove the adversarial patterns that were generated on LDMs to protect the images, which means that most protection methods nowadays, to some extent, cannot protect our images from malicious attacks. We hope that our insights will inspire the community to rethink the adversarial samples for diffusion models as protection methods and move forward to more effective protection. Codes are available in https://github.com/xavihart/PDM-Pure.

Haotian Xue, Yunhao Ge, Yu Zeng et al. · gatech

Vision-Language Models (VLMs) have demonstrated impressive world knowledge across a wide range of tasks, making them promising candidates for embodied reasoning applications. However, existing benchmarks primarily evaluate the embodied reasoning ability of VLMs through multiple-choice questions based on image annotations -- for example, selecting which trajectory better describes an event in the image. In this work, we introduce the Point-It-Out (PIO) benchmark, a novel benchmark designed to systematically assess the embodied reasoning abilities of VLMs through precise visual grounding. We propose a hierarchical evaluation protocol spanning three stages (S1: referred-object localization, S2: task-driven pointing, and S3: visual trace prediction), with data collected from critical domains for embodied intelligence, including indoor, kitchen, driving, and robotic manipulation scenarios. Extensive experiments with over ten state-of-the-art VLMs reveal several interesting findings. For example, strong general-purpose models such as GPT-4o, while excelling on many benchmarks (e.g., language, perception, and reasoning), underperform compared to some open-source models in precise visual grounding; models such as MoLMO perform well in S1 and S2 but struggle in S3, where requires grounding combined with visual trace planning.

Haotian Xue, Alexandre Araujo, Bin Hu et al.

Neural networks are known to be susceptible to adversarial samples: small variations of natural examples crafted to deliberately mislead the models. While they can be easily generated using gradient-based techniques in digital and physical scenarios, they often differ greatly from the actual data distribution of natural images, resulting in a trade-off between strength and stealthiness. In this paper, we propose a novel framework dubbed Diffusion-Based Projected Gradient Descent (Diff-PGD) for generating realistic adversarial samples. By exploiting a gradient guided by a diffusion model, Diff-PGD ensures that adversarial samples remain close to the original data distribution while maintaining their effectiveness. Moreover, our framework can be easily customized for specific tasks such as digital attacks, physical-world attacks, and style-based attacks. Compared with existing methods for generating natural-style adversarial samples, our framework enables the separation of optimizing adversarial loss from other surrogate losses (e.g., content/smoothness/style loss), making it more stable and controllable. Finally, we demonstrate that the samples generated using Diff-PGD have better transferability and anti-purification power than traditional gradient-based methods. Code will be released in https://github.com/xavihart/Diff-PGD

Haotian Xue, Yipu Chen, Liqian Ma et al.

Action-conditioned world models (ACWMs) have shown strong promise for video prediction and decision-making. However, existing benchmarks are largely restricted to egocentric navigation or narrow, task-specific robotics datasets, offering only limited coverage of the rich physical interactions required for generalized world understanding. We introduce ACWM-Phys, a new benchmark for evaluating action-conditioned prediction under diverse physical dynamics in a clean, controllable simulation environment with a carefully designed action space. ACWM-Phys contains training and evaluation data spanning rigid-body dynamics, kinematics, deformable-object interactions, and particle dynamics. To evaluate both interpolation and generalization, we design in-distribution and out-of-distribution protocols with controlled shifts in interaction patterns or scene configurations. By building the benchmark in a fully controllable simulator, ACWM-Phys enables precise data collection, reproducible evaluation, and systematic analysis of model capabilities for physically grounded world modeling. Through systematic experiments on ACWM-DiT, we find that OoD generalization depends not only on the physical regime but also on effective task complexity: models generalize well on visually simple, low-dimensional interactions with clear geometric structure, but suffer larger drops on deformable contacts, high-dimensional control, and complex articulated motion. This suggests that the model still relies heavily on visual appearance patterns instead of fully learning the underlying physics. Ablations show that cross-attention improves high-dimensional action conditioning, causal VAEs outperform frame-wise encoders, and larger action spaces are harder to model but can improve generalization by providing richer control signals. These findings guide the design of physically grounded world models.

Zirong Chen, Haotian Xue

Due to the unfamiliarity to particular words(or proper nouns) for ingredients, non-native English speakers can be extremely confused about the ordering process in restaurants like Subway. Thus, We developed a dialogue system, which supports Chinese(Mandarin)1 and English2 at the same time. In other words, users can switch arbitrarily between Chinese(Mandarin) and English as the conversation is being conducted. This system is specifically designed for Subway ordering3. In BilDOS, we designed a Discriminator module to tell the language is being used in inputted user utterance, a Translator module to translate used language into English if it is not English, and a Dialogue Manager module to detect the intention within inputted user utterances, handle outlier inputs by throwing clarification requests, map detected Intention and detailed Keyword4 into a particular intention class, locate the current ordering process, continue to give queries to finish the order, conclude the order details once the order is completed, activate the evaluation process when the conversation is done.

Haotian Xue, Kaixiong Zhou, Tianlong Chen et al.

Despite the recent advances of graph neural networks (GNNs) in modeling graph data, the training of GNNs on large datasets is notoriously hard due to the overfitting. Adversarial training, which augments data with the worst-case adversarial examples, has been widely demonstrated to improve model's robustness against adversarial attacks and generalization ability. However, while the previous adversarial training generally focuses on protecting GNNs from spiteful attacks, it remains unclear how the adversarial training could improve the generalization abilities of GNNs in the graph analytics problem. In this paper, we investigate GNNs from the lens of weight and feature loss landscapes, i.e., the loss changes with respect to model weights and node features, respectively. We draw the conclusion that GNNs are prone to falling into sharp local minima in these two loss landscapes, where GNNs possess poor generalization performances. To tackle this problem, we construct the co-adversarial perturbation (CAP) optimization problem in terms of weights and features, and design the alternating adversarial perturbation algorithm to flatten the weight and feature loss landscapes alternately. Furthermore, we divide the training process into two stages: one conducting the standard cross-entropy minimization to ensure the quick convergence of GNN models, the other applying our alternating adversarial training to avoid falling into locally sharp minima. The extensive experiments demonstrate our CAP can generally improve the generalization performance of GNNs on a variety of benchmark graph datasets.

Xu Cheng, Xin Wang, Haotian Xue et al.

This paper proposes a hypothesis for the aesthetic appreciation that aesthetic images make a neural network strengthen salient concepts and discard inessential concepts. In order to verify this hypothesis, we use multi-variate interactions to represent salient concepts and inessential concepts contained in images. Furthermore, we design a set of operations to revise images towards more beautiful ones. In experiments, we find that the revised images are more aesthetic than the original ones to some extent.

Hao Zhang, Jiayi Chen, Haotian Xue et al.

This paper proposes a set of criteria to evaluate the objectiveness of explanation methods of neural networks, which is crucial for the development of explainable AI, but it also presents significant challenges. The core challenge is that people usually cannot obtain ground-truth explanations of the neural network. To this end, we design four metrics to evaluate explanation results without ground-truth explanations. Our metrics can be broadly applied to nine benchmark methods of interpreting neural networks, which provides new insights of explanation methods.