Yicheng Jiang

Yicheng Jiang, Jiaxu Wang, Junhao He et al.

Current 3D-aware pretraining methods for embodied perception and manipulation are largely built on differentiable rendering frameworks, producing either fully implicit neural fields or fully explicit geometric primitives. Implicit representations, while expressive, lack explicit structural cues, whereas explicit ones preserve geometry but suffer from resolution limits and weak generalization. To address these limitations, we propose a novel pretraining framework that learns a hybrid representation-structural latent points. Specifically, we insert a point-wise latent variational autoencoder into the latent space of a point-cloud autoencoder, jointly regularizing point-wise features and coordinates toward a Gaussian prior. The resulting compact latent preserves coarse structural tendencies, which do not encode precise geometry but capture richer rough shape and semantic information, effectively combining the expressiveness of implicit representations with the structural priors of explicit ones. In addition, informed by shared design choices in prior work, we develop a streamlined, efficient 3DGS-based rendering pipeline that is deliberately kept lightweight, improving efficiency while leaving greater representational capacity to the front-end latent module. Extensive evaluations on RLBench, ManiSkill2, and a real-robot platform demonstrate consistent gains in task success, sample efficiency, and robustness to viewpoint and scene variations over strong baselines. Ablation studies further confirm that each component of our framework is critical to overall performance.

Jiaxu Wang, Yicheng Jiang, Tianlun He et al.

World-model-based imagine-then-act becomes a promising paradigm for robotic manipulation, yet existing approaches typically support either purely image-based forecasting or reasoning over partial 3D geometry, limiting their ability to predict complete 4D scene dynamics. This work proposes a novel embodied 4D world model that enables geometrically consistent, arbitrary-view RGBD generation: given only a single-view RGBD observation as input, the model imagines the remaining viewpoints, which can then be back-projected and fused to assemble a more complete 3D structure across time. To efficiently learn the multi-view, cross-modality generation, we explicitly design cross-view and cross-modality feature fusion that jointly encourage consistency between RGB and depth and enforce geometric alignment across views. Beyond prediction, converting generated futures into actions is often handled by inverse dynamics, which is ill-posed because multiple actions can explain the same transition. We address this with a test-time action optimization strategy that backpropagates through the generative model to infer a trajectory-level latent best matching the predicted future, and a residual inverse dynamics model that turns this trajectory prior into accurate executable actions. Experiments on three datasets demonstrate strong performance on both 4D scene generation and downstream manipulation, and ablations provide practical insights into the key design choices.

Jeffrey Joan Sam, Janhavi Sathe, Nikhil Chigali et al.

Spacecraft deployed in outer space are routinely subjected to various forms of damage due to exposure to hazardous environments. In addition, there are significant risks to the subsequent process of in-space repairs through human extravehicular activity or robotic manipulation, incurring substantial operational costs. Recent developments in image segmentation could enable the development of reliable and cost-effective autonomous inspection systems. While these models often require large amounts of training data to achieve satisfactory results, publicly available annotated spacecraft segmentation data are very scarce. Here, we present a new dataset of nearly 64k annotated spacecraft images that was created using real spacecraft models, superimposed on a mixture of real and synthetic backgrounds generated using NASA's TTALOS pipeline. To mimic camera distortions and noise in real-world image acquisition, we also added different types of noise and distortion to the images. Finally, we finetuned YOLOv8 and YOLOv11 segmentation models to generate performance benchmarks for the dataset under well-defined hardware and inference time constraints to mimic real-world image segmentation challenges for real-time onboard applications in space on NASA's inspector spacecraft. The resulting models, when tested under these constraints, achieved a Dice score of 0.92, Hausdorff distance of 0.69, and an inference time of about 0.5 second. The dataset and models for performance benchmark are available at https://github.com/RiceD2KLab/SWiM.

Boyu Shi, YiCheng Jiang, Chang Liu et al.

Large language models (LLMs) achieve strong performance but remain costly to deploy in resource-constrained settings. Training small language models (SLMs) from scratch is computationally expensive, while conventional knowledge distillation requires repeated access to large teachers for different target sizes, leading to poor scalability. To solve these problems, we propose \textbf{Chain-based Distillation (CBD)}, a scalable paradigm for efficiently initializing variable-sized language models. A sparse and limited sequence of intermediate models (called anchors) is constructed via stepwise distillation, forming a distillation chain that progressively transfers knowledge from the source LLMs. To support heterogeneous settings, we introduce \emph{bridge distillation} for cross-architecture and cross-vocabulary transfer. Models of variable sizes are initialized via parameter interpolation between adjacent anchors, eliminating repeated large teacher inference. Experiments show that the proposed method substantially improves efficiency and downstream performance. A 138M-parameter SLM without recovery pre-training, outperforms scratch-trained models on a 10B-token corpus on the specific task. CBD also demonstrates versatility in heterogeneous settings for initialize models with different architectures and vocabularies.

Xinhao Liu, Jintong Li, Yicheng Jiang et al.

Navigating dynamic urban environments presents significant challenges for embodied agents, requiring advanced spatial reasoning and adherence to common-sense norms. Despite progress, existing visual navigation methods struggle in map-free or off-street settings, limiting the deployment of autonomous agents like last-mile delivery robots. To overcome these obstacles, we propose a scalable, data-driven approach for human-like urban navigation by training agents on thousands of hours of in-the-wild city walking and driving videos sourced from the web. We introduce a simple and scalable data processing pipeline that extracts action supervision from these videos, enabling large-scale imitation learning without costly annotations. Our model learns sophisticated navigation policies to handle diverse challenges and critical scenarios. Experimental results show that training on large-scale, diverse datasets significantly enhances navigation performance, surpassing current methods. This work shows the potential of using abundant online video data to develop robust navigation policies for embodied agents in dynamic urban settings. Project homepage is at https://ai4ce.github.io/CityWalker/.

Sean Wang, Yicheng Jiang, Yuxin Tang et al.

Uncertainty Quantification (UQ) for Natural Language Generation (NLG) is crucial for assessing the performance of Large Language Models (LLMs), as it reveals confidence in predictions, identifies failure modes, and gauges output reliability. Conformal Prediction (CP), a model-agnostic method that generates prediction sets with a specified error rate, has been adopted for UQ in classification tasks, where the size of the prediction set indicates the model's uncertainty. However, when adapting CP to NLG, the sampling-based method for generating candidate outputs cannot guarantee the inclusion of the ground truth, limiting its applicability across a wide range of error rates. To address this, we propose \ourmethod, a method that explicitly adds the ground truth to the candidate outputs and uses logit scores to measure nonconformity. Our experiments with six LLMs on four NLG tasks show that \ourmethod outperforms baseline methods in calibrating error rates and empirical cover rates, offering accurate UQ across a wide range of user-specified error rates.

Yicheng Jiang, Jin Yuan, Hua Yuan et al.

Conditional image generation models have achieved remarkable results by leveraging text-based control to generate customized images. However, the high resource demands of these models and the scarcity of well-annotated data have hindered their deployment on edge devices, leading to enormous costs and privacy concerns, especially when user data is sent to a third party. To overcome these challenges, we propose Refine-Control, a semi-supervised distillation framework. Specifically, we improve the performance of the student model by introducing a tri-level knowledge fusion loss to transfer different levels of knowledge. To enhance generalization and alleviate dataset scarcity, we introduce a semi-supervised distillation method utilizing both labeled and unlabeled data. Our experiments reveal that Refine-Control achieves significant reductions in computational cost and latency, while maintaining high-fidelity generation capabilities and controllability, as quantified by comparative metrics.

Hua Yuan, Jin Yuan, Yicheng Jiang et al.

Object removal, a challenging task within image inpainting, involves seamlessly filling the removed region with content that matches the surrounding context. Despite advancements in diffusion models, current methods still face two critical challenges. The first is mask hallucination, where the model generates irrelevant or spurious content inside the masked region, and the second is mask-shape bias, where the model fills the masked area with an object that mimics the mask's shape rather than surrounding content. To address these issues, we propose Mask Consistency Regularization (MCR), a novel training strategy designed specifically for object removal tasks. During training, our approach introduces two mask perturbations: dilation and reshape, enforcing consistency between the outputs of these perturbed branches and the original mask. The dilated masks help align the model's output with the surrounding content, while reshaped masks encourage the model to break the mask-shape bias. This combination of strategies enables MCR to produce more robust and contextually coherent inpainting results. Our experiments demonstrate that MCR significantly reduces hallucinations and mask-shape bias, leading to improved performance in object removal.

Wenting Jiang, Yicheng Jiang, Lu Zhang et al.

Automatic segmentation of hepatocellular carcinoma (HCC) in Digital Subtraction Angiography (DSA) videos can assist radiologists in efficient diagnosis of HCC and accurate evaluation of tumors in clinical practice. Few studies have investigated HCC segmentation from DSA videos. It shows great challenging due to motion artifacts in filming, ambiguous boundaries of tumor regions and high similarity in imaging to other anatomical tissues. In this paper, we raise the problem of HCC segmentation in DSA videos, and build our own DSA dataset. We also propose a novel segmentation network called DSA-LTDNet, including a segmentation sub-network, a temporal difference learning (TDL) module and a liver region segmentation (LRS) sub-network for providing additional guidance. DSA-LTDNet is preferable for learning the latent motion information from DSA videos proactively and boosting segmentation performance. All of experiments are conducted on our self-collected dataset. Experimental results show that DSA-LTDNet increases the DICE score by nearly 4% compared to the U-Net baseline.