arXiv subjects

Phong Nguyen

Publications and source records attributed to Phong Nguyen.

At least 19 recordsLinked to original sources

arXiv2026

Cross-Space Distillation: Teaching One-Step Students with Modern Diffusion Teachers

Modern one-step diffusion models achieve impressive quality through distribution-based timestep distillation. Yet, they rely on a critical assumption: Teacher and Student must inhabit the same latent space. This Shared-Space constraint prevents knowledge transfer from modern high-capacity Teachers (e.g., SD 3.5 and Flux) into compact, deployment-friendly Students such as SD 1.5, whose latent resolution and VAE parameterization differ from the Teacher. We formalize this overlooked regime as Cross-Space Distillation, where Teacher and Student differ in both latent resolution and VAE space. To enable distillation under this mismatch, we introduce the Bridge, a lightweight latent interface that maps Student latents into the Teacher space without modifying the Student backbone. Bridge combines a frozen Student VAE decoder as a spatial prior with a compact learnable projector, and is trained with latent reconstruction and attention fidelity objectives for stable Teacher-space alignment. Across diverse modern Teachers, Bridge enables substantial gains for compact one-step Students; for example, it improves SD 1.5 from 5.4 to 9.4 HPSv3 while preserving one-step inference, low latency, and broad ecosystem compatibility. These results show that heterogeneous large Teachers can be distilled into efficient, deployable backbones through a lightweight latent-space interface.

cs.CV↗

arXiv2026

Threading Optimization for Vision-Language-Action Model Inference in Low-Cost Smart Agricultural Manipulation

Vision-Language Action (VLA) models continue to face challenges such as slow inference speed and difficulty performing fine-grained motion adjustments, limiting their widespread adoption in industry. While the Real-Time Action Chunking (RTAC) algorithm has been proposed to address these bottlenecks, bridging the gap between the algorithm provided in pseudocode to a stable, real-world deployment on a low-cost robotic arm remains a challenge. In this work, we present a complete system-level implementation of RTAC tailored for a low-cost robotic manipulation system. We advance beyond the original high-level pseudocode by optimizing the threading implementation for the policy inference and control pipeline, reducing end-to-end latency and improving responsiveness without modifying the underlying policy. We evaluate this system on tasks involving the manipulation of agricultural produce, specifically garlic bulbs and walnuts. Experimental results demonstrate that our custom threading implementation significantly improves control stability and speed compared to the base implementation of RTAC.

cs.RO↗

arXiv2026

When In-Distribution Gains Fail: Evaluating Weak-to-Strong Reward Models under Preference Shift

Weak-to-strong (W2S) generalization is a promising framework for scalable oversight, yet existing evaluations often test students under matched train-test distributions. Therefore, we study W2S preference learning under zero-shot distribution shift and find that strong students trained on weak preference labels can appear successful in-distribution while failing to transfer across preference datasets. We provide evidence for a representational failure mode in which weak-supervised fine-tuning can pull the strong model toward source-domain features instead of maintaining broadly transferable preference representations. To mitigate this, we propose Representation Anchoring (Anchor), a simple yet effective regularizer that constrains excessive drift from the pretrained strong model's representation space during fine-tuning, while still allowing task-relevant adaptation. Across preference domains, datasets, and model families, Anchor consistently improves out-of-distribution transfer while maintaining competitive in-distribution performance. Together, our evaluation protocol, transfer-aware metrics, and method expose hidden brittleness in current W2S reward modeling and provide a practical path toward more robust preference transfer.

cs.CL↗

arXiv2026

Alice v1: Distillation-Enhanced Video Generation Surpassing Closed-Source Models

Wepresent Alice v1, a 14-billion parameter open-source video generation model that achieves state-of-the-art quality through consistency distillation with score regularization (rCM). Contrary to conventional distillation-which trades quality for speed-we demonstrate that rCM-based distillation can exceed teacher model quality. We attribute this to three mechanisms: (1) the score regularization term acts as a mode-seeking objective that concentrates probability mass on high-quality outputs rather than covering the full teacher distribution, (2) our targeted synthetic data pipeline with hard example mining provides training signal specifically for failure modes (physics, hands, faces) that the teacher handles inconsistently, and (3) consistency enforcement acts as implicit regularization, eliminating "lucky path" dependence on specific noise samples. Alice v1 generates 5-second 720p videos at 24fps in 4 denoising steps (~8 seconds on H100), a 7x speedup over the 50-step teacher while improving VBench score from 84.0 (Wan2.2) to 91.2. This surpasses both the teacher and closed-source systems including Veo3 (~90) and Sora2 (~88) on automated benchmarks, with competitive results in human preference studies. We release all model weights, training code, synthetic data pipelines, and evaluation scripts to advance open research in video generation.

cs.GR↗

arXiv2026

InverFill: One-Step Inversion for Enhanced Few-Step Diffusion Inpainting

Recent diffusion-based models achieve photorealism in image inpainting but require many sampling steps, limiting practical use. Few-step text-to-image models offer faster generation, but naively applying them to inpainting yields poor harmonization and artifacts between the background and inpainted region. We trace this cause to random Gaussian noise initialization, which under low function evaluations causes semantic misalignment and reduced fidelity. To overcome this, we propose InverFill, a one-step inversion method tailored for inpainting that injects semantic information from the input masked image into the initial noise, enabling high-fidelity few-step inpainting. Instead of training inpainting models, InverFill leverages few-step text-to-image models in a blended sampling pipeline with semantically aligned noise as input, significantly improving vanilla blended sampling and even matching specialized inpainting models at low NFEs. Moreover, InverFill does not require real-image supervision and only adds minimal inference overhead. Extensive experiments show that InverFill consistently boosts baseline few-step models, improving image quality and text coherence without costly retraining or heavy iterative optimization.

cs.CV↗

arXiv2026

SwiftTailor: Efficient 3D Garment Generation with Geometry Image Representation

Realistic and efficient 3D garment generation remains a longstanding challenge in computer vision and digital fashion. Existing methods typically rely on large vision- language models to produce serialized representations of 2D sewing patterns, which are then transformed into simulation-ready 3D meshes using garment modeling framework such as GarmentCode. Although these approaches yield high-quality results, they often suffer from slow inference times, ranging from 30 seconds to a minute. In this work, we introduce SwiftTailor, a novel two-stage framework that unifies sewing-pattern reasoning and geometry-based mesh synthesis through a compact geometry image representation. SwiftTailor comprises two lightweight modules: PatternMaker, an efficient vision-language model that predicts sewing patterns from diverse input modalities, and GarmentSewer, an efficient dense prediction transformer that converts these patterns into a novel Garment Geometry Image, encoding the 3D surface of all garment panels in a unified UV space. The final 3D mesh is reconstructed through an efficient inverse mapping process that incorporates remeshing and dynamic stitching algorithms to directly assemble the garment, thereby amortizing the cost of physical simulation. Extensive experiments on the Multimodal GarmentCodeData demonstrate that SwiftTailor achieves state-of-the-art accuracy and visual fidelity while significantly reducing inference time. This work offers a scalable, interpretable, and high-performance solution for next-generation 3D garment generation.

cs.CV↗

arXiv2026

PixelRush: Ultra-Fast, Training-Free High-Resolution Image Generation via One-step Diffusion

Pre-trained diffusion models excel at generating high-quality images but remain inherently limited by their native training resolution. Recent training-free approaches have attempted to overcome this constraint by introducing interventions during the denoising process; however, these methods incur substantial computational overhead, often requiring more than five minutes to produce a single 4K image. In this paper, we present PixelRush, the first tuning-free framework for practical high-resolution text-to-image generation. Our method builds upon the established patch-based inference paradigm but eliminates the need for multiple inversion and regeneration cycles. Instead, PixelRush enables efficient patch-based denoising within a low-step regime. To address artifacts introduced by patch blending in few-step generation, we propose a seamless blending strategy. Furthermore, we mitigate over-smoothing effects through a noise injection mechanism. PixelRush delivers exceptional efficiency, generating 4K images in approximately 20 seconds representing a 10$\times$ to 35$\times$ speedup over state-of-the-art methods while maintaining superior visual fidelity. Extensive experiments validate both the performance gains and the quality of outputs achieved by our approach.

cs.CV↗

arXiv2025

Larger Is Not Always Better: Leveraging Structured Code Diffs for Comment Inconsistency Detection

Ensuring semantic consistency between source code and its accompanying comments is crucial for program comprehension, effective debugging, and long-term maintainability. Comment inconsistency arises when developers modify code but neglect to update the corresponding comments, potentially misleading future maintainers and introducing errors. Recent approaches to code-comment inconsistency (CCI) detection leverage Large Language Models (LLMs) and rely on capturing the semantic relationship between code changes and outdated comments. However, they often ignore the structural complexity of code evolution, including historical change activities, and introduce privacy and resource challenges. In this paper, we propose a Just-In-Time CCI detection approach built upon the CodeT5+ backbone. Our method decomposes code changes into ordered sequences of modification activities such as replacing, deleting, and adding to more effectively capture the correlation between these changes and the corresponding outdated comments. Extensive experiments conducted on publicly available benchmark datasets-JITDATA and CCIBENCH--demonstrate that our proposed approach outperforms recent state-of-the-art models by up to 13.54% in F1-Score and achieves an improvement ranging from 4.18% to 10.94% over fine-tuned LLMs including DeepSeek-Coder, CodeLlama and Qwen2.5-Coder.

cs.SE↗

arXiv2025

CodeLSI: Leveraging Foundation Models for Automated Code Generation with Low-Rank Optimization and Domain-Specific Instruction Tuning

Context: Automated code generation using Foundation Models (FMs) offers promising solutions for enhancing software development efficiency. However, challenges remain in ensuring domain specificity, cost-effectiveness, and security - especially when relying on third-party APIs. This paper introduces CodeLSI, a framework that combines low-rank optimization and domain-specific instruction tuning to address these challenges. Objectives: The aim of this study is to develop and evaluate CodeLSI, a novel approach for generating high-quality code tailored to specific domains, using FMs fine-tuned on company infrastructure without dependence on external APIs. Methods: CodeLSI applies low-rank adaptation techniques to reduce the computational cost of model pre-training and fine-tuning. Domain-specific instruction tuning is employed to align code generation with organizational needs. We implemented and tested the framework on real-world JavaScript coding tasks using datasets drawn from internal software projects. Results: Experimental evaluations show that CodeLSI produces high-quality, context aware code. It outperforms baseline models in terms of relevance, accuracy, and domain fit. The use of low-rank optimization significantly reduced resource requirements, enabling scalable training on company-owned infrastructure. Conclusion: CodeLSI demonstrates that combining low-rank optimization with domain specific tuning can enhance the practicality and performance of FMs for automated code generation. This approach provides a secure, cost-efficient alternative to commercial API based solutions and supports faster, more targeted innovation in software development.

cs.SE↗

arXiv2025

MOOSE: Pay Attention to Temporal Dynamics for Video Understanding via Optical Flows

Many motion-centric video analysis tasks, such as atomic actions, detecting atypical motor behavior in individuals with autism, or analyzing articulatory motion in real-time MRI of human speech, require efficient and interpretable temporal modeling. Capturing temporal dynamics is a central challenge in video analysis, often requiring significant computational resources and fine-grained annotations that are not widely available. This paper presents MOOSE (Motion Flow Over Spatial Space), a novel temporally-centric video encoder explicitly integrating optical flow with spatial embeddings to model temporal information efficiently, inspired by human perception of motion. Unlike prior models, MOOSE takes advantage of rich, widely available pre-trained visual and optical flow encoders instead of training video models from scratch. This significantly reduces computational complexity while enhancing temporal interpretability. Our primary contributions includes (1) proposing a computationally efficient temporally-centric architecture for video understanding (2) demonstrating enhanced interpretability in modeling temporal dynamics; and (3) achieving state-of-the-art performance on diverse benchmarks, including clinical, medical, and standard action recognition datasets, confirming the broad applicability and effectiveness of our approach.

cs.CV↗

arXiv2025

Semise: Semi-supervised learning for severity representation in medical image

This paper introduces SEMISE, a novel method for representation learning in medical imaging that combines self-supervised and supervised learning. By leveraging both labeled and augmented data, SEMISE addresses the challenge of data scarcity and enhances the encoder's ability to extract meaningful features. This integrated approach leads to more informative representations, improving performance on downstream tasks. As result, our approach achieved a 12% improvement in classification and a 3% improvement in segmentation, outperforming existing methods. These results demonstrate the potential of SIMESE to advance medical image analysis and offer more accurate solutions for healthcare applications, particularly in contexts where labeled data is limited.

eess.IV↗

arXiv2024

SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation

We propose SharpDepth, a novel approach to monocular metric depth estimation that combines the metric accuracy of discriminative depth estimation methods (e.g., Metric3D, UniDepth) with the fine-grained boundary sharpness typically achieved by generative methods (e.g., Marigold, Lotus). Traditional discriminative models trained on real-world data with sparse ground-truth depth can accurately predict metric depth but often produce over-smoothed or low-detail depth maps. Generative models, in contrast, are trained on synthetic data with dense ground truth, generating depth maps with sharp boundaries yet only providing relative depth with low accuracy. Our approach bridges these limitations by integrating metric accuracy with detailed boundary preservation, resulting in depth predictions that are both metrically precise and visually sharp. Our extensive zero-shot evaluations on standard depth estimation benchmarks confirm SharpDepth effectiveness, showing its ability to achieve both high depth accuracy and detailed representation, making it well-suited for applications requiring high-quality depth perception across diverse, real-world environments.

cs.CV↗

arXiv2024

Semi-supervised 3D Semantic Scene Completion with 2D Vision Foundation Model Guidance

Accurate prediction of 3D semantic occupancy from 2D visual images is vital in enabling autonomous agents to comprehend their surroundings for planning and navigation. State-of-the-art methods typically employ fully supervised approaches, necessitating a huge labeled dataset acquired through expensive LiDAR sensors and meticulous voxel-wise labeling by human annotators. The resource-intensive nature of this annotating process significantly hampers the application and scalability of these methods. We introduce a novel semi-supervised framework to alleviate the dependency on densely annotated data. Our approach leverages 2D foundation models to generate essential 3D scene geometric and semantic cues, facilitating a more efficient training process. Our framework exhibits notable properties: (1) Generalizability, applicable to various 3D semantic scene completion approaches, including 2D-3D lifting and 3D-2D transformer methods. (2) Effectiveness, as demonstrated through experiments on SemanticKITTI and NYUv2, wherein our method achieves up to 85% of the fully-supervised performance using only 10% labeled data. This approach not only reduces the cost and labor associated with data annotation but also demonstrates the potential for broader adoption in camera-based systems for 3D semantic occupancy prediction.

cs.CV↗

arXiv2024

Design and Implementation of Smart Infrastructures and Connected Vehicles in A Mini-city Platform

This paper presents a 1/10th scale mini-city platform used as a testing bed for evaluating autonomous and connected vehicles. Using the mini-city platform, we can evaluate different driving scenarios including human-driven and autonomous driving. We provide a unique, visual feature-rich environment for evaluating computer vision methods. The conducted experiments utilize onboard sensors mounted on a robotic platform we built, allowing them to navigate in a controlled real-world urban environment. The designed city is occupied by cars, stop signs, a variety of residential and business buildings, and complex intersections mimicking an urban area. Furthermore, We have designed an intelligent infrastructure at one of the intersections in the city which helps safer and more efficient navigation in the presence of multiple cars and pedestrians. We have used the mini-city platform for the analysis of three different applications: city mapping, depth estimation in challenging occluded environments, and smart infrastructure for connected vehicles. Our smart infrastructure is among the first to develop and evaluate Vehicle-to-Infrastructure (V2I) communication at intersections. The intersection-related result shows how inaccuracy in perception, including mapping and localization, can affect safety. The proposed mini-city platform can be considered as a baseline environment for developing research and education in intelligent transportation systems.

cs.RO↗

arXiv2024

Nemotron-4 340B Technical Report

We release the Nemotron-4 340B model family, including Nemotron-4-340B-Base, Nemotron-4-340B-Instruct, and Nemotron-4-340B-Reward. Our models are open access under the NVIDIA Open Model License Agreement, a permissive model license that allows distribution, modification, and use of the models and its outputs. These models perform competitively to open access models on a wide range of evaluation benchmarks, and were sized to fit on a single DGX H100 with 8 GPUs when deployed in FP8 precision. We believe that the community can benefit from these models in various research studies and commercial applications, especially for generating synthetic data to train smaller language models. Notably, over 98% of data used in our model alignment process is synthetically generated, showcasing the effectiveness of these models in generating synthetic data. To further support open research and facilitate model development, we are also open-sourcing the synthetic data generation pipeline used in our model alignment process.

cs.CL↗

arXiv2022

EmbryosFormer: Deformable Transformer and Collaborative Encoding-Decoding for Embryos Stage Development Classification

The timing of cell divisions in early embryos during the In-Vitro Fertilization (IVF) process is a key predictor of embryo viability. However, observing cell divisions in Time-Lapse Monitoring (TLM) is a time-consuming process and highly depends on experts. In this paper, we propose EmbryosFormer, a computational model to automatically detect and classify cell divisions from original time-lapse images. Our proposed network is designed as an encoder-decoder deformable transformer with collaborative heads. The transformer contracting path predicts per-image labels and is optimized by a classification head. The transformer expanding path models the temporal coherency between embryo images to ensure monotonic non-decreasing constraint and is optimized by a segmentation head. Both contracting and expanding paths are synergetically learned by a collaboration head. We have benchmarked our proposed EmbryosFormer on two datasets: a public dataset with mouse embryos with 8-cell stage and an in-house dataset with human embryos with 4-cell stage. Source code: https://github.com/UARK-AICV/Embryos.

cs.CV↗

arXiv2022

Network-aware Prefetching Method for Short-Form Video Streaming

Recent years have witnessed the rising of short-form video platforms such as TikTok. Apart from conventional videos, short-form videos are much shorter and users frequently change the content to watch. Thus, it is crucial to have an effective streaming method for this new type of video. In this paper, we propose a resource-efficient prefetching method for short-form video streaming. Taking into account network throughput conditions and user viewing behaviors, the proposed method dynamically adapts the amount of prefetched video data. Experiment results show that our method can reduce the data waste by 37~52% compared to other existing methods.

cs.MM↗

arXiv2021

SP-GPT2: Semantics Improvement in Vietnamese Poetry Generation

Automatic text generation has garnered growing attention in recent years as an essential step towards computer creativity. Generative Pretraining Transformer 2 (GPT2) is one of the state of the art approaches that have excellent successes. In this paper, we took the first step to investigate the power of GPT2 in traditional Vietnamese poetry generation. In the earlier time, our experiment with base GPT2 was quite good at generating the poem in the proper template. Though it can learn the patterns, including rhyme and tone rules, from the training data, like almost all other text generation approaches, the poems generated still has a topic drift and semantic inconsistency. To improve the cohesion within the poems, we proposed a new model SP-GPT2 (semantic poem GPT2) which was built on the top GPT2 model and an additional loss to constrain context throughout the entire poem. For better evaluation, we examined the methods by both automatic quantitative evaluation and human evaluation. Both automatic and human evaluation demonstrated that our approach can generate poems that have better cohesion without losing the quality due to additional loss. At the same time, we are the pioneers of this topic. We released the first computational scoring module for poems generated in the template containing the style rule dictionary. Additionally, we are the first to publish a Luc-Bat dataset, including 87609 Luc Bat poems, which is equivalent to about 2.6 million sentences, combined with about 83579 poems in other styles was also published for further exploration. The code is available at https://github.com/fsoft-ailab/Poem-Generator

cs.CL↗