arXiv subjects

Fusheng Hao

Publications and source records attributed to Fusheng Hao.

5 recordsLinked to original sources

arXiv2026

SAVLA: Symmetry-Aware Vision-Language-Action Models for Robotic Manipulation

Vision-language-action (VLA) models have become the dominant paradigm for language-conditioned robot manipulation. However, although images and language instructions inherently encode geometric information, VLAs acquire their spatial competence purely from demonstrations. As a result, they are reliable only within the range of scene poses that the demonstrations cover. We propose SAVLA, an end-to-end symmetry-aware VLA model for robust and data-efficient policy learning. Our approach keeps the pretrained vision-language backbone entirely frozen while combining it with an equivariant flow-matching action head and a learned canonicalizer. The head decomposes its state, action, and conditioning inputs into invariant and equivariant channels, and preserves this typing throughout all of its layers. The canonicalizer transforms oblique-view images into a canonical frame and rotates the geometric conditions consistently. We evaluate our model on LIBERO. Compared with the GR00T N1.5 baseline, SAVLA improves the success rate averaged over all four LIBERO suites by 5.1 points and increases the mean success rate under rotation on LIBERO-Goal from 41.5% to 90.4%.

cs.RO↗

arXiv2026

IVR-R1: Refining Trajectories through Iterative Visual-Grounded Reasoning in Reinforcement Learning

Multimodal large language models via reinforcement learning (RL) have demonstrated remarkable capabilities in complex visual reasoning tasks, yet they remain limited in long-horizon multimodal scenarios, often suffering from visual hallucination and logical error. Current methods typically pre-encode high-dimensional visual scenes into discrete textual proxies to facilitate downstream reasoning. As the reasoning chain unfolds, however, the inherent information asymmetry between text and visual scenes tends to erode visual grounding, resulting in misguided reasoning and erroneous outputs. To address this issue, we introduce IVR-R1 (Iterative Visual-grounded Reasoning), a novel RL training framework that facilitates dynamic visual re-alignment that actively rectifies reasoning trajectories to guide policy optimization. Specifically, by leveraging a reward-driven screening mechanism to identify flawed rollouts, IVR-R1 executes a fine-grained, step-level error attribution within the multimodal context. By iteratively cross-referencing intermediate reasoning states against pristine visual priors, a Re-Reasoning Loop enables automated trajectory rectification, effectively synthesizing expert-level demonstrations that serve as high-fidelity reasoning templates for the policy model. Our experiments across diverse multimodal benchmarks demonstrate that IVR-R1 consistently outperforms existing reinforcement learning methods, establishing a superior paradigm for maintaining logical and visual consistency in complex multimodal reasoning.

cs.CV↗

arXiv2025

Gradient-based Fine-Tuning through Pre-trained Model Regularization

Large pre-trained models have demonstrated extensive applications across various fields. However, fine-tuning these models for specific downstream tasks demands significant computational resources and storage. One fine-tuning method, gradient-based parameter selection (GPS), focuses on fine-tuning only the parameters with high gradients in each neuron, thereby reducing the number of training parameters. Nevertheless, this approach increases computational resource requirements and storage demands. In this paper, we propose an efficient gradient-based and regularized fine-tuning method (GRFT) that updates the rows or columns of the weight matrix. We theoretically demonstrate that the rows or columns with the highest sum of squared gradients are optimal for updating. This strategy effectively reduces storage overhead and improves the efficiency of parameter selection. Additionally, we incorporate regularization to enhance knowledge transfer from the pre-trained model. GRFT achieves state-of-the-art performance, surpassing existing methods such as GPS, Adapter Tuning, and LoRA. Notably, GRFT requires updating only 1.22% and 0.30% of the total parameters on FGVC and VTAB datasets, respectively, demonstrating its high efficiency and effectiveness. The source code will be released soon.

cs.LG↗

arXiv2023

Human-imperceptible, Machine-recognizable Images

Massive human-related data is collected to train neural networks for computer vision tasks. A major conflict is exposed relating to software engineers between better developing AI systems and distancing from the sensitive training data. To reconcile this conflict, this paper proposes an efficient privacy-preserving learning paradigm, where images are first encrypted to become ``human-imperceptible, machine-recognizable'' via one of the two encryption strategies: (1) random shuffling to a set of equally-sized patches and (2) mixing-up sub-patches of the images. Then, minimal adaptations are made to vision transformer to enable it to learn on the encrypted images for vision tasks, including image classification and object detection. Extensive experiments on ImageNet and COCO show that the proposed paradigm achieves comparable accuracy with the competitive methods. Decrypting the encrypted images requires solving an NP-hard jigsaw puzzle or an ill-posed inverse problem, which is empirically shown intractable to be recovered by various attackers, including the powerful vision transformer-based attacker. We thus show that the proposed paradigm can ensure the encrypted images have become human-imperceptible while preserving machine-recognizable information. The code is available at \url{https://github.com/FushengHao/PrivacyPreservingML.}

cs.CV↗

arXiv2018

Anchor-based Nearest Class Mean Loss for Convolutional Neural Networks

Discriminative features are critical for machine learning applications. Most existing deep learning approaches, however, rely on convolutional neural networks (CNNs) for learning features, whose discriminant power is not explicitly enforced. In this paper, we propose a novel approach to train deep CNNs by imposing the intra-class compactness and the inter-class separability, so as to enhance the learned features' discriminant power. To this end, we introduce anchors, which are predefined vectors regarded as the centers for each class and fixed during training. Discriminative features are obtained by constraining the deep CNNs to map training samples to the corresponding anchors as close as possible. We propose two principles to select the anchors, and measure the proximity of two points using the Euclidean and cosine distance metric functions, which results in two novel loss functions. These loss functions require no sample pairs or triplets and can be efficiently optimized by batch stochastic gradient descent. We test the proposed method on three benchmark image classification datasets and demonstrate its promising results.

cs.CV↗