Author name cluster

Huan Ma

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

9 papers

2 author rows

AAAI Conference 2025 Conference Paper

Exploring Task-Level Optimal Prompts for Visual In-Context Learning

Yan Zhu
Huan Ma
Changqing Zhang

With the development of Vision Foundation Models (VFMs) in recent years, Visual In-Context Learning (VICL) has become a better choice compared to modifying models in most scenarios. Different from retraining or fine-tuning models, VICL does not require modifications to the model's weights and architecture, and only needs a prompt with demonstrations to teach VFM how to solve tasks. Currently, significant computational cost for finding optimal prompts for every test sample hinders the deployment of VICL, as determining which demonstrations to use for constructing the prompt is very costly. In this paper, however, we find a counterintuitive phenomenon that most test samples actually achieve optimal performance under the same prompts, and searching for sample-level prompts only costs much time but results in completely identical prompts actually. Therefore, we propose task-level prompting to reduce the cost of searching for prompts during the inference stage and introduce two time-saving yet effective task-level prompt search strategies accordingly. Extensive experimental results show that our proposed method can identify near-optimal prompts and reach the best VICL performance with a minimal cost that prior work has never achieved.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Spurious Feature Eraser: Stabilizing Test-Time Adaptation for Vision-Language Foundation Model

Huan Ma
Yan Zhu
Changqing Zhang
Peilin Zhao
Baoyuan Wu
Long-Kai Huang
Qinghua Hu
Bingzhe Wu

Vision-language foundation models have exhibited remarkable success across a multitude of downstream tasks due to their scalability on extensive image-text paired data. However, these models also display significant limitations when applied to downstream tasks, such as fine-grained image classification, as a result of ``decision shortcuts'' that hinder their generalization capabilities. In this work, we find that the CLIP model possesses a rich set of features, encompassing both desired invariant causal features and undesired decision shortcuts. Moreover, the underperformance of CLIP on downstream tasks originates from its inability to effectively utilize pre-trained features in accordance with specific task requirements. To address this challenge, we propose a simple yet effective method, Spurious Feature Eraser (SEraser), to alleviate the decision shortcuts by erasing the spurious features. Specifically, we introduce a test-time prompt tuning paradigm that optimizes a learnable prompt, thereby compelling the model to exploit invariant features while disregarding decision shortcuts during the inference phase. The proposed method effectively alleviates excessive dependence on potentially misleading spurious information. We conduct comparative analysis of the proposed method against various approaches which validates the significant superiority.

PDF Details DOI

ICRA Conference 2024 Conference Paper

GelRoller: A Rolling Vision-based Tactile Sensor for Large Surface Reconstruction Using Self-Supervised Photometric Stereo Method

Zhiyuan Zhang
Huan Ma
Yulin Zhou
Jingjing Ji
Hua Yang 0002

Accurate perception of the surrounding environment stands as a primary objective for robots. Through tactile interaction, vision-based tactile sensors provide the capability to capture high-resolution and multi-modal surface information of objects, thereby facilitating robots in achieving more dexterous manipulations. However, the prevailing GelSight sensors entail intricate calibration procedures, posing challenges in their application on curved surfaces and requiring the maintenance of stable lighting conditions throughout experimentation. Additionally, constrained by shape and structure, current vision-based tactile sensors are predominantly applied to measurements within a limited area. In this study, we design a novel cylindrical vision-based tactile sensor that enables continuous and swift perception of large-scale object surfaces through rolling. To tackle the challenges posed by laborious calibration processes, we propose a self-supervised photometric stereo method based on deep learning, which eliminates pre-calibration requirements and enables the derivation of surface normals from a single image without relying on stable lighting conditions. Finally, we perform surface reconstruction from normal and point cloud registration on the multiple frames of images obtained by rolling the cylindrical sensor, resulting in large surface reconstruction. We compare our method with the representative lookup table method in the GelSight sensors. The results show that the proposed method enhances both reconstruction accuracy and robustness, thereby demonstrating the potential of the proposed sensor in large-scale surface reconstruction. Codes and mechanical structures are available at: https://github.com/ZhangZhiyuanZhang/GelRoller

Details

NeurIPS Conference 2024 Conference Paper

Scaling Law for Time Series Forecasting

Jingzhe Shi
Qinwei Ma
Huan Ma
Lei Li

Scaling law that rewards large datasets, complex models and enhanced data granularity has been observed in various fields of deep learning. Yet, studies on time series forecasting have cast doubt on scaling behaviors of deep learning methods for time series forecasting: while more training data improves performance, more capable models do not always outperform less capable models, and longer input horizon may hurt performance for some models. We propose a theory for scaling law for time series forecasting that can explain these seemingly abnormal behaviors. We take into account the impact of dataset size and model complexity, as well as time series data granularity, particularly focusing on the look-back horizon, an aspect that has been unexplored in previous theories. Furthermore, we empirically evaluate various models using a diverse set of time series forecasting datasets, which (1) verifies the validity of scaling law on dataset size and model complexity within the realm of time series forecasting, and (2) validates our theoretical framework, particularly regarding the influence of look back horizon. We hope our findings may inspire new models targeting time series forecasting datasets of limited size, as well as large foundational datasets and models for time series forecasting in future works.

PDF Details DOI

ICRA Conference 2024 Conference Paper

STNet: Spatio-Temporal Fusion-Based SelfAttention for Slip Detection in Visuo-Tactile Sensors

Jin Lu
Bangyan Niu
Huan Ma
Jiafeng Zhu
Jingjing Ji

Slip detection plays a pivotal role in the dexterity of robotics, improving the reliability and precision of manipulations but also contributing to safety, efficiency, and adaptability. Deep learning-based slip detection algorithms commonly difficult to concentrate on key features when faced with dense 3D shape data obtained by visuo-tactile sensors. Data from noncontact locations can interfere with slip judgements and the ignorance of interframe linkage can also lead to slip detection failure. In this paper, a new spatio-temporal sequences fusion-based self-attention, STNet, is proposed to perform slip detection by allocating more attention to the object-sensor contact area when processing complex 3D shape data. A binocular visuo-tactile system (BVTS) is designed and fabricated for dataset construction. The entire 3D shape dataset containing 4 motion patterns, including stationary, pressing, rolling and slipping. Self-attention architecture with and without spatio-temporal sequences fusion mechanism (denoted as STNet and TemNet, respectively) are trained based on the same dataset. The experiments show the validity of STNet, which can reach 98. 91% slip detection accuracy. Meanwhile, the ablation studies confirm the effectiveness of the spatio-temporal sequences fusion mechanism.

Details

NeurIPS Conference 2023 Conference Paper

Fairness-guided Few-shot Prompting for Large Language Models

Huan Ma
Changqing Zhang
Yatao Bian
Lemao Liu
Zhirui Zhang
Peilin Zhao
Shu Zhang
Huazhu Fu

Large language models have demonstrated surprising ability to perform in-context learning, i. e. , these models can be directly applied to solve numerous downstream tasks by conditioning on a prompt constructed by a few input-output examples. However, prior research has shown that in-context learning can suffer from high instability due to variations in training examples, example order, and prompt formats. Therefore, the construction of an appropriate prompt is essential for improving the performance of in-context learning. In this paper, we revisit this problem from the view of predictive bias. Specifically, we introduce a metric to evaluate the predictive bias of a fixed prompt against labels or a given attributes. Then we empirically show that prompts with higher bias always lead to unsatisfactory predictive quality. Based on this observation, we propose a novel search strategy based on the greedy search to identify the near-optimal prompt for improving the performance of in-context learning. We perform comprehensive experiments with state-of-the-art mainstream models such as GPT-3 on various downstream tasks. Our results indicate that our method can enhance the model's in-context learning performance in an effective and interpretable manner.

PDF Details

NeurIPS Conference 2021 Conference Paper

Trustworthy Multimodal Regression with Mixture of Normal-inverse Gamma Distributions

Huan Ma
Zongbo Han
Changqing Zhang
Huazhu Fu
Joey Tianyi Zhou
Qinghua Hu

Multimodal regression is a fundamental task, which integrates the information from different sources to improve the performance of follow-up applications. However, existing methods mainly focus on improving the performance and often ignore the confidence of prediction for diverse situations. In this study, we are devoted to trustworthy multimodal regression which is critical in cost-sensitive domains. To this end, we introduce a novel Mixture of Normal-Inverse Gamma distributions (MoNIG) algorithm, which efficiently estimates uncertainty in principle for adaptive integration of different modalities and produces a trustworthy regression result. Our model can be dynamically aware of uncertainty for each modality, and also robust for corrupted modalities. Furthermore, the proposed MoNIG ensures explicitly representation of (modality-specific/global) epistemic and aleatoric uncertainties, respectively. Experimental results on both synthetic and different real-world data demonstrate the effectiveness and trustworthiness of our method on various multimodal regression tasks (e. g. , temperature prediction for superconductivity, relative location prediction for CT slices, and multimodal sentiment analysis).

PDF Details

TCS Journal 2014 Journal Article

Mining hidden links in social networks to achieve equilibrium

Huan Ma
Zaixin Lu
Deying Li
Yuqing Zhu
Lidan Fan
Weili Wu

Although more connections between individuals in a social network can be identified with the development of high techniques, to obtain the complete relation information between individuals is still hard due to complex structure and individual privacy. However, the social networks have communities. In our work, we aim at mining the invisible or missing relations between individuals within a community in social networks. We propose our algorithm according to the fact that the individuals exist in communities satisfying Nash equilibrium, which is borrowed from game-theoretic concepts often used in economic researches. Each hidden relation is explored through the individual's loyalty to their community. To the best of our knowledge, this is the first work that studies the problem of mining hidden links from the aspect of Nash equilibrium. Eventually we confirm our approach's superiority from extensive experiments over real-world social networks.

Details DOI

TCS Journal 2013 Journal Article

Minimum energy multicast/broadcast routing with reception cost in wireless sensor networks

Deying Li
Zewen Liu
Yi Hong
Wenping Chen
Huan Ma

In this paper, we study the minimum energy multicast/broadcast problem with reception cost in wireless sensor networks. Suppose there are n sensors in the network. Each node v has l ( v ) transmission power levels to choose and its reception cost is B ( v ) if it receives a message. The problem of our concern is: given a multicast (broadcast) request, how to find a multicast (broadcast) tree such that the total energy cost of the multicast tree including transmitting cost and reception cost is minimized. There are two cases for reception cost: one is that for any node v, the reception cost of v only relies on v itself and is irrelevant with its transmitting node, the other is that the reception cost of v relies on not only itself but also its transmitting node. For the first case, we firstly propose a general approximation algorithm MEB-R-G for the broadcast problem. Moreover, for the multicast problem, we propose a general algorithm MEM-R-G and prove its approximation ratio, we also present a greedy algorithm. For the second case, we also propose a general approximation algorithm MEM-RT-G, and prove its approximation ratio.

Details DOI