arXiv subjects

Shafei Wang

Publications and source records attributed to Shafei Wang.

11 recordsLinked to original sources

arXiv2025

Enhancing Large Language Models with Reward-guided Tree Search for Knowledge Graph Question and Answering

Recently, large language models (LLMs) have demonstrated impressive performance in Knowledge Graph Question Answering (KGQA) tasks, which aim to find answers based on knowledge graphs (KGs) for natural language questions. Existing LLMs-based KGQA methods typically follow the Graph Retrieval-Augmented Generation (GraphRAG) paradigm, which first retrieves reasoning paths from the large KGs, and then generates the answers based on them. However, these methods emphasize the exploration of new optimal reasoning paths in KGs while ignoring the exploitation of historical reasoning paths, which may lead to sub-optimal reasoning paths. Additionally, the complex semantics contained in questions may lead to the retrieval of inaccurate reasoning paths. To address these issues, this paper proposes a novel and training-free framework for KGQA tasks called Reward-guided Tree Search on Graph (RTSoG). RTSoG decomposes an original question into a series of simpler and well-defined sub-questions to handle the complex semantics. Then, a Self-Critic Monte Carlo Tree Search (SC-MCTS) guided by a reward model is introduced to iteratively retrieve weighted reasoning paths as contextual knowledge. Finally, it stacks the weighted reasoning paths according to their weights to generate the final answers. Extensive experiments on four datasets demonstrate the effectiveness of RTSoG. Notably, it achieves 8.7\% and 7.0\% performance improvement over the state-of-the-art method on the GrailQA and the WebQSP respectively.

cs.CL↗

arXiv2025

EPERM: An Evidence Path Enhanced Reasoning Model for Knowledge Graph Question and Answering

Due to the remarkable reasoning ability, Large language models (LLMs) have demonstrated impressive performance in knowledge graph question answering (KGQA) tasks, which find answers to natural language questions over knowledge graphs (KGs). To alleviate the hallucinations and lack of knowledge issues of LLMs, existing methods often retrieve the question-related information from KGs to enrich the input context. However, most methods focus on retrieving the relevant information while ignoring the importance of different types of knowledge in reasoning, which degrades their performance. To this end, this paper reformulates the KGQA problem as a graphical model and proposes a three-stage framework named the Evidence Path Enhanced Reasoning Model (EPERM) for KGQA. In the first stage, EPERM uses the fine-tuned LLM to retrieve a subgraph related to the question from the original knowledge graph. In the second stage, EPERM filters out the evidence paths that faithfully support the reasoning of the questions, and score their importance in reasoning. Finally, EPERM uses the weighted evidence paths to reason the final answer. Since considering the importance of different structural information in KGs for reasoning, EPERM can improve the reasoning ability of LLMs in KGQA tasks. Extensive experiments on benchmark datasets demonstrate that EPERM achieves superior performances in KGQA tasks.

cs.CL↗

arXiv2025

Flatness Guided Test-Time Adaptation for Vision-Language Models

Test-time adaptation (TTA) of Vision-Language Models (VLMs) has emerged as a technique for tackling distribution shifts during the test time. Recent research indicates that the test-time adaptation is intrinsically linked to the model's training history. However, existing TTA methods, such as Test-time Prompt Tuning, often design adaptation strategies in isolation from the models' training characteristics, which degrade their performance. This paper argues that the flatness acquired via sharpness-aware training is an efficient clue for the test-time adaptation of VLMs. Built on this insight, this paper proposes a novel Flatness-Guided Adaptation framework (FGA) for VLMs to cohesively unify training and test-time procedures. Its core idea is to leverage the alignment between the training minimum and test loss flat regions to guide the adaptation process. Specifically, our FGA consists of a prompt-tuning stage and a test-time adaptation stage. In the tuning stage, a Sharpness-Aware Prompt Tuning method is utilized to identify the training flat minimum, offering a geometric clue of flatness for subsequent adaptation. In the test stage, a Sharpness-based Test Sample Selection approach is proposed to ensure the alignment of flat minima between the training and each augmented test sample's loss landscape. In comparison to existing TTA methods, our FGA avoids the expensive prompt parameter updates during test time, and substantially reduces the computation overhead. Extensive experiments on both domain generalization and cross-dataset benchmarks demonstrate that our FGA achieves superior performance over prevalent TTA methods. Notably, when employing a ViT-B/16 image encoder, FGA even outperforms TPT+CoOp by an average of 4.88% across all four ImageNet out-of-domain variants.

cs.CV↗

arXiv2025

Infinite Factorial Linear Dynamical Systems for Transient Signal Detection

Accurately detecting the transient signal of interest from the background signal is one of the fundamental tasks in signal processing. The most recent approaches assume the existence of a single background source and represent the background signal using a linear dynamical system (LDS). This assumption might fail to capture the complexities of modern electromagnetic environments with multiple sources. To address this limitation, this paper proposes a method for detecting the transient signal in a background composed of an unknown number of emitters. The proposed method consists of two main tasks. First, a Bayesian nonparametric model called the infinite factorial linear dynamical system (IFLDS) is developed. The developed model is based on the sticky Indian buffet process and enables the representation and parameter learning of the unbounded number of background sources. This study also designs a parameter learning method for the IFLDS using slice sampling and particle Gibbs with ancestor sampling. Second, the finite moving average (FMA) stopping time is introduced to minimize the worst-case probability of missed detection, and the statistical performance of the stopping time is investigated. To facilitate the computation of the FMA stopping time, this study derives the factorial Kalman forward filtering (FKFF) method and designs a dependence structure for the underlying model, allowing the stopping time to be defined by a recursive function. Numerical simulations demonstrate the effectiveness of the proposed method and the validity of the theoretical results. The experimental results of the pulse signal detection under the condition of communication interference confirm the effectiveness and superiority of the proposed method.

eess.SP↗

arXiv2024

Seeking Consistent Flat Minima for Better Domain Generalization via Refining Loss Landscapes

Domain generalization aims to learn a model from multiple training domains and generalize it to unseen test domains. Recent theory has shown that seeking the deep models, whose parameters lie in the flat minima of the loss landscape, can significantly reduce the out-of-domain generalization error. However, existing methods often neglect the consistency of loss landscapes in different domains, resulting in models that are not simultaneously in the optimal flat minima in all domains, which limits their generalization ability. To address this issue, this paper proposes an iterative Self-Feedback Training (SFT) framework to seek consistent flat minima that are shared across different domains by progressively refining loss landscapes during training. It alternatively generates a feedback signal by measuring the inconsistency of loss landscapes in different domains and refines these loss landscapes for greater consistency using this feedback signal. Benefiting from the consistency of the flat minima within these refined loss landscapes, our SFT helps achieve better out-of-domain generalization. Extensive experiments on DomainBed demonstrate superior performances of SFT when compared to state-of-the-art sharpness-aware methods and other prevalent DG baselines. On average across five DG benchmarks, SFT surpasses the sharpness-aware minimization by 2.6% with ResNet-50 and 1.5% with ViT-B/16, respectively. The code will be available soon.

cs.CV↗

arXiv2024

Cost-Effective RF Fingerprinting Based on Hybrid CVNN-RF Classifier with Automated Multi-Dimensional Early-Exit Strategy

While the Internet of Things (IoT) technology is booming and offers huge opportunities for information exchange, it also faces unprecedented security challenges. As an important complement to the physical layer security technologies for IoT, radio frequency fingerprinting (RFF) is of great interest due to its difficulty in counterfeiting. Recently, many machine learning (ML)-based RFF algorithms have emerged. In particular, deep learning (DL) has shown great benefits in automatically extracting complex and subtle features from raw data with high classification accuracy. However, DL algorithms face the computational cost problem as the difficulty of the RFF task and the size of the DNN have increased dramatically. To address the above challenge, this paper proposes a novel costeffective early-exit neural network consisting of a complex-valued neural network (CVNN) backbone with multiple random forest branches, called hybrid CVNN-RF. Unlike conventional studies that use a single fixed DL model to process all RF samples, our hybrid CVNN-RF considers differences in the recognition difficulty of RF samples and introduces an early-exit mechanism to dynamically process the samples. When processing "easy" samples that can be well classified with high confidence, the hybrid CVNN-RF can end early at the random forest branch to reduce computational cost. Conversely, subsequent network layers will be activated to ensure accuracy. To further improve the early-exit rate, an automated multi-dimensional early-exit strategy is proposed to achieve scheduling control from multiple dimensions within the network depth and classification category. Finally, our experiments on the public ADS-B dataset show that the proposed algorithm can reduce the computational cost by 83% while improving the accuracy by 1.6% under a classification task with 100 categories.

eess.SP↗

arXiv2024

Representation and De-interleaving of Mixtures of Hidden Markov Processes

De-interleaving of the mixtures of Hidden Markov Processes (HMPs) generally depends on its representation model. Existing representation models consider Markov chain mixtures rather than hidden Markov, resulting in the lack of robustness to non-ideal situations such as observation noise or missing observations. Besides, de-interleaving methods utilize a search-based strategy, which is time-consuming. To address these issues, this paper proposes a novel representation model and corresponding de-interleaving methods for the mixtures of HMPs. At first, a generative model for representing the mixtures of HMPs is designed. Subsequently, the de-interleaving process is formulated as a posterior inference for the generative model. Secondly, an exact inference method is developed to maximize the likelihood of the complete data, and two approximate inference methods are developed to maximize the evidence lower bound by creating tractable structures. Then, a theoretical error probability lower bound is derived using the likelihood ratio test, and the algorithms are shown to get reasonably close to the bound. Finally, simulation results demonstrate that the proposed methods are highly effective and robust for non-ideal situations, outperforming baseline methods on simulated and real-life data.

stat.ML↗

arXiv2024

Class Information Guided Reconstruction for Automatic Modulation Open-Set Recognition

Automatic Modulation Recognition (AMR) is a crucial technology in the domains of radar and communications. Traditional AMR approaches assume a closed-set scenario, where unknown samples are forcibly misclassified into known classes, leading to serious consequences for situation awareness and threat assessment. To address this issue, Automatic Modulation Open-set Recognition (AMOSR) defines two tasks as Known Class Classification (KCC) and Unknown Class Identification (UCI). However, AMOSR faces core challenges in terms of inappropriate decision boundaries and sparse feature distributions. To overcome the aforementioned challenges, we propose a Class Information guided Reconstruction (CIR) framework, which leverages reconstruction losses to distinguish known and unknown classes. To enhance distinguishability, we design Class Conditional Vectors (CCVs) to match the latent representations extracted from input samples, achieving perfect reconstruction for known samples while yielding poor results for unknown ones. We also propose a Mutual Information (MI) loss function to ensure reliable matching, with upper and lower bounds of MI derived for tractable optimization and mathematical proofs provided. The mutually beneficial CCVs and MI facilitate the CIR attaining optimal UCI performance without compromising KCC accuracy, especially in scenarios with a higher proportion of unknown classes. Additionally, a denoising module is introduced before reconstruction, enabling the CIR to achieve a significant performance improvement at low SNRs. Experimental results on simulated and measured signals validate the effectiveness and the robustness of the proposed method.

eess.SP↗

arXiv2024

Mitigating Receiver Impact on Radio Frequency Fingerprint Identification via Domain Adaptation

Radio Frequency Fingerprint Identification (RFFI), which exploits non-ideal hardware-induced unique distortion resident in the transmit signals to identify an emitter, is emerging as a means to enhance the security of communication systems. Recently, machine learning has achieved great success in developing state-of-the-art RFFI models. However, few works consider cross-receiver RFFI problems, where the RFFI model is trained and deployed on different receivers. Due to altered receiver characteristics, direct deployment of RFFI model on a new receiver leads to significant performance degradation. To address this issue, we formulate the cross-receiver RFFI as a model adaptation problem, which adapts the trained model to unlabeled signals from a new receiver. We first develop a theoretical generalization error bound for the adaptation model. Motivated by the bound, we propose a novel method to solve the cross-receiver RFFI problem, which includes domain alignment and adaptive pseudo-labeling. The former aims at finding a feature space where both domains exhibit similar distributions, effectively reducing the domain discrepancy. Meanwhile, the latter employs a dynamic pseudo-labeling scheme to implicitly transfer the label information from the labeled receiver to the new receiver. Experimental results indicate that the proposed method can effectively mitigate the receiver impact and improve the cross-receiver RFFI performance.

eess.SP↗

arXiv2023

Bayesian Non-parametric Hidden Markov Model for Agile Radar Pulse Sequences Streaming Analysis

Multi-function radars (MFRs) are sophisticated types of sensors with the capabilities of complex agile inter-pulse modulation implementation and dynamic work mode scheduling. The developments in MFRs pose great challenges to modern electronic reconnaissance systems or radar warning receivers for recognition and inference of MFR work modes. To address this issue, this paper proposes an online processing framework for parameter estimation and change point detection of MFR work modes. At first, this paper designed a fully-conjugate Bayesian non-parametric hidden Markov model with a designed prior distribution (agile BNP-HMM) to represent the MFR pulse agility characteristics. The proposed model allows fully-variational Bayesian inference. Then, the proposed framework is constructed by two main parts. The first part is the agile BNP-HMM model for automatically inferring the number of HMM hidden states and emission distribution of the corresponding hidden states. An estimation error lower bound on performance is derived and the proposed algorithm is shown to be close to the bound. The second part utilizes the streaming Bayesian updating to facilitate computation, and designed an online work mode change detection framework based upon a weighted sequential probability ratio test. We demonstrate that the proposed framework is consistently highly effective and robust to baseline methods on diverse simulated data-sets.

eess.SP↗

arXiv2022

Latency Minimization for Multiuser Computation Offloading in Fog-Radio Access Networks

This paper considers computation offloading in fog-radio access networks (F-RAN), where multiple user equipments (UEs) offload their computation tasks to the F-RAN through a number of fog nodes. Each UE can choose one of the fog nodes to offload its task, and each fog node may serve multiple UEs. Depending on the computation burden at the fog nodes, the tasks may be computed by the fog nodes or further offloaded to the cloud via capacity-limited fronthaul links. To compute all UEs' tasks as fast as possible, joint optimization of UE-Fog association, radio and computation resources of F-RAN is proposed to minimize the maximum latency of all UEs. This min-max problem is formulated as a mixed integer nonlinear program (MINP). We first show that the MINP can be reformulated as a continuous optimization problem, and then employ the majorization minimization (MM) approach to find a solution. The MM approach that we develop is unconventional in that -- each MM subproblem can be solved inexactly with the same provable convergence guarantee as the conventional exact MM, thereby reducing the complexity of each MM iteration. In addition, we also consider a cooperative offloading model, where the fog nodes compress-and-forward their received signals to the cloud. Under this model, a similar min-max latency optimization problem is formulated and tackled again by the inexact MM approach. Simulation results show that the proposed algorithms outperform some heuristic offloading strategies, and that the cooperative offloading can better exploit the transmission diversity to attain better latency performance than the non-cooperative one.

eess.SP↗