arXiv subjects

Haohan Shi

Publications and source records attributed to Haohan Shi.

6 recordsLinked to original sources

arXiv2026

Audio Deepfake Detection at the First Greeting: "Hi!"

This paper focuses on audio deepfake detection under real-world communication degradations, with an emphasis on ultra-short inputs (0.5-2.0s), targeting the capability to detect synthetic speech at a conversation opening, e.g., when a scammer says "Hi." We propose Short-MGAA (S-MGAA), a novel lightweight extension of Multi-Granularity Adaptive Time-Frequency Attention, designed to enhance discriminative representation learning for short, degraded inputs subjected to communication processing and perturbations. The S-MGAA integrates two tailored modules: a Pixel-Channel Enhanced Module (PCEM) that amplifies fine-grained time-frequency saliency, and a Frequency Compensation Enhanced Module (FCEM) to supplement limited temporal evidence via multi-scale frequency modeling and adaptive frequency-temporal interaction. Extensive experiments demonstrate that S-MGAA consistently surpasses nine state-of-the-art baselines while achieving strong robustness to degradations and favorable efficiency-accuracy trade-offs, including low RTF, competitive GFLOPs, compact parameters, and reduced training cost, highlighting its strong potential for real-time deployment in communication systems and edge devices.

eess.AS↗

arXiv2025

The Persistence of Retracted Papers on Wikipedia

Wikipedia serves as a key infrastructure for public access to scientific knowledge, but it faces challenges in maintaining the credibility of cited sources--especially when scientific papers are retracted. This paper investigates how citations to retracted research are handled on English Wikipedia. We construct a novel dataset that integrates Wikipedia revision histories with metadata from Retraction Watch, Crossref, Altmetric, and OpenAlex, identifying 1,181 citations of retracted papers. We find that 71.6% of the citations were initially problematic and in need of reader-facing repair, defined as those added before the paper's retraction (51.5%) or introduced afterwards without proper warning (20.1%). While many are eventually corrected, our analysis reveals that these citations persist for a median of 3.68 years (1,344 days). Through survival analysis, we find that bot-mediated flagging (RetractionBot), open access availability, pre-existing online visibility (e.g., Twitter/X mention counts), and page-level organization (e.g., number of categories on a Wikipedia page) are associated with a higher hazard of correction. Conversely, a paper's established scholarly authority--a higher academic citation count--is associated with a slower time to correction. Our findings highlight how the Wikipedia community supports collaborative maintenance but leaves gaps in citation-level repair. We contribute to CSCW research by advancing our understanding of this sociotechnical vulnerability, which takes the form of a community coordination challenge, and by offering design directions to support citation credibility at scale.

cs.HC↗

arXiv2025

Multi-Granularity Adaptive Time-Frequency Attention Framework for Audio Deepfake Detection under Real-World Communication Degradations

The rise of highly convincing synthetic speech poses a growing threat to audio communications. Although existing Audio Deepfake Detection (ADD) methods have demonstrated good performance under clean conditions, their effectiveness drops significantly under degradations such as packet losses and speech codec compression in real-world communication environments. In this work, we propose the first unified framework for robust ADD under such degradations, which is designed to effectively accommodate multiple types of Time-Frequency (TF) representations. The core of our framework is a novel Multi-Granularity Adaptive Attention (MGAA) architecture, which employs a set of customizable multi-scale attention heads to capture both global and local receptive fields across varying TF granularities. A novel adaptive fusion mechanism subsequently adjusts and fuses these attention branches based on the saliency of TF regions, allowing the model to dynamically reallocate its focus according to the characteristics of the degradation. This enables the effective localization and amplification of subtle forgery traces. Extensive experiments demonstrate that the proposed framework consistently outperforms state-of-the-art baselines across various real-world communication degradation scenarios, including six speech codecs and five levels of packet losses. In addition, comparative analysis reveals that the MGAA-enhanced features significantly improve separability between real and fake audio classes and sharpen decision boundaries. These results highlight the robustness and practical deployment potential of our framework in real-world communication environments.

eess.AS↗

arXiv2025

Benchmarking Audio Deepfake Detection Robustness in Real-world Communication Scenarios

Existing Audio Deepfake Detection (ADD) systems often struggle to generalise effectively due to the significantly degraded audio quality caused by audio codec compression and channel transmission effects in real-world communication scenarios. To address this challenge, we developed a rigorous benchmark to evaluate the performance of the ADD system under such scenarios. We introduced ADD-C, a new test dataset to evaluate the robustness of ADD systems under diverse communication conditions, including different combinations of audio codecs for compression and packet loss rates. Benchmarking three baseline ADD models on the ADD-C dataset demonstrated a significant decline in robustness under such conditions. A novel Data Augmentation (DA) strategy was proposed to improve the robustness of ADD systems. Experimental results demonstrated that the proposed approach significantly enhances the performance of ADD systems on the proposed ADD-C dataset. Our benchmark can assist future efforts towards building practical and robustly generalisable ADD systems.

eess.AS↗

arXiv2025

Rethinking the Upsampling Layer in Hyperspectral Image Super Resolution

Deep learning has achieved significant success in single hyperspectral image super-resolution (SHSR); however, the high spectral dimensionality leads to a heavy computational burden, thus making it difficult to deploy in real-time scenarios. To address this issue, this paper proposes a novel lightweight SHSR network, i.e., LKCA-Net, that incorporates channel attention to calibrate multi-scale channel features of hyperspectral images. Furthermore, we demonstrate, for the first time, that the low-rank property of the learnable upsampling layer is a key bottleneck in lightweight SHSR methods. To address this, we employ the low-rank approximation strategy to optimize the parameter redundancy of the learnable upsampling layer. Additionally, we introduce a knowledge distillation-based feature alignment technique to ensure the low-rank approximated network retains the same feature representation capacity as the original. We conducted extensive experiments on the Chikusei, Houston 2018, and Pavia Center datasets compared to some SOTAs. The results demonstrate that our method is competitive in performance while achieving speedups of several dozen to even hundreds of times compared to other well-performing SHSR methods.

eess.IV↗

arXiv2023

Differential impact from individual versus collective misinformation tagging on the diversity of Twitter (X) information engagement and mobility

Fears about the destabilizing impact of misinformation online have motivated individuals and platforms to respond. Individuals have increasingly challenged others' online claims with fact-checks in pursuit of a healthier information ecosystem and to break down echo chambers of self-reinforcing opinion. Using Twitter (now X) data, here we show the consequences of individual misinformation tagging: tagged posters had explored novel political information and expanded topical interests immediately prior, but being tagged caused posters to retreat into information bubbles. These unintended consequences were softened by a collective verification system for misinformation moderation. In Twitter's new feature, Community Notes, misinformation tagging was peer-reviewed by other fact-checkers before revelation to the poster. With collective misinformation tagging, posters were less likely to retreat from diverse information engagement. Detailed comparison demonstrated differences in toxicity, sentiment, readability, and delay in individual versus collective misinformation tagging messages. These findings provide evidence for differential impacts from individual versus collective moderation strategies on the diversity of information engagement and mobility across the information ecosystem.

cs.CY↗