Wushao Wen

CIDER: A Causal Cure for Brand-Obsessed text-to-Image Models

Fangjian Shen, Zifeng Liang, Chao Wang, Wushao Wen

Text-to-image (T2I) models exhibit a significant yet under-explored "brand bias", a tendency to generate contents featuring dominant commercial brands from generic prompts, posing ethical and legal risks. We propose CIDER, a novel, model-agnostic framework to mitigate bias at inference-time through prompt refinement …

cn (code pays fourni par la source)

Spectral-Adaptive Individual Preference Learning with Bootstrap Alignment for Multimodal Recommendation

Hui Wang, Jinghui Qin, Wushao Wen

cn (code pays fourni par la source)

TAGT: A Time-Aware Graph Transformer for Securing Time-Sensitive Networking

Liang Qi, Wushao Wen

The precision of time synchronization is the cornerstone of safety-critical applications in Automotive Ethernet and Time-Sensitive Networking (TSN). The IEEE 802.1AS protocol (gPTP), while widely adopted, lacks built-in security mechanisms, making it vulnerable to sophisticated delay attacks and packet manipulation. Existing Deep …

cn (code pays fourni par la source)

UPRPRC: Unified Pipeline for Reproducing Parallel Resources -- Corpus from the United Nations

Qingyi Lu, Fangjian Shen, Z.X. Tang, Qiang Liu et autres

The quality and accessibility of multilingual datasets are crucial for advancing machine translation. However, previous corpora built from United Nations documents have suffered from issues such as opaque process, difficulty of reproduction, and limited scale. To address these challenges, we introduce a …

AssoCiAm: A Benchmark for Evaluating Association Thinking while Circumventing Ambiguity

Yifan Liu, Wenkuan Zhao, Shanshan Zhong, Jinghui Qin et autres

Recent advancements in multimodal large language models (MLLMs) have garnered significant attention, offering a promising pathway toward artificial general intelligence (AGI). Among the essential capabilities required for AGI, creativity has emerged as a critical trait for MLLMs, with association serving as its …

Transformer-based Reinforcement Learning for Net Ordering in Detailed Routing

Zhanwen Zhou, Hankz Hankui Zhuo, Jinghua Zhou, Wushao Wen

With feature size shrinking and design complexity increasing, detailed routing has become a crucial challenge in VLSI design. Although detailed routers have been proposed to judiciously handle hard-to-access pins and various design rules, their performances are sensitive to the order of nets …

cn (code pays fourni par la source)

CrossGCL: cross-view graph contrastive learning with dual tasks for drug recommendation

Wushao Wen, Liang Wang, Lihuang Fang, Qiangpu Chen et autres

cn (code pays fourni par la source)

DVIB: Towards Robust Multimodal Recommender Systems via Variational Information Bottleneck Distillation

Wenkuan Zhao, Shanshan Zhong, Wushao Wen, Jinghui Qin et autres

In multimodal recommender systems (MRS), integrating various modalities helps to model user preferences and item characteristics more accurately, thereby assisting users in discovering items that match their interests. Although the introduction of multimodal information offers opportunities for performance improvement, it will increase …

cn, us (code pays fourni par la source)

SHARP: A Study on Safe Heterogeneous Agent Reinforcement Learning Paradigm for 5G-TSN Traffic Scheduling

S. M. Tan, Chongwu Dong, Wushao Wen

With the rapid development of the Industrial Internet of Things (IIoT), 5G-TSN (5G-Time Sensitive Network) has emerged as a promising solution in IIoT to address stringent requirements for low latency and high reliability in industrial networks. However, it introduces a critical challenge …

cn (code pays fourni par la source)

A Reinforcement Learning-Based PQoS Framework for V2X Networks in NS-3

Wushao Wen

The emergence of the sixth-generation mobile networks (6G) introduces massive ultra-reliable and low-latency communications (mURLLC), critical for applications like teleoperated driving, which demand ultra-high reliability and sub-50 ms latency. This paper addresses the challenge of Predictive Quality of Service (PQoS) in Vehicle-to-Everything …

cn (code pays fourni par la source)

Boundary-Driven Table-Filling with Cross-Granularity Contrastive Learning for Aspect Sentiment Triplet Extraction

Qingling Li, Wushao Wen, Jinghui Qin

The Aspect Sentiment Triplet Extraction (ASTE) task aims to extract aspect terms, opinion terms, and their corresponding sentiment polarity from a given sentence. It remains one of the most prominent subtasks in fine-grained sentiment analysis. Most existing approaches frame triplet extraction as …

cn (code pays fourni par la source)

Delay-Sensitive Task Offloading With Edge Caching Through Martingale-Based Deep Reinforcement Learning

Chongwu Dong, Weidong Li, Zhi Zhou, Xu Chen et autres

In the forthcoming era of 6G networks, delay-sensitive applications for Internet of Things (IoT) are poised to become the prevailing services with ultra-reliable and low-latency (URLLC) requirements. Unlike traditional video caching, IoT-based edge caching faces unique challenges due to diverse data types, …

cn (code pays fourni par la source)