11 papers
Advancing Electrolaryngeal Speech Enhancement Through Speech-Text Representation Learning
Ding Ma, Jinyi Mi, Fengji Li +5
Objective: laryngectomees depend on an electromechanical device to generate electrolaryngeal (EL) speech. Compared with normal speech, EL speech suffers from severe distortion, lim…
Evaluating Large Language Models in Scientific Discovery
Zhangde Song, Jieyu Lu, Yuanqi Du +53
Large language models (LLMs) are increasingly applied to scientific research, yet prevailing science benchmarks probe decontextualized knowledge and overlook the iterative reasonin…
VisionCreator: A Native Visual-Generation Agentic Model with Understanding, Thinking, Planning and Creation
Jinxiang Lai, Zexin Lu, Jiajun He +11
Visual content creation tasks demand a nuanced understanding of design conventions and creative workflows-capabilities challenging for general models, while workflow-based agents l…
A Comprehensive Study on the Effectiveness of ASR Representations for Noise-Robust Speech Emotion Recognition
Xiaohan Shi, Jiajun He, Xingfeng Li +1
This paper proposes an efficient attempt to noisy speech emotion recognition (NSER). Conventional NSER approaches have proven effective in mitigating the impact of artificial noise…
BudgetLeak: Membership Inference Attacks on RAG Systems via the Generation Budget Side Channel
Hao Li, Jiajun He, Guangshuo Wang +3
Retrieval-Augmented Generation (RAG) enhances large language models by integrating external knowledge, but reliance on proprietary or sensitive corpora poses various data risks, in…
M4SER: Multimodal, Multirepresentation, Multitask, and Multistrategy Learning for Speech Emotion Recognition
Jiajun He, Xiaohan Shi, Cheng-Hung Hu +3
Multimodal speech emotion recognition (SER) has emerged as pivotal for improving human-machine interaction. Researchers are increasingly leveraging both speech and textual informat…