SLP-L6.2
KEEPING MODELS LISTENING: SEGMENT- AND TIME-AWARE ATTENTION RESCALING AT DECODING TIME
Hangyu Du, National University of Singapore, China; Jingxing Zhong, Fuzhou University, China
Session:
SLP-L6: Speech LLM: Alignment & Decoding Oral
Track:
Speech and Language Processing [SL]
Location:
Room 115
Presentation Time:
Wed, 6 May, 09:20 - 09:40
Presentation
Discussion
Resources
No resources available.
Session SLP-L6
SLP-L6.1: AQUA-Bench: Beyond Finding Answers to Knowing When There Are None in Audio Question Answering
Chun-Yi Kuan, Hung-yi Lee, National Taiwan University, Taiwan
SLP-L6.2: KEEPING MODELS LISTENING: SEGMENT- AND TIME-AWARE ATTENTION RESCALING AT DECODING TIME
Hangyu Du, National University of Singapore, China; Jingxing Zhong, Fuzhou University, China
SLP-L6.3: Still Thinking or Stopped Talking? Dialogue Silence Intention Classification Using Multimodal Large Language Model
Muyun Wu, Zi Haur Pang, Koji Inoue, Tatsuya Kawahara, Kyoto university, Japan
SLP-L6.4: SPEECHMAPPER: SPEECH-TO-TEXT EMBEDDING PROJECTOR FOR LLMS
Biswesh Mohapatra, Inria Paris, France; Marcely Zanon boito, Ioan Calapodescu, NAVER LABS Europe, France
SLP-L6.5: UNDERSTANDING TEXTUAL CAPABILITY DEGRADATION IN SPEECH LLMS VIA PARAMETER IMPORTANCE ANALYSIS
Chao Wang, Rui-Chen Zheng, Yang Ai, Zhen-Hua Ling, University of Science and Technology of China, China
SLP-L6.6: TASU: TEXT-ONLY ALIGNMENT FOR SPEECH UNDERSTANDING
Jing Peng, Yi Yang, Shanghai Jiao Tong University, China; Xu Li, AIspeech, China; Yu Xi, Shanghai Jiao Tong University, China; Quanwei Tang, Suzhou University, China; Yangui Fang, Huazhong Techonology University, China; Junjie Li, Kai Yu, Shanghai Jiao Tong University, China
Contacts