Do We Need Distinct Representations for Every Speech Token? Unveiling and Exploiting Redundancy in Large Speech Language Models

Documenti analoghi

• Understanding the Modality Gap: An Empirical Study on the Speech-Text Alignment Mechanism of Large Speech Language Models
  di: Xiang, Bajian, et al.
  Pubblicazione: (2025)
• Advancing Speech Language Models by Scaling Supervised Fine-Tuning with Over 60,000 Hours of Synthetic Speech Dialogue Data
  di: Zhao, Shuaijiang, et al.
  Pubblicazione: (2024)
• SpeechR: A Benchmark for Speech Reasoning in Large Audio-Language Models
  di: Yang, Wanqi, et al.
  Pubblicazione: (2025)
• Every Token Counts: Generalizing 16M Ultra-Long Context in Large Language Models
  di: Hu, Xiang, et al.
  Pubblicazione: (2025)
• Measuring the Redundancy of Decoder Layers in SpeechLLMs
  di: Moumen, Adel, et al.
  Pubblicazione: (2026)