AccKV: Towards Efficient Audio-Video LLMs Inference via Adaptive-Focusing and Cross-Calibration KV Cache Optimization

Documenti analoghi

• PureKV: Plug-and-Play KV Cache Optimization with Spatial-Temporal Sparse Attention for Vision-Language Large Models
  di: Jiang, Zhonghua, et al.
  Pubblicazione: (2025)
• FlowMM: Cross-Modal Information Flow Guided KV Cache Merging for Efficient Multimodal Context Inference
  di: Li, Kunxi, et al.
  Pubblicazione: (2025)
• MadaKV: Adaptive Modality-Perception KV Cache Eviction for Efficient Multimodal Long-Context Inference
  di: Li, Kunxi, et al.
  Pubblicazione: (2025)
• DreamFoley: Scalable VLMs for High-Fidelity Video-to-Audio Generation
  di: Li, Fu, et al.
  Pubblicazione: (2025)
• Listen, Pause, and Reason: Toward Perception-Grounded Hybrid Reasoning for Audio Understanding
  di: Wang, Jieyi, et al.
  Pubblicazione: (2026)