From the 1 of 5 linked papers with an AI index.
5 papers
OmniDelta: Skill-Driven Budget Allocation for Token Compression in OmniLLMs
Haoyang Huang, Wenjie Huang, Tianqi Xu +14
OmniDelta is a training-free framework that dynamically allocates token budgets for audio and video inputs in omni-modal large language models, using skill pools and local complexi…
HDRFace: Rethinking Face Restoration with High-Dimensional Representation
Zirui Wang, Xianhui Lin, Yi Dong +7
Face restoration under complex degradations still remains an ill-posed inverse problem due to severe information loss. Although diffusion models benefit from strong generative prio…
TileQ: Efficient Low-Rank Quantization of Mixture-of-Experts with 2D Tiling
Hongyaoxing Gu, Xinzhe Chen, Lijuan Hu +1
Mixture-of-Experts (MoE) models achieve remarkable performance by sparsely activating specialized experts, yet their massive parameters in experts pose significant challenges for d…
LoPRo: Enhancing Low-Rank Quantization via Permuted Block-Wise Rotation
Hongyaoxing Gu, Lijuan Hu, Liye Yu +2
Post-training quantization (PTQ) enables effective model compression while preserving relatively high accuracy. Current weight-only PTQ methods primarily focus on the challenging s…
FLRQ: Faster LLM Quantization with Flexible Low-Rank Matrix Sketching
Hongyaoxing Gul, Lijuan Hu, Shuzi Niu +1
Traditional post-training quantization (PTQ) is considered an effective approach to reduce model size and accelerate inference of large-scale language models (LLMs). However, exist…