Abstract: Vector-Quantization (VQ) based discrete generative models are widely used to learn powerful high-quality (HQ) priors for blind image restoration (BIR). In this paper, we diagnose the ...
EntropyKV is an online, 100% attention-free key-value (KV) cache eviction policy designed to run long-context LLMs on consumer-grade GPUs. Traditional KV cache compression methods (e.g., H2O, SnapKV) ...
Abstract: 3D Gaussian Splatting (3DGS) is rapidly gaining popularity for its photorealistic rendering quality and real-time performance, but it generates massive amounts of data. Hence compressing ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results