Search papers, labs, and topics across Lattice.
5
0
8
2
High variance in draft quality is tackled head-on, leading to a remarkable 66% increase in throughput for speculative decoding in large language models.
AdaptSim transforms CRS evaluation by generating realistic dialogues that adapt to diverse user styles and preferences, overcoming traditional simulator limitations.
MXFP4 quantization just got a whole lot better: BATQuant recovers up to 96.43% of full-precision performance in LLMs and MLLMs, even under aggressive W4A4KV16 settings, by preventing outlier propagation across quantization blocks.
Existing deep feature selection methods for recommender systems suffer from layer, baseline, and approximation biases, leading to suboptimal feature selection, which FairFS effectively mitigates.
User-defined response prefixes in LLMs are a major safety risk, enabling CoT attacks to achieve near-perfect success rates on some models.
