Publications (7)
Apple Intelligence Foundation Language Models
Tom Gunter, Zirui Wang, Chong Wang +152
We present foundation language models developed to power Apple Intelligence features, including a ~3 billion parameter model designed to run efficiently on devices and a large serv…
FastSR-NeRF: Improving NeRF Efficiency on Consumer Devices with A Simple Super-Resolution Pipeline
Chien-Yu Lin, Qichen Fu, Thomas Merth +2
Super-resolution (SR) techniques have recently been proposed to upscale the outputs of neural radiance fields (NeRF) and generate high-quality images with enhanced inference speeds…
LazyLLM: Dynamic Token Pruning for Efficient Long Context LLM Inference
Qichen Fu, Minsik Cho, Thomas Merth +3
The inference of transformer-based large language models consists of two sequential stages: 1) a prefilling stage to compute the KV cache of prompts and generate the first token, a…
Do Compressed LLMs Forget Knowledge? An Experimental Study with Practical Implications
Duc N. M Hoang, Minsik Cho, Thomas Merth +2
Compressing Large Language Models (LLMs) often leads to reduced performance, especially for knowledge-intensive tasks. In this work, we dive into how compression damages LLMs' inhe…
On the Efficacy of Multi-scale Data Samplers for Vision Applications
Elvis Nunez, Thomas Merth, Anish Prabhu +4
Multi-scale resolution training has seen an increased adoption across multiple vision tasks, including classification and detection. Training with smaller resolutions enables faste…
SPIN: An Empirical Evaluation on Sharing Parameters of Isotropic Networks
Chien-Yu Lin, Anish Prabhu, Thomas Merth +4
Recent isotropic networks, such as ConvMixer and vision transformers, have found significant success across visual recognition tasks, matching or outperforming non-isotropic convol…
Superposition Prompting: Improving and Accelerating Retrieval-Augmented Generation
Thomas Merth, Qichen Fu, Mohammad Rastegari +1
Despite the successes of large language models (LLMs), they exhibit significant drawbacks, particularly when processing long contexts. Their inference cost scales quadratically wit…