6 papers
These Are Not All the Features You Are Looking For: A Fundamental Bottleneck in Supervised Pretraining
Xingyu Alice Yang, Jianyu Zhang, Léon Bottou
Transfer learning is widely used to adapt large pretrained models to new tasks with only a small amount of new data. However, a challenge persists -- the features from the original…
Memory Mosaics at scale
Jianyu Zhang, Léon Bottou
Memory Mosaics [Zhang et al., 2025], networks of associative memories, have demonstrated appealing compositional and in-context learning capabilities on medium-scale networks (GPT-…
A Single Character can Make or Break Your LLM Evals
Jingtong Su, Jianyu Zhang, Karen Ullrich +2
Common Large Language model (LLM) evaluations rely on demonstration examples to steer models' responses to the desired style. While the number of examples used has been studied and…
Memory Mosaics
Jianyu Zhang, Niklas Nolte, Ranajoy Sadhukhan +2
Memory Mosaics are networks of associative memories working in concert to achieve a prediction task of interest. Like transformers, memory mosaics possess compositional capabilitie…
Fine-tuning with Very Large Dropout
Jianyu Zhang, Léon Bottou
It is impossible today to pretend that the practice of machine learning is always compatible with the idea that training and testing data follow the same distribution. Several auth…
MagicPIG: LSH Sampling for Efficient LLM Generation
Zhuoming Chen, Ranajoy Sadhukhan, Zihao Ye +8
Large language models (LLMs) with long context windows have gained significant attention. However, the KV cache, stored to avoid re-computation, becomes a bottleneck. Various dynam…