Search papers, labs, and topics across Lattice.
4
0
11
2
Conditional memory can either enhance or hinder scientific reasoning, and knowing when to activate it is crucial for optimal performance.
LLMs can learn to synthesize data more effectively by accumulating and transferring experience across a stream of sequential synthesis tasks, opening the door to more efficient and adaptable synthetic data generation.
Freezing a Sparse Autoencoder's encoder creates a reusable "safety dictionary" that generalizes to new risks in text-to-image diffusion models, offering a more robust alternative to fixed-layer steering.
On-device LLM performance is heavily influenced by sequence length and model depth, with hardware heterogeneity creating efficiency traps that can be mitigated by architectural refinements like Multi-head Latent Attention.
