Search papers, labs, and topics across Lattice.
2
0
3
Particle filtering reveals a fundamental limit to inference-time sampling methods for LLMs, suggesting that simply increasing the number of samples has diminishing returns.
Sampling from diffusion models with quadratic rewards can be surprisingly hard: negative-definite tilts are intractable even in simple cases, while a new algorithm makes low-rank positive-definite tilts tractable.
