Search papers, labs, and topics across Lattice.
2
0
4
Forget massive models: fine-tuning a small 4B parameter model on carefully curated data can match or even approach the performance of 100B+ parameter LLMs in program verification tasks.
Provably minimal explanations for neural networks are now tractable: a new algorithm slashes the query complexity for Neural Additive Models from exponential to logarithmic.
