Search papers, labs, and topics across Lattice.
Department of Computer Science and Engineering, Indian Institute of Technology Bombay, Mumbai, India
2
0
4
7
Text-to-SQL models can get a 36% accuracy boost and run 2.2x faster by exploiting the predictable patterns in real-world query workloads.
LLMs can learn to balance in-context and in-weights learning, but only if you fine-tune them with a Goldilocks mix of relevant and irrelevant context examples.
