-
Cerebra: Aligning Implicit Knowledge in Interactive SQL Authoring
Cerebra aligns implicit knowledge between users and LLMs during SQL authoring by retrieving context from historical scripts and supporting iterative refinement.
-
RAG Without the Lag: Enabling "What-If" Analysis for Retrieval-Augmented Generation Pipelines
Raggy enables rapid experimentation in retrieval-augmented generation pipelines through composable primitives and interactive debugging, reducing development friction.
-
Retrieval-augmented Generation of Enhanced Trigger-action Programming Rules in Smart Home
Retrieval-augmented generation system for automated generation of complex trigger-action programming rules in smart home automation, achieving 84% accuracy through semantic exemplar matching.
-
FLAT uses formal languages to type strings
FLAT uses context-free grammars as type definitions to distinguish semantically different string encodings, enabling type-safe handling of URLs, file paths, and other structured data.
