Product Experimentation with Instrumental Variables: Unconfounding LLM Routing Decisions in Python⭐9

Learn how to overcome confounded routing decisions in LLM gateways using instrumental variables for unbiased causal inference in Python product experiments.

CausalGate: Causal Importance Distillation for Transformer Module Pruning⭐8

CausalGate uses causal importance distillation to prune transformer modules, reducing compute and latency in LLMs without runtime overhead.