CASCADE enables LLMs to continually adapt at deployment via case-based episodic memory and contextual bandits, improving macro-averaged success by 20.9% over zero-shot on 16 tasks spanning medicine, law, code, and robotics.
Online multi-llm selection via contextual bandits under unstructured context evolution
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
citing papers explorer
-
CASCADE: Case-Based Continual Adaptation for Large Language Models During Deployment
CASCADE enables LLMs to continually adapt at deployment via case-based episodic memory and contextual bandits, improving macro-averaged success by 20.9% over zero-shot on 16 tasks spanning medicine, law, code, and robotics.
- Latency-Quality Routing for Functionally Equivalent Tools in LLM Agents