Nilesh Auradkar

Understanding and Mitigating Inference-Time Overreliance Using Agentic Memory

Agentic memory allows LLM agents to reuse past experience, yet retrieved memories can also distort inference even when they are benign, correctly stored, and appropriately retrieved. We study this failure mode, which we call memory over-reliance. Across benchmarks and memory architectures, we find that memory is useful...