Search for a command to run...
Can Large Language Models Keep Up? Benchmarking Online Adaptation to Continual Knowledge Streams