Can Large Language Models Keep Up? Benchmarking Online Adaptation to Continual Knowledge Streams
This benchmark addresses the critical problem of LLMs maintaining accuracy in dynamic real-world contexts for developers and users of LLMs.
This paper introduces OAKS, a benchmark for evaluating the online adaptation capabilities of Large Language Models (LLMs) to continually evolving knowledge streams. The evaluation of 14 models on OAKS-BABI and OAKS-Novel datasets revealed significant limitations, with models failing to adapt robustly and exhibiting delays in state-tracking and susceptibility to distraction.
LLMs operating in dynamic real-world contexts often encounter knowledge that evolves continuously or emerges incrementally. To remain accurate and effective, models must adapt to newly arriving information on the fly. We introduce Online Adaptation to Continual Knowledge Streams(OAKS) to evaluate this capability, establishing a benchmark for online adaptation over streaming, continually updating knowledge. Specifically, the benchmark is structured as a sequence of fine-grained context chunks where facts change dynamically across time intervals. OAKS comprises two datasets: OAKS-BABI and OAKS-Novel, where individual facts evolve multiple times across context chunks. These datasets include dense annotations to measure whether models track changes accurately. Evaluating 14 models with varied inference approaches, we observe significant limitations in current methodologies. Both state-of-the-art models and agentic memory systems fail to adapt robustly on OAKS, demonstrating delays in state-tracking and susceptibility to distraction within streaming environments.