Recall with Reasoning: Chain-of-Thought Distillation for Mamba's Long-Context Memory and Extrapolation

Mamba's theoretical infinite-context potential is limited in practice when sequences far exceed training lengths. This work explores unlocking Mamba's long-context memory ability by a simple-yet-effective method, Recall with Reasoning (RwR), by distilling chain-ofthought (CoT) summarization from a teacher model. Specifically, RwR prepends these summarization as CoT prompts during fine-tuning, teaching Mamba to actively recall and reason over long contexts. Experiments on LONG-MEMEVAL and HELMET show RwR boosts Mamba's long-context performance against comparable Transformer/hybrid baselines under similar pretraining conditions, while preserving short-context capabilities, all without architectural changes. Method In this section, we present Recall with Reasoning (RwR). The overview of the framework is presented in Figure 2 .