Large reasoning models (LRMs) achieve strong accuracy through test-time scaling, generating longer chains of thought or sampling multiple solutions, but at steep costs in tokens and latency. We argue that memory is a core ingredient for efficient reasoning: when evidence already exists, models should think less by reusing structured memory instead of recomputing derivations. We present ENGRAM-R, an inference-time memory layer that integrates typed retrieval with compact fact card representations and explicit citation control. On the LoCoMo benchmark, ENGRAM-R reduces input tokens by 85% and reasoning tokens by 75% compared to full context while maintaining high accuracy. On a multi-hop slice of the LongMemEval benchmark, it achieves similar efficiency with substantial accuracy gains. These results show that memory is not only critical for long-horizon correctness but also a practical lever for efficient reasoning under tight compute, memory, and latency budgets.

翻译：大型推理模型（LRMs）通过测试时扩展（如生成更长的思维链或采样多个解）实现了强大的准确性，但代价是高昂的令牌消耗和延迟。我们认为，内存是高效推理的核心要素：当证据已存在时，模型应通过重用结构化内存来减少计算，而非重新推导。我们提出了ENGRAM-R，一种推断时内存层，它将类型化检索与紧凑的事实卡片表示及显式引用控制相结合。在LoCoMo基准测试中，与完整上下文相比，ENGRAM-R将输入令牌减少了85%，推理令牌减少了75%，同时保持了高准确性。在LongMemEval基准测试的多跳子集上，它实现了相似的效率，并带来显著的准确性提升。这些结果表明，内存不仅对长程推理的正确性至关重要，也是在有限计算、内存和延迟预算下实现高效推理的实用杠杆。