Cognitive deficits may occur after mild traumatic brain injury (mTBI), but effective treatment modalities are presently unavailable. Caloric restriction (CR) has beneficial effects on neurodegenerative diseases and brain injury. However, the underlying mechanisms have not yet been clearly defined. Therefore, the aim of the present study was to investigate the short-term effects of CR treatment on cognitive function in mice after mTBI. Forty-five 12-week-old C57/BL6 mice were subjected to closed-head mTBI using a weight drop device. The mice were then randomly divided into three groups according to their diet for 30 days: the normal calorie group (mTBI + NC group, n = 15), the caloric restriction group (mTBI + CR group, n = 15), and the high energy group (mTBI + HE group, n = 15). After 30 days, the Morris water maze test was performed to evaluate learning abilities. Nissl staining, immunohistochemistry, and western blotting were used to monitor pathological changes and changes in autophagy-associated proteins in the hippocampus. The average escape latency was significantly shorter in the mTBI + CR group than in the mTBI + NC and mTBI + HE groups, and the number of target platform crossings in the mTBI + CR group was significantly higher than in the other two groups. In the hippocampus, the expression of GFAP and mTOR was increased in the mTBI + HE group and decreased in the mTBI + CR group. Conversely, the expression of LC3B was decreased in the mTBI + HE group and increased in the mTBI + CR group. Our findings suggest that short-term CR after mTBI may ameliorate cognitive dysfunction induced by mTBI by increasing the level of autophagy and suppressing astrocyte activation.