Few studies report on therapy to improve language comprehension in children with specific language impairment (SLI). We address this gap by measuring the effect of a systematic intervention to improve inferential comprehension using dialogic reading tasks in conjunction with pre-determined questions and cues. Sixteen children with a diagnosis of SLI aged 4-6 participated in 10 weekly treatment sessions carried out by their regular therapists. Baseline and maintenance periods were also tabulated. Two experimental measures and a standardized test revealed that children's total scores and the quality of their responses post-treatment were better than those obtained pre-treatment. However, perhaps due to the use of non-equivalent probes, this change could not be interpreted solely as a significant effect of intervention. These results nevertheless suggest that a systematically designed intervention focusing on the comprehension of specific types of questions requiring inferencing and using a carefully scaffolded cueing strategy can be beneficial.