In this paper we present results on developing robust natural language interfaces by combining shallowand partialinterpretationwith dialoguemanagement. The key issue is to reduce the e ort needed to adapt the knowledge sources for parsing and interpretation to a necessary minimum. In the paper we identifydi erent types of informationand present corresponding computational models. The approach utilizes an automatically generated lexicon which is updated with information from a corpus of simulated dialogues. The grammar is developed manually from the same knowledge sources. We also present results from evaluations that support the approach.