A Revision-Based Generation Architecture for Reporting Facts in their Historical Context

Natural language reports generated by existing systems ignore the historical context of the facts and events they relate. In this paper, I argue that going beyond this limitation requires abandoning the pipelined architecture of existing report generators. I propose a new architecture in which a rst draft of the report is organized around new information and then incrementally revised to opportunistically add related historical information. This type of information-adding revision allows to elaborate inside a clause or a nominal while taking into account surface structure constraints from any other portion of the report. In addition to providing the additional exibility required to convey historical information, the proposed architecture constitutes an interesting testbed to investigate a wide range of open questions concerning content planning below the sentence level, generation with revision and generation architecture.