作者位次 #4/6ACL 2026Poster· Natural Language Processing
S ci MDR : Advancing Scientific Multimodal Document Reasoning
Ziyu Chen · Yilun Zhao · Chengye Wang · Rilyn R. Han · Manasi Patwardhan · Arman Cohan
摘要
Constructing scientific multimodal document reasoning datasets for foundation model training involves an inherent trade-off among scale, faithfulness, and realism. To address this challenge, we introduce the synthesize-and-reground framework, a two-stage pipeline comprising: (1) Claim-Centric QA...