作者位次 #6/8ACL 2026Poster· Natural Language Processing
Learning More from Less: Exploiting Counterfactuals for Data-Efficient Chart Understanding
Jianzhu Bao · Haozhen Zhang · Kuicai Dong · Bozhi Wu · Sarthak Ketanbhai Modi · Zi Pong Lim · Yon Shin Teo · Wenya Wang
摘要
Vision-Language Models (VLMs) have demonstrated remarkable progress in chart understanding, largely driven by supervised fine-tuning (SFT) on increasingly large synthetic datasets.However, scaling SFT data alone is inefficient and overlooks a key property of charts: charts are programmatically ge...