Other people with similar names: Xin Yao, Xin Yao

Unverified author pages with similar names: Xin Yao


2026

Large Vision-Language Models (LVLMs) have recently demonstrated remarkable progress, yet hallucination remains a critical barrier, particularly in chart understanding, which requires sophisticated perceptual and cognitive abilities and demands rigorous factual accuracy. While prior work has investigated hallucinations and chart comprehension independently, their intersection remains largely unexplored. To address this gap, we present ChartHal, a benchmark that features a fine-grained taxonomy of hallucination scenarios in chart understanding, along with a human-validated dataset of 1,062 samples. Our evaluation shows that state-of-the-art LVLMs suffer from severe hallucinations on ChartHal, with even GPT-5 achieving only 34.46% accuracy. Further analysis reveals that questions involving information absent from or contradictory to charts are especially likely to trigger hallucinations. We also evaluate several representative hallucination mitigation methods and find that they exhibit notable limitations, which underscores the urgent need for more robust chart-specific mitigation strategies.

2025

Chart2code has recently received significant attention in the multimodal community due to its potential to reduce the burden of visualization and promote a more detailed understanding of charts. However, existing Chart2code-related training datasets suffer from at least one of the following issues: (1) limited scale, (2) limited type coverage, and (3) inadequate complexity. To address these challenges, we seek more diverse sources that better align with real-world user distributions and propose dual data synthesis pipelines: (1) synthesize based on online plotting code. (2) synthesize based on chart images in the academic paper. We create a large-scale Chart2code training dataset Chart2code53, including 53 chart types, 130K Chart-code pairs based on the pipeline. Experimental results demonstrate that even with few parameters, the model finetuned on Chart2code53 achieves state-of-the-art performance on multiple Chart2code benchmarks within open-source models.