AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

ChartCap: Mitigating Hallucination of Dense Chart Captioning

Junyoung Lim, Jaewoo Ahn, Gunhee Kim

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Multimodal LLM benchmarks · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fine-tune on ChartCapPhi3.5-Vision-4B · ChartCap test setsacreBLEU: 8.41 -> 23.82Improved
fine-tune on ChartCapInternVL2.5-8B · ChartCap test setsacreBLEU: 3.60 -> 19.47Improved
fine-tune on ChartCapPhi3.5-Vision-4B · ChartCap test setOCRScore: 0.4875 -> 0.5179Improved
fine-tune on 250K original captions instead of ChartCap (ablation)Phi3.5-Vision-4B · ChartCap test setsacreBLEU: 8.41 -> 0.09Hurt
fine-tune on ChartSumm instead of ChartCap (ablation)Phi3.5-Vision-4B · ChartCap test setsacreBLEU: 8.41 -> 1.31Hurt
fine-tune on ChartSumm instead of ChartCap (ablation)Phi3.5-Vision-4B · VisText test (zero-shot)Visual Consistency Score: 0.8814 -> 0.8677Hurt
fine-tune on ChartCapQwen2.5-VL-7B · VisText test (zero-shot)Visual Consistency Score: 0.9044 -> 0.9328Improved
fine-tune on ChartCapQwen2.5-VL-7B · Chart-to-Text PEW subset (zero-shot)Visual Consistency Score: 0.7739 -> 0.8084Improved
fine-tune on ChartCapPhi3.5-Vision-4B · VisText test (zero-shot)Visual Consistency Score: 0.8814 -> 0.9382Improved
fine-tune on ChartCapPhi3.5-Vision-4B · VisText test (zero-shot)OCRScore: 0.3414 -> 0.3826Improved
fine-tune on ChartCapInternVL2.5-8B · VisText test (zero-shot)OCRScore: 0.3172 -> 0.3360Mixed
train on ChartCap data rather than ChartSumm captionscaptioning dataset · Dataset caption quality (1K samples)Visual Consistency Score: 0.9008 -> 0.9089Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy