arXiv cs.CL· Somraj Gautam, Gaurav Harit·· 12 小时前AI 评分27
Cycle-of-Thought：面向文档图像表格视觉问答的自评估迭代推理框架
Cycle-of-Thought: Self-Evaluative Iterative Reasoning for Table Visual Question Answering in Document Images
AI 导读
Self-RefineVQA 是一个无需训练、在推理阶段自我纠错的表格 VQA 框架，让单个 VLM 迭代扮演生成器、批评者和精炼者，通过提示词引导的重新评估修正空间错位与逻辑幻觉。在表格 VQA 基准上，该方法在单语和跨语言理解中均取得稳定准确率提升，2B-7B 参数的小模型提升尤为明显，可媲美更大模型。其推理成本线性增长，但省去了人工标注的微调瓶颈，数据效率更高。
正文 · 原文
Abstract:Recent advances in Vision-Language Models (VLMs) have significantly improved Visual Question Answering (VQA) performance, yet their reasoning over structured content such as tables remains inconsistent, often plagued by spatial misalignment and logical hallucinations. Existing approaches for improving table VQA typically rely on architectural modifications or expensive supervised fine-tuning (SFT), which impose high barriers to entry due to the need for large-scale, expert-annotated datasets and significant computational resources. In this work, we propose Self-RefineVQA, a training-free, inference-time self-correction framework that improves table VQA performance without additional labeled data or model updates. Inspired by human revision behavior, our method enables a single VLM to iteratively act as a generator, critic, and refiner, forming a closed self-feedback loop that corrects common reasoning errors through prompt-guided re-evaluation. Experiments on tabular VQA benchmarks demonstrate that Self-RefineVQA yields consistent accuracy gains in monolingual and cross-lingual understanding, with a particularly significant boost for small models, enabling lightweight models (e.g., 2B-7B parameters) to rival the performance of significantly larger counterparts. Furthermore, we provide a comparative complexity analysis demonstrating that while inference cost increases linearly, our framework offers superior data efficiency and a lower cumulative cost by eliminating the prohibitive annotation bottleneck of human labeling required for fine-tuning. Our results highlight the potential of prompt-based self-improvement as a scalable, accessible paradigm for reliable table reasoning in VLMs.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.32423 [cs.CV] |
| (or arXiv:2610.32423v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.32423 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Somraj Gautam [view email]
[v1]
Fri, 23 Oct 2026 09:50:37 UTC (1,591 KB)
来源：arXiv cs.CL · arxiv.org