Why Is Table Extraction with VLM Models Still Challenging? [D]

Table extraction using Vision Language Models (VLM) presents unique challenges, particularly when dealing with complex formats like borderless tables or those with numerous columns. Users often find traditional tools insufficient for converting PDFs to Markdown, especially in financial contexts. While some solutions, such as LandingAI, show promise, they often come with a cost. Many are seeking effective open-source alternatives that can simplify this process. If you have recommendations or insights into viable options, your expertise could greatly benefit those navigating these challenges.