Summary of Do Vision & Language Decoders Use Images and Text Equally? How Self-consistent Are Their Explanations?, by Letitia Parcalabescu and Anette Frank
Do Vision & Language Decoders use Images and Text equally? How Self-consistent are their Explanations?
by Letitia Parcalabescu, Anette Frank
First submitted to arxiv on: 29 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper investigates how vision and language model (VLM) decoders utilize input modalities when generating answers or explanations. It finds that most VLMs are less self-consistent than LLMs, but image contributions are more important for generating explanations compared to answers, especially in CoT settings. The study also provides an updated benchmarking of state-of-the-art VL decoders on the VALSE benchmark, showing that they still struggle with certain phenomena. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary VLMs can answer questions and provide natural language explanations. But do they rely more heavily on vision or text when generating answers or explanations? This paper looks at how well VLMs use their input modalities to generate answers and explanations. It finds that while most VLMs are not as consistent as LLMs, the images contribute more to explanation generation than answer generation. |
Keywords
» Artificial intelligence » Language model
