CogCoM: Train Large Vision-Language Models Diving into Details through Chain of Manipulations
CoRR(2024)
摘要
Vision-Language Models (VLMs) have demonstrated their widespread viability
thanks to extensive training in aligning visual instructions to answers.
However, this conclusive alignment leads models to ignore critical visual
reasoning, and further result in failures on meticulous visual problems and
unfaithful responses. In this paper, we propose Chain of Manipulations, a
mechanism that enables VLMs to solve problems with a series of manipulations,
where each manipulation refers to an operation on the visual input, either from
intrinsic abilities (e.g., grounding) acquired through prior training or from
imitating human-like behaviors (e.g., zoom in). This mechanism encourages VLMs
to generate faithful responses with evidential visual reasoning, and permits
users to trace error causes in the interpretable paths. We thus train CogCoM, a
general 17B VLM with a memory-based compatible architecture endowed this
reasoning mechanism. Experiments show that our model achieves the
state-of-the-art performance across 8 benchmarks from 3 categories, and a
limited number of training steps with the data swiftly gains a competitive
performance. The code and data are publicly available at
https://github.com/THUDM/CogCoM.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要