Summary of Towards Compositionality in Concept Learning, by Adam Stein et al.
Towards Compositionality in Concept Learning
by Adam Stein, Aaditya Naik, Yinjun Wu, Mayur Naik, Eric Wong
First submitted to arxiv on: 26 Jun 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: 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 This paper explores concept-based interpretability methods for foundation models, aiming to decompose their embeddings into high-level concepts. The authors identify the importance of compositional concepts, which can explain a full sample by composing individual concepts. To achieve this, they propose Compositional Concept Extraction (CCE) and evaluate it on five datasets covering image and text data. CCE outperforms baselines in finding compositional concept representations and achieves better accuracy in four downstream classification tasks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us understand how foundation models work by breaking down their “thoughts” into simple ideas called concepts. The authors want to make these concepts useful by making them “compositional”, meaning they can be combined to explain a whole picture or text. To do this, they developed a new way to find these compositional concepts and tested it on many different datasets. It worked better than other methods in most cases! |
Keywords
* Artificial intelligence * Classification
