Can Foundation Models Talk Causality?

Willig, Moritz; Zečević, Matej; Dhami, Devendra Singh; Kersting, Kristian

doi:10.48550/arxiv.2206.10591

Cited by 3 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Though there is growing interest, to date, there are few studies exploring the utility of LLM in causal diagram development. A recent study by Willig et al [2022] compared the performance of three query LLMs in making causal graph predictions in a general context. There also has been some interesting works applying causal inference in the LLM context.…”

Section: Discussionmentioning

confidence: 99%

Can large language models build causal graphs?

Long¹,

Schuster²,

Piche³

et al. 2023

Preprint

View full text Add to dashboard Cite

Building causal graphs can be a laborious process. To ensure all relevant causal pathways have been captured, researchers often have to discuss with clinicians and experts while also reviewing extensive relevant medical literature. By encoding common and medical knowledge, large language models (LLMs) represent an opportunity to ease this process by automatically scoring edges (i.e., connections between two variables) in potential graphs. LLMs however have been shown to be brittle to the choice of probing words, context, and prompts that the user employs. In this work, we evaluate if LLMs can be a useful tool in complementing causal graph development.

show abstract

Section: Discussionmentioning

confidence: 99%

Can large language models build causal graphs?

Long¹,

Schuster²,

Piche³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

Knowledge Graph Structure as Prompt: Improving Small Language Models Capabilities for Knowledge-Based Causal Discovery

Susanti,

Färber

2024

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Molecular causality in the advent of foundation models

Lobentanzer,

Rodriguez-Mier,

Bauer

et al. 2024

Mol Syst Biol

View full text Add to dashboard Cite

Correlation is not causation: this simple and uncontroversial statement has far-reaching implications. Defining and applying causality in biomedical research has posed significant challenges to the scientific community. In this perspective, we attempt to connect the partly disparate fields of systems biology, causal reasoning, and machine learning to inform future approaches in the field of systems biology and molecular medicine.

show abstract

Can Foundation Models Talk Causality?

Cited by 3 publications

References 0 publications

Can large language models build causal graphs?

Can large language models build causal graphs?

Knowledge Graph Structure as Prompt: Improving Small Language Models Capabilities for Knowledge-Based Causal Discovery

Molecular causality in the advent of foundation models

Contact Info

Product

Resources

About