REVIEW 5 cited by
Understanding Causality with Large Language Models: Feasibility and Opportunities
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
We assess the ability of large language models (LLMs) to answer causal questions by analyzing their strengths and weaknesses against three types of causal question. We believe that current LLMs can answer causal questions with existing causal knowledge as combined domain experts. However, they are not yet able to provide satisfactory answers for discovering new knowledge or for high-stakes decision-making tasks with high precision. We discuss possible future directions and opportunities, such as enabling explicit and implicit causal modules as well as deep causal-aware LLMs. These will not only enable LLMs to answer many different types of causal questions for greater impact but also enable LLMs to be more trustworthy and efficient in general.
Forward citations
Cited by 5 Pith papers
-
Linear-LLM-SCM: Benchmarking LLMs for Coefficient Elicitation in Linear-Gaussian Causal Models
LLMs asked to fill in linear-Gaussian causal equations give inaccurate, unstable, and perturbation-sensitive coefficients; the open-source Linear-LLM-SCM benchmark measures this, with Gemini 2.5 Flash leading on scale...
-
Unveiling Causal Reasoning in Large Language Models: Reality or Mirage?
LLMs perform much worse on causal questions built from post-cutoff news articles, suggesting their apparent causal skill is mostly memorization, and a general-knowledge prompt method only partly closes the gap.
-
LLM Cannot Discover Causality, and Should Be Restricted to Non-Decisional Support in Causal Discovery
LLMs are unreliable causal reasoners, so they should be limited to non-decisional search support in causal discovery algorithms.
-
Paths to Causality: Finding Informative Subgraphs Within Knowledge Graphs for Knowledge-Based Causal Discovery
A learning-to-rank model chooses informative knowledge-graph paths between entity pairs, and adding the top path to zero-shot prompts improves LLM causal classification by up to 44.4 F1 points.
-
Causal Distillation: Transferring Structured Explanations from Large to Compact Language Models
Small language models fine-tuned on GPT-4 causal explanations score high on a new teacher-similarity metric, but the paper provides no independent evidence that causal reasoning was transferred.
Discussion (0). Sign in to comment.