CLEAR-3K: Assessing Causal Explanatory Capabilities in Language Models

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Liu, Naiming, Baraniuk, Richard, Sonkar, Shashank
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915353278283776
author Liu, Naiming
Baraniuk, Richard
Sonkar, Shashank
author_facet Liu, Naiming
Baraniuk, Richard
Sonkar, Shashank
contents We introduce CLEAR-3K, a dataset of 3,000 assertion-reasoning questions designed to evaluate whether language models can determine if one statement causally explains another. Each question present an assertion-reason pair and challenge language models to distinguish between semantic relatedness and genuine causal explanatory relationships. Through comprehensive evaluation of 21 state-of-the-art language models (ranging from 0.5B to 72B parameters), we identify two fundamental findings. First, language models frequently confuse semantic similarity with causality, relying on lexical and semantic overlap instead of inferring actual causal explanatory relationships. Second, as parameter size increases, models tend to shift from being overly skeptical about causal relationships to being excessively permissive in accepting them. Despite this shift, performance measured by the Matthews Correlation Coefficient plateaus at just 0.55, even for the best-performing models.Hence, CLEAR-3K provides a crucial benchmark for developing and evaluating genuine causal reasoning in language models, which is an essential capability for applications that require accurate assessment of causal relationships.
format Preprint
id arxiv_https___arxiv_org_abs_2506_17180
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CLEAR-3K: Assessing Causal Explanatory Capabilities in Language Models
Liu, Naiming
Baraniuk, Richard
Sonkar, Shashank
Computation and Language
We introduce CLEAR-3K, a dataset of 3,000 assertion-reasoning questions designed to evaluate whether language models can determine if one statement causally explains another. Each question present an assertion-reason pair and challenge language models to distinguish between semantic relatedness and genuine causal explanatory relationships. Through comprehensive evaluation of 21 state-of-the-art language models (ranging from 0.5B to 72B parameters), we identify two fundamental findings. First, language models frequently confuse semantic similarity with causality, relying on lexical and semantic overlap instead of inferring actual causal explanatory relationships. Second, as parameter size increases, models tend to shift from being overly skeptical about causal relationships to being excessively permissive in accepting them. Despite this shift, performance measured by the Matthews Correlation Coefficient plateaus at just 0.55, even for the best-performing models.Hence, CLEAR-3K provides a crucial benchmark for developing and evaluating genuine causal reasoning in language models, which is an essential capability for applications that require accurate assessment of causal relationships.
title CLEAR-3K: Assessing Causal Explanatory Capabilities in Language Models
topic Computation and Language
url https://arxiv.org/abs/2506.17180