Can LLMs Reconcile Knowledge Conflicts in Counterfactual Reasoning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915771993554944 |
|---|---|
| author | Yamin, Khurram Ghosal, Gaurav Wilder, Bryan |
| author_facet | Yamin, Khurram Ghosal, Gaurav Wilder, Bryan |
| contents | Large Language Models have been shown to contain extensive world knowledge in their parameters, enabling impressive performance on many knowledge intensive tasks. However, when deployed in novel settings, LLMs often encounter situations where they must integrate parametric knowledge with new or unfamiliar information. In this work, we explore whether LLMs can combine knowledge in-context with their parametric knowledge through the lens of counterfactual reasoning. Through synthetic and real experiments in multi-hop reasoning problems, we show that LLMs generally struggle with counterfactual reasoning, often resorting to exclusively using their parametric knowledge. Moreover, we show that simple post-hoc finetuning can struggle to instill counterfactual reasoning ability -- often leading to degradation in stored parametric knowledge. Ultimately, our work reveals important limitations of current LLM's abilities to re-purpose parametric knowledge in novel settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_15732 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Can LLMs Reconcile Knowledge Conflicts in Counterfactual Reasoning Yamin, Khurram Ghosal, Gaurav Wilder, Bryan Artificial Intelligence Machine Learning Large Language Models have been shown to contain extensive world knowledge in their parameters, enabling impressive performance on many knowledge intensive tasks. However, when deployed in novel settings, LLMs often encounter situations where they must integrate parametric knowledge with new or unfamiliar information. In this work, we explore whether LLMs can combine knowledge in-context with their parametric knowledge through the lens of counterfactual reasoning. Through synthetic and real experiments in multi-hop reasoning problems, we show that LLMs generally struggle with counterfactual reasoning, often resorting to exclusively using their parametric knowledge. Moreover, we show that simple post-hoc finetuning can struggle to instill counterfactual reasoning ability -- often leading to degradation in stored parametric knowledge. Ultimately, our work reveals important limitations of current LLM's abilities to re-purpose parametric knowledge in novel settings. |
| title | Can LLMs Reconcile Knowledge Conflicts in Counterfactual Reasoning |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2506.15732 |