Enregistré dans:
Détails bibliographiques
Auteurs principaux: Liu, Tianyang, Li, Tianyi, Cheng, Liang, Steedman, Mark
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2408.14467
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912002312503296
author Liu, Tianyang
Li, Tianyi
Cheng, Liang
Steedman, Mark
author_facet Liu, Tianyang
Li, Tianyi
Cheng, Liang
Steedman, Mark
contents Large Language Models (LLMs) are reported to hold undesirable attestation bias on inference tasks: when asked to predict if a premise P entails a hypothesis H, instead of considering H's conditional truthfulness entailed by P, LLMs tend to use the out-of-context truth label of H as a fragile proxy. In this paper, we propose a pipeline that exploits this bias to do explicit inductive inference. Our pipeline uses an LLM to transform a premise into a set of attested alternatives, and then aggregate answers of the derived new entailment inquiries to support the original inference prediction. On a directional predicate entailment benchmark, we demonstrate that by applying this simple pipeline, we can improve the overall performance of LLMs on inference and substantially alleviate the impact of their attestation bias.
format Preprint
id arxiv_https___arxiv_org_abs_2408_14467
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Explicit Inductive Inference using Large Language Models
Liu, Tianyang
Li, Tianyi
Cheng, Liang
Steedman, Mark
Computation and Language
Large Language Models (LLMs) are reported to hold undesirable attestation bias on inference tasks: when asked to predict if a premise P entails a hypothesis H, instead of considering H's conditional truthfulness entailed by P, LLMs tend to use the out-of-context truth label of H as a fragile proxy. In this paper, we propose a pipeline that exploits this bias to do explicit inductive inference. Our pipeline uses an LLM to transform a premise into a set of attested alternatives, and then aggregate answers of the derived new entailment inquiries to support the original inference prediction. On a directional predicate entailment benchmark, we demonstrate that by applying this simple pipeline, we can improve the overall performance of LLMs on inference and substantially alleviate the impact of their attestation bias.
title Explicit Inductive Inference using Large Language Models
topic Computation and Language
url https://arxiv.org/abs/2408.14467