Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Zheng, Xiao
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2509.13702
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918142578524160
author Zheng, Xiao
author_facet Zheng, Xiao
contents Large Language Model (LLM) hallucination is a significant barrier to their reliable deployment. Current methods like Retrieval-Augmented Generation (RAG) are often reactive. We introduce **Dynamic Self-reinforcing Calibration for Hallucination Suppression (DSCC-HS)**, a novel, proactive framework that intervenes during autoregressive decoding. Inspired by dual-process cognitive theory, DSCC-HS uses a compact proxy model, trained in adversarial roles as a Factual Alignment Proxy (FAP) and a Hallucination Detection Proxy (HDP). During inference, these proxies dynamically steer a large target model by injecting a real-time steering vector, which is the difference between FAP and HDP logits, at each decoding step. This plug-and-play approach requires no modification to the target model. Our experiments on TruthfulQA and BioGEN show DSCC-HS achieves state-of-the-art performance. On TruthfulQA, it reached a 99.2% Factual Consistency Rate (FCR). On the long-form BioGEN benchmark, it attained the highest FActScore of 46.50. These results validate DSCC-HS as a principled and efficient solution for enhancing LLM factuality.
format Preprint
id arxiv_https___arxiv_org_abs_2509_13702
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DSCC-HS: A Dynamic Self-Reinforcing Framework for Hallucination Suppression in Large Language Models
Zheng, Xiao
Computation and Language
Artificial Intelligence
Large Language Model (LLM) hallucination is a significant barrier to their reliable deployment. Current methods like Retrieval-Augmented Generation (RAG) are often reactive. We introduce **Dynamic Self-reinforcing Calibration for Hallucination Suppression (DSCC-HS)**, a novel, proactive framework that intervenes during autoregressive decoding. Inspired by dual-process cognitive theory, DSCC-HS uses a compact proxy model, trained in adversarial roles as a Factual Alignment Proxy (FAP) and a Hallucination Detection Proxy (HDP). During inference, these proxies dynamically steer a large target model by injecting a real-time steering vector, which is the difference between FAP and HDP logits, at each decoding step. This plug-and-play approach requires no modification to the target model. Our experiments on TruthfulQA and BioGEN show DSCC-HS achieves state-of-the-art performance. On TruthfulQA, it reached a 99.2% Factual Consistency Rate (FCR). On the long-form BioGEN benchmark, it attained the highest FActScore of 46.50. These results validate DSCC-HS as a principled and efficient solution for enhancing LLM factuality.
title DSCC-HS: A Dynamic Self-Reinforcing Framework for Hallucination Suppression in Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2509.13702