paper-with-me

Papers

A Cause-Effect Look at Alleviating Hallucination of Knowledge-grounded Dialogue Generation

2024-04-04 · Jifan Yu, Xiaohan Zhang, Yifan Xu, Xuanyu Lei, Zijun Yao, Jing Zhang, Lei Hou, Juanzi Li

Empowered by the large-scale pretrained language models, existing dialogue systems have demonstrated impressive performance conducting fluent and natural-sounding conversations. However, they are still plagued by the hallucination problem, causing unpredictable factual errors in the generated responses. Recently, knowledge-grounded dialogue generation models, that intentionally invoke external knowledge resources to more informative responses, are also proven to be effective in reducing hallucination. Following the idea of getting high-quality knowledge, a few efforts have achieved pretty good performance on this issue. As some inevitable knowledge noises may also lead to hallucinations, it is emergent to investigate the reason and future directions for building noise-tolerant methods in KGD tasks. In this paper, we analyze the causal story behind this problem with counterfactual reasoning methods. Based on the causal effect analysis, we propose a possible solution for alleviating the hallucination in KGD by exploiting the dialogue-knowledge interaction. Experimental results of our example implementation show that this method can reduce hallucination without disrupting other dialogue performance, while keeping adaptive to different generation models. We hope our efforts can support and call for more attention to developing lightweight techniques towards robust and trusty dialogue systems.

📄 PDF Abstract BibTeX arXiv:2404.03491

Code (0)

등록된 구현이 없습니다.

Tasks

counterfactualCounterfactual ReasoningDialogue GenerationHallucination

Similar Papers 제목 키워드 기반

Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factuality

2025-09-28 · Junliang Li, Yucheng Wang, Yan Chen, Yu Ran 외 arxiv

Hallucination in large language models (LLMs) during long-form generation remains difficult to address under existing reinforcement learning from human feedback (RLHF) frameworks, as their preference rewards often overlo…

Reinforcement Learning

Look Closer! An Adversarial Parametric Editing Framework for Hallucination Mitigation in VLMs

2025-12-26 · Jiayu Hu, Beibei Li, Jiangwei Xia, Yanjun Qin 외 arxiv

While Vision-Language Models (VLMs) have garnered increasing attention in the AI community due to their promising practical applications, they exhibit persistent hallucination issues, generating outputs misaligned with v…

Hallucination Diversity-Aware Active Learning for Text Summarization

2024-04-02 · Yu Xia, Xu Liu, Tong Yu, Sungchul Kim 외

Large Language Models (LLMs) have shown propensity to generate hallucinated outputs, i.e., texts that are factually incorrect or unsupported. Existing methods for alleviating hallucinations typically require costly human…

Active LearningDiversityHallucinationText Summarization

HACK: Hallucinations Along Certainty and Knowledge Axes

2025-10-28 · Adi Simhi, Jonathan Herzig, Itay Itzhak, Dana Arad 외 arxiv

Hallucinations in LLMs present a critical barrier to their reliable usage. Existing research usually categorizes hallucination by their external properties rather than by the LLMs' underlying internal properties. This ex…

Octopus: Alleviating Hallucination via Dynamic Contrastive Decoding

2025-01-01 · CVPR 2025 1 · Wei Suo, Lijun Zhang, Mengyang Sun, Lin Yuanbo Wu 외

Large Vision-Language Models (LVLMs) have obtained impressive performance in visual content understanding and multi-modal reasoning. Unfortunately, these large models suffer from serious hallucination problems and te…

Hallucination