paper-with-me

홈 › Papers

$\textit{Dial BeInfo for Faithfulness}$: Improving Factuality of Information-Seeking Dialogue via Behavioural Fine-Tuning

2023-11-16 · Evgeniia Razumovskaia, Ivan Vulić, Pavle Marković, Tomasz Cichy, Qian Zheng, Tsung-Hsien Wen, Paweł Budzianowski

Factuality is a crucial requirement in information seeking dialogue: the system should respond to the user's queries so that the responses are meaningful and aligned with the knowledge provided to the system. However, most modern large language models suffer from hallucinations, that is, they generate responses not supported by or contradicting the knowledge source. To mitigate the issue and increase faithfulness of information-seeking dialogue systems, we introduce BeInfo, a simple yet effective method that applies behavioural tuning to aid information-seeking dialogue. Relying on three standard datasets, we show that models tuned with BeInfo} become considerably more faithful to the knowledge source both for datasets and domains seen during BeInfo-tuning, as well as on unseen domains, when applied in a zero-shot manner. In addition, we show that the models with 3B parameters (e.g., Flan-T5) tuned with BeInfo demonstrate strong performance on data from real `production' conversations and outperform GPT4 when tuned on a limited amount of such realistic in-domain dialogues.

📄 PDF Abstract BibTeX arXiv:2311.09800

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

HalluDial: A Large-Scale Benchmark for Automatic Dialogue-Level Hallucination Evaluation

2024-06-11 · Wen Luo, Tianshu Shen, Wei Li, Guangyue Peng 외

Large Language Models (LLMs) have significantly advanced the field of Natural Language Processing (NLP), achieving remarkable performance across diverse tasks and enabling widespread real-world applications. However, LLM…

HallucinationHallucination EvaluationLanguage ModellingSentence

DiaHalu: A Dialogue-level Hallucination Evaluation Benchmark for Large Language Models

2024-03-01 · Kedi Chen, Qin Chen, Jie zhou, Yishen He 외

Since large language models (LLMs) achieve significant success in recent years, the hallucination issue remains a challenge, numerous benchmarks are proposed to detect the hallucination. Nevertheless, some of these bench…

HallucinationHallucination EvaluationSentence

Is Factuality Enhancement a Free Lunch For LLMs? Better Factuality Can Lead to Worse Context-Faithfulness

2024-03-30 · Baolong Bi, Shenghua Liu, Yiwei Wang, Lingrui Mei 외

As the modern tools of choice for text understanding and generation, large language models (LLMs) are expected to accurately output answers by leveraging the input context. This requires LLMs to possess both context-fait…

knowledge editing

TODSum: Task-Oriented Dialogue Summarization with State Tracking

2021-10-25 · Lulu Zhao, Fujia Zheng, Keqing He, Weihao Zeng 외

Previous dialogue summarization datasets mainly focus on open-domain chitchat dialogues, while summarization datasets for the broadly used task-oriented dialogue haven't been explored yet. Automatically summarizing such …

INFACT: A Diagnostic Benchmark for Induced Faithfulness and Factuality Hallucinations in Video-LLMs

2026-03-12 · Junqi Yang, Yuecong Min, Jie Zhang, Shiguang Shan 외 arxiv

Despite rapid progress, Video Large Language Models (Video-LLMs) remain unreliable due to hallucinations, which are outputs that contradict either video evidence (faithfulness) or verifiable world knowledge (factuality).…