CHARP: Conversation History AwaReness Probing for Knowledge-grounded Dialogue Systems
In this work, we dive deep into one of the popular knowledge-grounded dialogue benchmarks that focus on faithfulness, FaithDial. We show that a significant portion of the FaithDial data contains annotation artifacts, which may bias models towards completely ignoring the conversation history. We therefore introduce CHARP, a diagnostic test set, designed for an improved evaluation of hallucinations in conversational model. CHARP not only measures hallucination but also the compliance of the models to the conversation task. Our extensive analysis reveals that models primarily exhibit poor performance on CHARP due to their inability to effectively attend to and reason over the conversation history. Furthermore, the evaluation methods of FaithDial fail to capture these shortcomings, neglecting the conversational history. Our findings indicate that there is substantial room for contribution in both dataset creation and hallucination evaluation for knowledge-grounded dialogue, and that CHARP can serve as a tool for monitoring the progress in this particular research area. CHARP is publicly available at https://huggingface.co/datasets/huawei-noah/CHARP
Code (0)
등록된 구현이 없습니다.
Tasks
DiagnosticHallucinationHallucination EvaluationMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
CharPoet: A Chinese Classical Poetry Generation System Based on Token-free LLM
Automatic Chinese classical poetry generation has attracted much research interest, but achieving effective control over format and content simultaneously remains challenging. Traditional systems usually accept keywords …
Language ModellingEnhancing LLM-Based Human-Robot Interaction with Nuances for Diversity Awareness
This paper presents a system for diversity-aware autonomous conversation leveraging the capabilities of large language models (LLMs). The system adapts to diverse populations and individuals, considering factors like bac…
DiversityMachine Learning-based Correlation of Charpy Impact Properties Between Sub-sized and Standard-sized Specimens for Nuclear Structural Materials
Reliable correlations of Charpy impact test results between sub-sized and full-sized specimens are essential for structural integrity assessments, particularly in nuclear applications, where spatial constraints and limit…
Human-like informative conversations: Better acknowledgements using conditional mutual information
This work aims to build a dialogue agent that can weave new factual content into conversations as naturally as humans. We draw insights from linguistic principles of conversational analysis and annotate human-human conve…
SpecificityLearning to Trust Your Feelings: Leveraging Self-awareness in LLMs for Hallucination Mitigation
We evaluate the ability of Large Language Models (LLMs) to discern and express their internal knowledge state, a key factor in countering factual hallucination and ensuring reliable application of LLMs. We observe a robu…
HallucinationKnowledge Probingreinforcement-learningReinforcement Learning