IRCLJan 21, 2024

Towards Reliable and Factual Response Generation: Detecting Unanswerable Questions in Information-Seeking Conversations

arXiv:2401.11452v16 citationsECIR
Originality Incremental advance
AI Analysis

This addresses the issue of unreliable responses in conversational AI systems, which can undermine user trust, though it is incremental as it builds on existing retrieval-based approaches.

The paper tackles the problem of hallucinations in generative AI by proposing a two-step method to detect unanswerable questions in information-seeking conversations, using a sentence-level classifier and aggregation to estimate answerability, and demonstrates that it outperforms a state-of-the-art LLM on this task.

Generative AI models face the challenge of hallucinations that can undermine users' trust in such systems. We approach the problem of conversational information seeking as a two-step process, where relevant passages in a corpus are identified first and then summarized into a final system response. This way we can automatically assess if the answer to the user's question is present in the corpus. Specifically, our proposed method employs a sentence-level classifier to detect if the answer is present, then aggregates these predictions on the passage level, and eventually across the top-ranked passages to arrive at a final answerability estimate. For training and evaluation, we develop a dataset based on the TREC CAsT benchmark that includes answerability labels on the sentence, passage, and ranking levels. We demonstrate that our proposed method represents a strong baseline and outperforms a state-of-the-art LLM on the answerability prediction task.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes