홈 › Papers

Fine-grained Hallucination Detection and Editing for Language Models

2024-01-12 · Abhika Mishra, Akari Asai, Vidhisha Balachandran, Yizhong Wang, Graham Neubig, Yulia Tsvetkov, Hannaneh Hajishirzi

Large language models (LMs) are prone to generate factual errors, which are often called hallucinations. In this paper, we introduce a comprehensive taxonomy of hallucinations and argue that hallucinations manifest in diverse forms, each requiring varying degrees of careful assessments to verify factuality. We propose a novel task of automatic fine-grained hallucination detection and construct a new evaluation benchmark, FavaBench, that includes about one thousand fine-grained human judgments on three LM outputs across various domains. Our analysis reveals that ChatGPT and Llama2-Chat (70B, 7B) exhibit diverse types of hallucinations in the majority of their outputs in information-seeking scenarios. We train FAVA, a retrieval-augmented LM by carefully creating synthetic data to detect and correct fine-grained hallucinations. On our benchmark, our automatic and human evaluations show that FAVA significantly outperforms ChatGPT and GPT-4 on fine-grained hallucination detection, and edits suggested by FAVA improve the factuality of LM-generated text.

📄 PDF Abstract BibTeX arXiv:2401.06855

Code (0)

등록된 구현이 없습니다.

Tasks

HallucinationRetrieval

Methods 이 논문이 사용한 방법론

Attention 설명 없음

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…

Adam 설명 없음

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Multi-Head Attention 설명 없음

PFME: A Modular Approach for Fine-grained Hallucination Detection and Editing of Large Language Models

2024-06-29 · Kunquan Deng, Zeyu Huang, Chen Li, Chenghua Lin 외

Large Language Models (LLMs) excel in fluency but risk producing inaccurate content, called "hallucinations." This paper outlines a standardized process for categorizing fine-grained hallucination types and proposes an i…

Hallucination Sentence

Multilingual Fine-Grained News Headline Hallucination Detection

2024-07-22 · Jiaming Shen, Tianqi Liu, Jialu Liu, Zhen Qin 외

The popularity of automated news headline generation has surged with advancements in pre-trained language models. However, these models often suffer from the ``hallucination'' problem, where the generated headline is not…

Hallucination Headline Generation In-Context Learning

FG-PRM: Fine-grained Hallucination Detection and Mitigation in Language Model Mathematical Reasoning

2024-10-08 · Ruosen Li, Ziming Luo, Xinya Du

Hallucinations in large language models (LLMs) pose significant challenges in tasks requiring complex multi-step reasoning, such as mathematical problem-solving. Existing approaches primarily detect the presence of hallu…

GSM8K Hallucination Language Modeling Language Modelling+3

Halluverse-M^3: A multitask multilingual benchmark for hallucination in LLMs

2026-02-06 · Samir Abdaljalil, Parichit Sharma, Erchin Serpedin, Hasan Kurban arxiv

Hallucinations in large language models remain a persistent challenge, particularly in multilingual and generative settings where factual consistency is difficult to maintain. While recent models show strong performance …

Question Answering

HalluScope: Fine-grained Hallucination Diagnosis for Multimodal Large Language Models

2026-07-23 · Weilin Jin, Mingyu Wang, Wenbo Li, Haoyang Huang 외 arxiv

Although Multimodal Large Language Models have achieved strong performance across a wide range of vision-language tasks, they still suffer from hallucinations, where model outputs become inconsistent with the visual cont…

Explanation Generation