Hallucinations in large multilingual translation models

NM Guerreiro, DM Alves, J Waldendorf… - Transactions of the …, 2023 - direct.mit.edu
Hallucinated translations can severely undermine and raise safety issues when machine
translation systems are deployed in the wild. Previous research on the topic focused on …

A primer on the inner workings of transformer-based language models

J Ferrando, G Sarti, A Bisazza, M Costa-jussà - 2024 - research.rug.nl
The rapid progress of research aimed at interpreting the inner workings of advanced
language models has highlighted a need for contextualizing the insights gained from years …

HalOmi: A manually annotated benchmark for multilingual hallucination and omission detection in machine translation

D Dale, E Voita, J Lam, P Hansanti, C Ropers… - arxiv preprint arxiv …, 2023 - arxiv.org
Hallucinations in machine translation are translations that contain information completely
unrelated to the input. Omissions are translations that do not include some of the input …

Literature Review of AI Hallucination Research Since the Advent of ChatGPT: Focusing on Papers from arxiv

DM Park, HJ Lee - Informatization Policy, 2024 - koreascience.kr
Hallucination is a significant barrier to the utilization of large-scale language models or
multimodal models. In this study, we collected 654 computer science papers with" …

Elastic weight removal for faithful and abstractive dialogue generation

N Daheim, N Dziri, M Sachan, I Gurevych… - arxiv preprint arxiv …, 2023 - arxiv.org
Ideally, dialogue systems should generate responses that are faithful to the knowledge
contained in relevant documents. However, many models generate hallucinated responses …

An Audit on the Perspectives and Challenges of Hallucinations in NLP

PN Venkit, T Chakravorti, V Gupta, H Biggs… - arxiv preprint arxiv …, 2024 - arxiv.org
We audit how hallucination in large language models (LLMs) is characterized in peer-
reviewed literature, using a critical examination of 103 publications across NLP research …

Investigating hallucinations in pruned large language models for abstractive summarization

G Chrysostomou, Z Zhao, M Williams… - Transactions of the …, 2024 - direct.mit.edu
Despite the remarkable performance of generative large language models (LLMs) on
abstractive summarization, they face two significant challenges: their considerable size and …

Non-fluent synthetic target-language data improve neural machine translation

VM Sánchez-Cartagena, M Esplà-Gomis… - … on Pattern Analysis …, 2023 - ieeexplore.ieee.org
When the amount of parallel sentences available to train a neural machine translation is
scarce, a common practice is to generate new synthetic training samples from them. A …

GraphER: A Structure-aware Text-to-Graph Model for Entity and Relation Extraction

U Zaratiana, N Tomeh, NE Khbir, P Holat… - arxiv preprint arxiv …, 2024 - arxiv.org
Information extraction (IE) is an important task in Natural Language Processing (NLP),
involving the extraction of named entities and their relationships from unstructured text. In …

Trucidator: Document-level Event Factuality Identification via Hallucination Enhancement and Cross-Document Inference

Z Zhang, Z Qian, X Zhu, P Li, Q Zhu - Proceedings of the 31st …, 2025 - aclanthology.org
Document-level event factuality identification (DEFI) assesses the veracity degree to which
an event mentioned in a document has happened, which is crucial for many natural …