Do Large Language Models Know What They Don't Know?

Z Yin, Q Sun, Q Guo, J Wu, X Qiu, X Huang - arxiv preprint arxiv …, 2023 - arxiv.org
Large language models (LLMs) have a wealth of knowledge that allows them to excel in
various Natural Language Processing (NLP) tasks. Current research focuses on enhancing …

Direct preference optimization with an offset

A Amini, T Vieira, R Cotterell - arxiv preprint arxiv:2402.10571, 2024 - arxiv.org
Direct preference optimization (DPO) is a successful fine-tuning strategy for aligning large
language models with human preferences without the need to train a reward model or …

Leveraging gpt-4 for automatic translation post-editing

V Raunak, A Sharaf, Y Wang, HH Awadallah… - arxiv preprint arxiv …, 2023 - arxiv.org
While Neural Machine Translation (NMT) represents the leading approach to Machine
Translation (MT), the outputs of NMT models still require translation post-editing to rectify …

Easyedit: An easy-to-use knowledge editing framework for large language models

P Wang, N Zhang, B Tian, Z **, Y Yao, Z Xu… - arxiv preprint arxiv …, 2023 - arxiv.org
Large Language Models (LLMs) usually suffer from knowledge cutoff or fallacy issues, which
means they are unaware of unseen events or generate text with incorrect facts owing to the …

Adapting large language models for document-level machine translation

M Wu, TT Vu, L Qu, G Foster, G Haffari - arxiv preprint arxiv:2401.06468, 2024 - arxiv.org
Large language models (LLMs) have made significant strides in various natural language
processing (NLP) tasks. Recent research shows that the moderately-sized LLMs often …

Merging generated and retrieved knowledge for open-domain QA

Y Zhang, M Khalifa, L Logeswaran, M Lee… - arxiv preprint arxiv …, 2023 - arxiv.org
Open-domain question answering (QA) systems are often built with retrieval modules.
However, retrieving passages from a given source is known to suffer from insufficient …

Knowledge-augmented language model verification

J Baek, S Jeong, M Kang, JC Park… - arxiv preprint arxiv …, 2023 - arxiv.org
Recent Language Models (LMs) have shown impressive capabilities in generating texts with
the knowledge internalized in parameters. Yet, LMs often generate the factually incorrect …

Chatreport: Democratizing sustainability disclosure analysis through llm-based tools

J Ni, J Bingler, C Colesanti-Senni, M Kraus… - arxiv preprint arxiv …, 2023 - arxiv.org
In the face of climate change, are companies really taking substantial steps toward more
sustainable operations? A comprehensive answer lies in the dense, information-rich …

Battle of the Large Language Models: Dolly vs LLaMA vs Vicuna vs Guanaco vs Bard vs ChatGPT--A Text-to-SQL Parsing Comparison

S Sun, Y Zhang, J Yan, Y Gao, D Ong, B Chen… - arxiv preprint arxiv …, 2023 - arxiv.org
The success of ChatGPT has ignited an AI race, with researchers striving to develop new
large language models (LLMs) that can match or surpass the language understanding and …

Aart: Ai-assisted red-teaming with diverse data generation for new llm-powered applications

B Radharapu, K Robinson, L Aroyo, P Lahoti - arxiv preprint arxiv …, 2023 - arxiv.org
Adversarial testing of large language models (LLMs) is crucial for their safe and responsible
deployment. We introduce a novel approach for automated generation of adversarial …