In spite of the potential for ground-breaking achievements offered by large language models (LLMs)(eg, GPT-3), they still lag significantly behind fully-supervised baselines (eg, fine …
The strong few-shot in-context learning capability of large pre-trained language models (PLMs) such as GPT-3 is highly appealing for application domains such as biomedicine …
Prompt tuning, in which a base pretrained model is adapted to each task via conditioning on learned prompt vectors, has emerged as a promising approach for efficiently adapting large …
B Liu, J Ash, S Goel… - Advances in Neural …, 2024 - proceedings.neurips.cc
Why do large language models sometimes output factual inaccuracies and exhibit erroneous reasoning? The brittleness of these models, particularly when executing long …
X Wang, C Li, Z Wang, F Bai, H Luo, J Zhang… - arXiv preprint arXiv …, 2023 - arxiv.org
Highly effective, task-specific prompts are often heavily engineered by experts to integrate detailed instructions and domain insights based on a deep understanding of both instincts of …
Autoregressive large language models (LLMs) compress knowledge from their training data through next-token conditional distributions. This limits tractable querying of this knowledge …
X Liu, Z Zhang, Y Wang, H Pu, Y Lan… - arXiv preprint arXiv …, 2022 - arxiv.org
Machine-Generated Text (MGT) detection, a task that discriminates MGT from Human- Written Text (HWT), plays a crucial role in preventing misuse of text generative models …
Classifier-Free Guidance (CFG) has recently emerged in text-to-image generation as a lightweight technique to encourage prompt-adherence in generations. In this work, we …
X Liu, Z Zhang, Y Wang, H Pu, Y Lan… - Proceedings of the 2023 …, 2023 - aclanthology.org
Abstract Machine-Generated Text (MGT) detection, a task that discriminates MGT from Human-Written Text (HWT), plays a crucial role in preventing misuse of text generative …