1 code implementation • 19 Feb 2024 • Xiao Ye, Andrew Wang, Jacob Choi, Yining Lu, Shreya Sharma, Lingfeng Shen, Vijay Tiyyala, Nicholas Andrews, Daniel Khashabi
Our benchmarking approach focuses on aspects of this ability that are common among humans: (i) recalling related experiences from a large amount of information, and (ii) applying analogical reasoning to complex and lengthy scenarios.
no code implementations • 23 Jan 2024 • Lingfeng Shen, Weiting Tan, Sihao Chen, Yunmo Chen, Jingyu Zhang, Haoran Xu, Boyuan Zheng, Philipp Koehn, Daniel Khashabi
As the influence of large language models (LLMs) spans across global communities, their safety challenges in multilingual settings become paramount for alignment research.
1 code implementation • 16 Jan 2024 • Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, Young Jin Kim
However, even the top-performing 13B LLM-based translation models, like ALMA, does not match the performance of state-of-the-art conventional encoder-decoder translation models or larger-scale LLMs such as GPT-4.
no code implementations • 4 Nov 2023 • Weiting Tan, Haoran Xu, Lingfeng Shen, Shuyue Stella Li, Kenton Murray, Philipp Koehn, Benjamin Van Durme, Yunmo Chen
Large language models trained primarily in a monolingual setting have demonstrated their ability to generalize to machine translation using zero- and few-shot examples with in-context learning.
no code implementations • 12 Oct 2023 • Lingfeng Shen, Aayush Mishra, Daniel Khashabi
We observe that ICL and GD have different sensitivity to the order in which they observe demonstrations.
1 code implementation • 6 Oct 2023 • Abe Bohan Hou, Jingyu Zhang, Tianxing He, Yichen Wang, Yung-Sung Chuang, Hongwei Wang, Lingfeng Shen, Benjamin Van Durme, Daniel Khashabi, Yulia Tsvetkov
Existing watermarking algorithms are vulnerable to paraphrase attacks because of their token-level design.
1 code implementation • 28 Sep 2023 • Lingfeng Shen, Sihao Chen, Linfeng Song, Lifeng Jin, Baolin Peng, Haitao Mi, Daniel Khashabi, Dong Yu
We propose Contrast Instructions -- a benchmarking strategy for the consistency of RM.
no code implementations • 4 Jun 2023 • Lingfeng Shen, Haiyun Jiang, Lemao Liu, Shuming Shi
Sentence embedding is one of the most fundamental tasks in Natural Language Processing and plays an important role in various tasks.
1 code implementation • 18 May 2023 • Lingfeng Shen, Weiting Tan, Boyuan Zheng, Daniel Khashabi
We provide theoretical foundations for this metric and its relationship with other prompt selection metrics, providing a comprehensive understanding of existing methods.
no code implementations • 13 May 2023 • Lingfeng Shen, Haiyun Jiang, Lemao Liu, Shuming Shi
Generating proper embedding of sentences through an unsupervised way is beneficial to semantic matching and retrieval problems in real-world scenarios.
no code implementations • 13 May 2023 • Lingfeng Shen, Haiyun Jiang, Lemao Liu, Ying Chen
Static word embedding is still useful, particularly for context-unavailable tasks, because in the case of no context available, pre-trained language models often perform worse than static word embeddings.
no code implementations • 3 Feb 2023 • Lingfeng Shen, Ze Zhang, Haiyun Jiang, Ying Chen
A recent line of work, detection-based defense, aims to distinguish adversarial sentences from benign ones.
1 code implementation • 17 Feb 2022 • Lingfeng Shen, Lemao Liu, Haiyun Jiang, Shuming Shi
In this paper we revisit automatic metrics for paraphrase evaluation and obtain two findings that disobey conventional wisdom: (1) Reference-free metrics achieve better performance than their reference-based counterparts.
no code implementations • 9 Jan 2022 • Lingfeng Shen, Haiyun Jiang, Lemao Liu, Shuming Shi
It has been shown that natural language processing (NLP) models are vulnerable to a kind of security threat called the Backdoor Attack, which utilizes a `backdoor trigger' paradigm to mislead the models.