Massive: A 1m-example multilingual natural language understanding dataset with 51 typologically-diverse languages
J FitzGerald, C Hench, C Peris, S Mackie… - arXiv preprint arXiv …, 2022 - arxiv.org
We present the MASSIVE dataset--Multilingual Amazon Slu resource package (SLURP) for
Slot-filling, Intent classification, and Virtual assistant Evaluation. MASSIVE contains 1M …
Slot-filling, Intent classification, and Virtual assistant Evaluation. MASSIVE contains 1M …
Megaverse: Benchmarking large language models across languages, modalities, models and tasks
S Ahuja, D Aggarwal, V Gumma, I Watts… - arXiv preprint arXiv …, 2023 - arxiv.org
Recently, there has been a rapid advancement in research on Large Language Models
(LLMs), resulting in significant progress in several Natural Language Processing (NLP) …
(LLMs), resulting in significant progress in several Natural Language Processing (NLP) …
Baichuan 2: Open large-scale language models
A Yang, B Xiao, B Wang, B Zhang, C Bian… - arXiv preprint arXiv …, 2023 - arxiv.org
Large language models (LLMs) have demonstrated remarkable performance on a variety of
natural language tasks based on just a few examples of natural language instructions …
natural language tasks based on just a few examples of natural language instructions …
Alexa teacher model: Pretraining and distilling multi-billion-parameter encoders for natural language understanding systems
J FitzGerald, S Ananthakrishnan, K Arkoudas… - Proceedings of the 28th …, 2022 - dl.acm.org
We present results from a large-scale experiment on pretraining encoders with non-
embedding parameter counts ranging from 700M to 9.3 B, their subsequent distillation into …
embedding parameter counts ranging from 700M to 9.3 B, their subsequent distillation into …
Culturax: A cleaned, enormous, and multilingual dataset for large language models in 167 languages
T Nguyen, C Van Nguyen, VD Lai, H Man… - arXiv preprint arXiv …, 2023 - arxiv.org
The driving factors behind the development of large language models (LLMs) with
impressive learning capabilities are their colossal model sizes and extensive training …
impressive learning capabilities are their colossal model sizes and extensive training …
Alignbench: Benchmarking chinese alignment of large language models
X Liu, X Lei, S Wang, Y Huang, Z Feng, B Wen… - arXiv preprint arXiv …, 2023 - arxiv.org
Alignment has become a critical step for instruction-tuned Large Language Models (LLMs)
to become helpful assistants. However, effective evaluation of alignment for emerging …
to become helpful assistants. However, effective evaluation of alignment for emerging …
LINGUIST: Language model instruction tuning to generate annotated utterances for intent classification and slot tagging
A Rosenbaum, S Soltan, W Hamza, Y Versley… - arXiv preprint arXiv …, 2022 - arxiv.org
We present LINGUIST, a method for generating annotated data for Intent Classification and
Slot Tagging (IC+ ST), via fine-tuning AlexaTM 5B, a 5-billion-parameter multilingual …
Slot Tagging (IC+ ST), via fine-tuning AlexaTM 5B, a 5-billion-parameter multilingual …
Flm-101b: An open llm and how to train it with $100 k budget
X Li, Y Yao, X Jiang, X Fang, X Meng, S Fan… - arXiv preprint arXiv …, 2023 - arxiv.org
Large language models (LLMs) have achieved remarkable success in NLP and multimodal
tasks. Despite these successes, their development faces two main challenges:(i) high …
tasks. Despite these successes, their development faces two main challenges:(i) high …
Api-bank: A comprehensive benchmark for tool-augmented llms
M Li, Y Zhao, B Yu, F Song, H Li, H Yu, Z Li… - arXiv preprint arXiv …, 2023 - arxiv.org
Recent research has demonstrated that Large Language Models (LLMs) can enhance their
capabilities by utilizing external tools. However, three pivotal questions remain …
capabilities by utilizing external tools. However, three pivotal questions remain …
XNLI: Evaluating cross-lingual sentence representations
A Conneau, G Lample, R Rinott, A Williams… - arXiv preprint arXiv …, 2018 - arxiv.org
State-of-the-art natural language processing systems rely on supervision in the form of
annotated data to learn competent models. These models are generally trained on data in a …
annotated data to learn competent models. These models are generally trained on data in a …