3 citations · 8 across the 11 of their papers we have counts for
7 papers · 1 filter
SimCT: A Simple Consistency Test Protocol in LLMs Development Lifecycle
Fufangchen Zhao, Guoqiang Jin, Rui Zhao +2
In this work, we report our efforts to advance the standard operation procedure of developing Large Language Models (LLMs) or LLMs-based systems or services in industry. We introdu…
CMR Scaling Law: Predicting Critical Mixture Ratios for Continual Pre-training of Language Models
Jiawei Gu, Zacc Yang, Chuanghao Ding +2
Large Language Models (LLMs) excel in diverse tasks but often underperform in specialized fields due to limited domain-specific or proprietary corpus. Continual pre-training (CPT)…
Balancing Speciality and Versatility: A Coarse to Fine Framework for Mitigating Catastrophic Forgetting in Large Language Models
Hengyuan Zhang, Yanru Wu, Dawei Li +4
Aligned Large Language Models (LLMs) showcase remarkable versatility, capable of handling diverse real-world tasks. Meanwhile, aligned LLMs are also expected to exhibit speciality,…
Consistency Matters: Explore LLMs Consistency From a Black-Box Perspective
Fufangchen Zhao, Guoqiang Jin, Jiaheng Huang +2
Nowadays both commercial and open-source academic LLM have become the mainstream models of NLP. However, there is still a lack of research on LLM consistency, meaning that througho…
PUnifiedNER: A Prompting-based Unified NER System for Diverse Datasets
Jinghui Lu, Rui Zhao, Brian Mac Namee +1
Much of named entity recognition (NER) research focuses on developing dataset-specific models based on data from the domain of interest, and a limited set of related entity types.…
SDA: Simple Discrete Augmentation for Contrastive Sentence Representation Learning
Dongsheng Zhu, Zhenyu Mao, Jinghui Lu +2
Contrastive learning has recently achieved compelling performance in unsupervised sentence representation. As an essential element, data augmentation protocols, however, have not b…
What Makes Pre-trained Language Models Better Zero-shot Learners?
Jinghui Lu, Dongsheng Zhu, Weidong Han +3
Current methods for prompt learning in zeroshot scenarios widely rely on a development set with sufficient human-annotated data to select the best-performing prompt template a post…