Zobrazeno 1 - 10
of 104
pro vyhledávání: '"CHARLES, Denis"'
Autor:
Hu, Xinyu, Tang, Pengfei, Zuo, Simiao, Wang, Zihan, Song, Bowen, Lou, Qiang, Jiao, Jian, Charles, Denis
Large language models (LLMs) have made impressive progress in natural language processing. These models rely on proper human instructions (or prompts) to generate suitable responses. However, the potential of LLMs are not fully harnessed by commonly-
Externí odkaz:
http://arxiv.org/abs/2310.13855
Autor:
Sun, Hong, Li, Xue, Xu, Yinchuan, Homma, Youkow, Cao, Qi, Wu, Min, Jiao, Jian, Charles, Denis
This paper presents AutoHint, a novel framework for automatic prompt engineering and optimization for Large Language Models (LLM). While LLMs have demonstrated remarkable ability in achieving high-quality annotation in various tasks, the key to apply
Externí odkaz:
http://arxiv.org/abs/2307.07415
Named entity recognition (NER) is a crucial task for online advertisement. State-of-the-art solutions leverage pre-trained language models for this task. However, three major challenges remain unresolved: web queries differ from natural language, on
Externí odkaz:
http://arxiv.org/abs/2306.17413
Autor:
Zuo, Simiao, Liu, Xiaodong, Jiao, Jian, Charles, Denis, Manavoglu, Eren, Zhao, Tuo, Gao, Jianfeng
Transformer models have achieved superior performance in various natural language processing tasks. However, the quadratic computational cost of the attention mechanism limits its practicality for long sequences. There are existing attention variants
Externí odkaz:
http://arxiv.org/abs/2212.08136
In this paper, we propose TEDL, a two-stage learning approach to quantify uncertainty for deep learning models in classification tasks, inspired by our findings in experimenting with Evidential Deep Learning (EDL) method, a recently proposed uncertai
Externí odkaz:
http://arxiv.org/abs/2209.05522
Autor:
Pfeiffer III, Joseph J., Charles, Denis, Gilton, Davis, Jung, Young Hun, Parsana, Mehul, Anderson, Erik
Today, many web advertising data flows involve passive cross-site tracking of users. Enabling such a mechanism through the usage of third party tracking cookies (3PC) exposes sensitive user data to a large number of parties, with little oversight on
Externí odkaz:
http://arxiv.org/abs/2110.14794
It is often critical for prediction models to be robust to distributional shifts between training and testing data. From a causal perspective, the challenge is to distinguish the stable causal relationships from the unstable spurious correlations acr
Externí odkaz:
http://arxiv.org/abs/2010.08710
Publikováno v:
Special issue on Causal Inference and Machine Learning with Network Data, Frontiers in Big Data, 2022
In classical causal inference, inferring cause-effect relations from data relies on the assumption that units are independent and identically distributed. This assumption is violated in settings where units are related through a network of dependenci
Externí odkaz:
http://arxiv.org/abs/2010.07458
Autor:
Deshmukh, Aniket Anand, Kumar, Abhimanu, Boyles, Levi, Charles, Denis, Manavoglu, Eren, Dogan, Urun
Contextual bandits are a common problem faced by machine learning practitioners in domains as diverse as hypothesis testing to product recommendations. There have been a lot of approaches in exploiting rich data representations for contextual bandit
Externí odkaz:
http://arxiv.org/abs/2003.08485
Self-supervision is key to extending use of deep learning for label scarce domains. For most of self-supervised approaches data transformations play an important role. However, up until now the impact of transformations have not been studied. Further
Externí odkaz:
http://arxiv.org/abs/2002.07384