Zobrazeno 1 - 10
of 69 296
pro vyhledávání: '"Jy On"'
Autor:
Choi, Chanyeol, Kim, Junseong, Lee, Seolhwa, Kwon, Jihoon, Gu, Sangmo, Kim, Yejin, Cho, Minkyung, Sohn, Jy-yong
This report explores the enhancement of text retrieval performance using advanced data refinement techniques. We develop Linq-Embed-Mistral\footnote{\url{https://huggingface.co/Linq-AI-Research/Linq-Embed-Mistral}} by building on the E5-mistral and M
Externí odkaz:
http://arxiv.org/abs/2412.03223
Autor:
Dai, Shenghong, Sohn, Jy-yong, Chen, Yicong, Alam, S M Iftekharul, Balakrishnan, Ravikumar, Banerjee, Suman, Himayat, Nageen, Lee, Kangwook
Continual Federated Learning (CFL) is essential for enabling real-world applications where multiple decentralized clients adaptively learn from continuous data streams. A significant challenge in CFL is mitigating catastrophic forgetting, where model
Externí odkaz:
http://arxiv.org/abs/2409.01585
Fine-tuning large pre-trained models is a common practice in machine learning applications, yet its mathematical analysis remains largely unexplored. In this paper, we study fine-tuning through the lens of memorization capacity. Our new measure, the
Externí odkaz:
http://arxiv.org/abs/2408.00359
Making decent multi-lingual sentence representations is critical to achieve high performances in cross-lingual downstream tasks. In this work, we propose a novel method to align multi-lingual embeddings based on the similarity of sentences measured b
Externí odkaz:
http://arxiv.org/abs/2405.16155
Autor:
So, Jae-hee, Chang, Joonhwan, Kim, Eunji, Na, Junho, Choi, JiYeon, Sohn, Jy-yong, Kim, Byung-Hoon, Chu, Sang Hui
Recent advancements in Large Language Models (LLMs) have accelerated their usage in various domains. Given the fact that psychiatric interviews are goal-oriented and structured dialogues between the professional interviewer and the interviewee, it is
Externí odkaz:
http://arxiv.org/abs/2403.17428
Improving the accessibility of psychotherapy with the aid of Large Language Models (LLMs) is garnering a significant attention in recent years. Recognizing cognitive distortions from the interviewee's utterances can be an essential part of psychother
Externí odkaz:
http://arxiv.org/abs/2403.14255
Mitigating hallucination issues is a key challenge that must be overcome to reliably deploy large language models (LLMs) in real-world scenarios. Recently, various methods have been proposed to detect and revise factual errors in LLM-generated texts,
Externí odkaz:
http://arxiv.org/abs/2402.17097
Publikováno v:
Proceedings of the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024, Valencia, Spain
Contrastive learning has emerged as a prominent branch of self-supervised learning for several years. Especially, CLIP, which applies contrastive learning to large sets of captioned images, has garnered significant attention. Recently, SigLIP, a vari
Externí odkaz:
http://arxiv.org/abs/2402.12613
Chain-of-thought (CoT) prompting is a simple and effective method for improving the reasoning capabilities of Large Language Models (LLMs). The basic idea of CoT is to let LLMs break down their thought processes step-by-step by putting exemplars in t
Externí odkaz:
http://arxiv.org/abs/2402.10645
Publikováno v:
Clinical, Cosmetic and Investigational Dermatology, Vol Volume 17, Pp 2945-2952 (2024)
Ji Young Um,1,* Min Gyu Choi,2,* Han Bi Kim,1 So Yeon Lee,1 Gi Hyun Park,3 Hea Won Kim,3 Bo Young Chung,1 Chun Wook Park,1 Hye One Kim1 1Department of Dermatology, College of Medicine, Hallym University, Kangnam Sacred Heart Hospital, Seoul,
Externí odkaz:
https://doaj.org/article/53f394646ca3406b9ac816ece059f768
