The llama 3 herd of models

A Dubey, A Jauhri, A Pandey, A Kadian… - arXiv preprint arXiv …, 2024 - arxiv.org
Modern artificial intelligence (AI) systems are powered by foundation models. This paper
presents a new set of foundation models, called Llama 3. It is a herd of language models …

Fine tuning vs. retrieval augmented generation for less popular knowledge

H Soudani, E Kanoulas, F Hasibi - … of the 2024 Annual International ACM …, 2024 - dl.acm.org
Language Models (LMs) memorize a vast amount of factual knowledge, exhibiting strong
performance across diverse tasks and domains. However, it has been observed that the …

[PDF][PDF] A comprehensive survey of small language models in the era of large language models: Techniques, enhancements, applications, collaboration with llms, and …

F Wang, Z Zhang, X Zhang, Z Wu, T Mo, Q Lu… - arXiv preprint arXiv …, 2024 - ai.radensa.ru
Large language models (LLM) have demonstrated emergent abilities in text generation,
question answering, and reasoning, facilitating various tasks and domains. Despite their …

Training on the test task confounds evaluation and emergence

R Dominguez-Olmedo, FE Dorner, M Hardt - arXiv preprint arXiv …, 2024 - arxiv.org
We study a fundamental problem in the evaluation of large language models that we call
training on the test task. Unlike wrongful practices like training on the test data, leakage, or …

Resolving discrepancies in compute-optimal scaling of language models

T Porian, M Wortsman, J Jitsev, L Schmidt… - arXiv preprint arXiv …, 2024 - arxiv.org
Kaplan et al. and Hoffmann et al. developed influential scaling laws for the optimal model
size as a function of the compute budget, but these laws yield substantially different …

Open-Ethical AI: Advancements in Open-Source Human-Centric Neural Language Models

S Sicari, JF Cevallos M, A Rizzardi… - ACM Computing …, 2024 - dl.acm.org
This survey summarises the most recent methods for building and assessing helpful, honest,
and harmless neural language models, considering small, medium, and large-size models …

Skywork-reward: Bag of tricks for reward modeling in llms

CY Liu, L Zeng, J Liu, R Yan, J He, C Wang… - arXiv preprint arXiv …, 2024 - arxiv.org
In this report, we introduce a collection of methods to enhance reward modeling for LLMs,
focusing specifically on data-centric techniques. We propose effective data selection and …

A survey on efficient inference for large language models

Z Zhou, X Ning, K Hong, T Fu, J Xu, S Li, Y Lou… - arXiv preprint arXiv …, 2024 - arxiv.org
Large Language Models (LLMs) have attracted extensive attention due to their remarkable
performance across various tasks. However, the substantial computational and memory …

Empirical guidelines for deploying llms onto resource-constrained edge devices

R Qin, D Liu, C Xu, Z Yan, Z Tan, Z Jia… - arXiv preprint arXiv …, 2024 - arxiv.org
The scaling laws have become the de facto guidelines for designing large language models
(LLMs), but they were studied under the assumption of unlimited computing resources for …

Preference tuning with human feedback on language, speech, and vision tasks: A survey

GI Winata, H Zhao, A Das, W Tang, DD Yao… - arXiv preprint arXiv …, 2024 - arxiv.org
Preference tuning is a crucial process for aligning deep generative models with human
preferences. This survey offers a thorough overview of recent advancements in preference …