Self-instruct: Aligning language models with self-generated instructions

Y Wang, Y Kordi, S Mishra, A Liu, NA Smith… - arXiv preprint arXiv …, 2022 - arxiv.org
Large" instruction-tuned" language models (ie, finetuned to respond to instructions) have
demonstrated a remarkable ability to generalize zero-shot to new tasks. Nevertheless, they …

Large language models can self-improve

J Huang, SS Gu, L Hou, Y Wu, X Wang, H Yu… - arXiv preprint arXiv …, 2022 - arxiv.org
Large Language Models (LLMs) have achieved excellent performances in various tasks.
However, fine-tuning an LLM requires extensive supervision. Human, on the other hand …

Data-driven causal effect estimation based on graphical causal modelling: A survey

D Cheng, J Li, L Liu, J Liu, TD Le - ACM Computing Surveys, 2024 - dl.acm.org
In many fields of scientific research and real-world applications, unbiased estimation of
causal effects from non-experimental data is crucial for understanding the mechanism …

Weak localization of radiographic manifestations in pulmonary tuberculosis from chest x-ray: A systematic review

DW Feyisa, YM Ayano, TG Debelee, F Schwenker - Sensors, 2023 - mdpi.com
Pulmonary tuberculosis (PTB) is a bacterial infection that affects the lung. PTB remains one
of the infectious diseases with the highest global mortalities. Chest radiography is a …

Balancing logit variation for long-tailed semantic segmentation

Y Wang, J Fei, H Wang, W Li, T Bao… - Proceedings of the …, 2023 - openaccess.thecvf.com
Semantic segmentation usually suffers from a long tail data distribution. Due to the
imbalanced number of samples across categories, the features of those tail classes may get …

Learning from future: A novel self-training framework for semantic segmentation

Y Du, Y Shen, H Wang, J Fei, W Li… - Advances in …, 2022 - proceedings.neurips.cc
Self-training has shown great potential in semi-supervised learning. Its core idea is to use
the model learned on labeled data to generate pseudo-labels for unlabeled samples, and in …

A survey on semi-supervised graph clustering

F Daneshfar, S Soleymanbaigi, P Yamini… - … Applications of Artificial …, 2024 - Elsevier
Abstract Semi-Supervised Graph Clustering (SSGC) has emerged as a pivotal field at the
intersection of graph clustering and semi-supervised learning (SSL), offering innovative …

Temporal-domain adaptation for satellite image time-series land-cover mapping with adversarial learning and spatially aware self-training

E Capliez, D Ienco, R Gaetano… - IEEE Journal of …, 2023 - ieeexplore.ieee.org
Nowadays, satellite image time series (SITS) are commonly employed to derive land-cover
maps (LCM) to support decision makers in a variety of land management applications. In the …

Linguistic steganalysis in few-shot scenario

H Wang, Z Yang, J Yang, C Chen… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
Due to the widespread use of text in cyberspace, linguistic steganography, which hides
secret information into normal texts, develops quickly in these years. While linguistic …

Self-training with direct preference optimization improves chain-of-thought reasoning

T Wang, S Li, W Lu - arXiv preprint arXiv:2407.18248, 2024 - arxiv.org
Effective training of language models (LMs) for mathematical reasoning tasks demands high-
quality supervised fine-tuning data. Besides obtaining annotations from human experts, a …