Xiao Liu

A new method ensures reliable explanations for DRL agent decisions.

New method enhances watermarking for better image ownership protection.

Learn about evaluating argument strength for better decision-making and communication.

Additive manufacturing faces a significant gap in available open datasets for machine learning.

This research investigates how tools can aid LLMs in managing complex environments.

A new benchmark assesses data handling skills of Large Language Models.

This study uses new methods to help AI learn Zhuang with limited resources.

A study on how teams use personas to improve software design.

This article discusses the effectiveness of MCQs in testing LLMs compared to long-form questions.

ChatGLM-RLHF improves AI interactions through human feedback and advanced training methods.

A new method enhances both language and math skills in language models.

New web navigation agent based on ChatGLM3-6B model enhances task performance.

This article examines how requirements are gathered in government software projects.

Coin3D makes 3D modeling easy for everyone, regardless of skill level.

GLM-4 models show improved capabilities in language understanding and generation.

BrainMAE approach enhances analysis of brain data using fMRI.

Improving how models handle evidence in long documents builds user trust.

Exploring how CSE task recommendations help software practitioners find suitable tasks.

Researchers enhance 3D shape learning using diverse data sources for improved machine understanding.

New methods improve automated question-answer relationships in writing analysis.

A new method to assess how well LLMs understand and apply rules.

Researchers manipulate light direction using atomic motion and topological models for new optical devices.

AutoGLM simplifies interactions with technology, enhancing everyday tasks efficiently.

LLaVA improves Visual Question Answering by blending local device power with cloud processing.

Learn how human feedback shapes AI language model responses.

A fresh approach to enhance instruction-following in language models.

Machines now generate images and videos based on human preferences.