Skip To Main Content
Mingkai Deng on Muck Rack

Mingkai Deng

Get in touch with Mingkai

Contact Mingkai, search articles and posts on X, monitor coverage, and track replies from one place.

Learn more about Muck Rack

Actions

Share this page

Articles

[2507.05169] Critiques of World Models

|
By Mingkai Deng, Jinyu Hou, Zhiting Hu
| arXiv
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Open in Who Shared View Duplicates (423) Issue with article?

RLPrompt: Optimizing discrete text prompts with reinforcement learning - ΑΙhub

|
By Mingkai Deng, Lucy Smith, Daniele Meli
| aihub.org
Figure 1: Overview of RL Prompt for discrete prompt optimization. All language models (LMs) are frozen. We build our policy network by training a task-specific multi-layer perceptron (MLP) network inserted into a frozen pre-trained LM. The figure above illustrates 1) generation of a prompt (left), 2) example usages in a masked LM for classification (top right) and a left-to-right LM for generation (bottom right), and 3) update of the MLP using RL reward signals (red arrows).
Open in Who Shared Issue with article?

Compression, transduction, and creation: a unified framework for evaluating natural language generation - ΑΙhub

|
By Mingkai Deng, Lucy Smith
| aihub.org
Figure 1: Our framework classifies language generation tasks into compression, transduction, and creation (left), and unifies the evaluation (middle) of key quality aspects with the common operation of information alignment (right). By Mingkai Deng TL;DR: Evaluating natural language generation (NLG) is hard. Our general framework helps solve the difficulty by unifying the evaluation with a common central operation.
Open in Who Shared Issue with article?
See all 3 articles