<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Mingming Sun · 孙明明</title><link>https://sunmingming.art/</link><description>Recent content on Mingming Sun · 孙明明</description><generator>Hugo -- gohugo.io</generator><language>en</language><lastBuildDate>Fri, 29 Sep 2023 00:00:00 +0000</lastBuildDate><atom:link href="https://sunmingming.art/index.xml" rel="self" type="application/rss+xml"/><item><title>News</title><link>https://sunmingming.art/blogs/acceptance-of-two-papers-at-neurips-2023/</link><pubDate>Fri, 29 Sep 2023 00:00:00 +0000</pubDate><guid>https://sunmingming.art/blogs/acceptance-of-two-papers-at-neurips-2023/</guid><description>Acceptance of Two Papers at NeurIPS 2023 Link to heading Differentiable Neuro-Symbolic Reasoning on Large-Scale Knowledge Graphs Link to heading Our second paper aims to enhance the scalability of MLN reasoning on large-scale knowledge graphs. The first paper in this series, MLN4KB, was published at WWW 2023. This paper introduces the DiffLogic algorithm, offering the following key contributions:
DiffLogic harnesses the strengths of KG-embedding models and rule-based models, blending efficiency, effectiveness, the capacity to leverage prior knowledge, and the ability to handle uncertainty.</description></item><item><title>About</title><link>https://sunmingming.art/about/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://sunmingming.art/about/</guid><description>I am a professor at the Beijing Institute of Mathematical Sciences and Applications (BIMSA). My research asks how AI systems can build usable representations of the world, check their reasoning against evidence, and improve through interaction. BIMSA faculty profile
I began with geometry and representation learning, then worked on structured language understanding, knowledge bases, and scalable inference. These threads now meet in my work on AI research and in my teaching.</description></item><item><title>News archive</title><link>https://sunmingming.art/news/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://sunmingming.art/news/</guid><description> December 2025 · Presented Analytic Natural Language Processing for Cognitive Computation at the joint BIMSA–USTB symposium. May 2025 · Spoke about AI and complex systems at BIMSA. September 2024 · Joined BIMSA.</description></item><item><title>Publications</title><link>https://sunmingming.art/publications/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://sunmingming.art/publications/</guid><description>Google Scholar · ACL Anthology
By field Link to heading NLP foundations · NLP applications · Knowledge and reasoning · Computer vision · Machine learning methods · Machine learning systems · Industry applications · Facial analysis · Manifold learning
Foundations of natural language processing Link to heading Lingua-Graph: A Unified Representation of Cross-Task Common Substructures for Analytic Language Processing. Mingming Sun, Runze Jiang, Zhu Zhangchenxi, Minlong Peng, Yunfeng Cai. ACL 2026, pp.</description></item><item><title>Research</title><link>https://sunmingming.art/research/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://sunmingming.art/research/</guid><description>Research philosophy Link to heading Understanding the world requires a system of concepts that supports computation and logical reasoning.
How AI and science deepen understanding AI proposes revisions Suggest new concepts, models, and rules to explain the evidence.
Conceptual, computational, and logical systems Define what to check, compute expected results, and specify what follows from the rules.
Verify AI outputs Compare predictions with observations and check conclusions against premises and rules.</description></item><item><title>Systems</title><link>https://sunmingming.art/systems/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://sunmingming.art/systems/</guid><description>Research systems Link to heading Lingua-Graph parser. First author of the research introducing Lingua-Graph and its baseline parser, which produces a shared graph for several language understanding tasks. Code · Paper Logician and Orator. First author of the work on Logician, which extracts structured facts from text, and of the dual system that pairs it with Orator to express facts in language. Logician · Dual system MLN4KB. Co-author of a Julia engine for Markov logic inference over large, sparse knowledge bases.</description></item><item><title>Teaching</title><link>https://sunmingming.art/teaching/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://sunmingming.art/teaching/</guid><description>Natural Language Understanding Link to heading Qiuzhen College, Tsinghua University · Spring 2025 to Autumn 2026
The course starts with how to represent language and knowledge and builds toward language modeling, reasoning, and intelligent systems. Students examine neural and symbolic approaches and practice testing what their systems infer.
Open the course website for classroom slides.</description></item><item><title>Writing &amp; Talks</title><link>https://sunmingming.art/writing/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://sunmingming.art/writing/</guid><description>December 18, 2025 · 15:00–15:20 · First Joint BIMSA–USTB Symposium on Mathematical and AI Applications
Analytic Natural Language Processing for Cognitive Computation Introduces Lingua-Graph and shows how entity recognition, coreference, and relation extraction can work with substructures of a shared graph.
BIMSA program and abstract October 24, 2025 · USTB Shing-Tung Yau Mathematics Experimental Class visit to BIMSA
The Revival of AI and the New Role of Mathematics 人工智能的复兴与数学的新使命</description></item></channel></rss>