Alfio Gliozzo

Michael Glass

Peter Hess

Dinesh Khandelwal

Dinesh Garg

Parag Singla

Shourya Aggarwal

Divyanshu Mandowara

Vishwajeet Agrawal

Salim Roukos

Alexander Gray

Pavan Kapanipathi

Pierre Dognin

Mustafa Canim

Chuang Gan

Abhishek Bhandwaldar

Dan Gutfreund

Research

Q & A

Release

IBM’s new AI outperforms competition in table entry search with question-answering

IBM, MIT and Harvard release “Common Sense AI” dataset at ICML 2021

AI, you have a lot of explaining to do

Getting AI to reason: using neuro-symbolic AI for knowledge-based question answering

 DualTKB: A Dual Learning Bridge between Text and Knowledge Base

How close are we to real AI reasoning? 

AAAI 2025

ACS Spring 2025

ICASSP 2025

ICLR 2025

Trustworthy AI

Trustworthy Generation

Knowledge and Reasoning

Natural Language Processing

Neuro-symbolic AI

Generative AI

Explainable AI

For AI to be useful, it needs to understand what’s being asked of it, and to know how to respond. At IBM Research, we’re working on systems to help AI better reason with the tasks it’s presented, such as understanding context and analogies, comprehension, and planning through scenarios. 

It can take over 10 years to come up with new materials. At IBM Research, we’re looking to accelerate the discovery process using new AI methods, robotics, the hybrid cloud, and quantum computers. Our goal is to unlock new properties and materials to address global challenges in years not decades.

Materials Discovery

Much of the information that can help transform enterprises is locked away in text, like documents, tables, and charts. We’re building advanced AI systems that can parse vast bodies of text to help unlock that data, but also ones flexible enough to be applied to any language problem.

We see Neuro-symbolic AI as a pathway to achieve artificial general intelligence. By augmenting and combining the strengths of statistical AI, like machine learning, with the capabilities of human-like symbolic knowledge and reasoning, we're aiming to create a revolution in AI, rather than an evolution. 

The demand for virtual agents that can handle customer needs has continued to increase dramatically. At IBM Research, we’re building the next generation of artificial intelligence systems that can understand what’s being asked of them and how best to respond as efficiently as possible.

Conversational AI

Francesca Rossi looks at the camera with her arms crossed. Over her shoulder, text reads "Can AI models actually reason?"

Knowledge and Reasoning

Our work

How close are we to real AI reasoning?

AI, you have a lot of explaining to do

IBM, MIT and Harvard release “Common Sense AI” dataset at ICML 2021

IBM’s new AI outperforms competition in table entry search with question-answering

Getting AI to reason: using neuro-symbolic AI for knowledge-based question answering

DualTKB: A Dual Learning Bridge between Text and Knowledge Base

Publications

A Generalist Hanabi Agent

Epigraph Based Multilevel Optimization (EMO) for Enhancing Chain-of-Thought Reasoning Capabilities

Comparative Study of Open-source LLMs for text classification and knowledge extraction in the Material Discovery Domain

Foundation models for materials discovery – current state and future directions

Neural Reasoning Networks: Efficient interpretable neural networks with automatic textual explanations

ULKB Logic: A HOL-based Framework for Reasoning over Knowledge Graphs

Related topics