Conversational AI
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.
Our work
IBM Granite led open-source LLMs on API calling
NewsKim MartineauPutting more knowledge at the fingertips of non-English speakers
ReleaseKim MartineauProject Debater argues for key point analysis of surveys, reviews, and social media
ReleaseNoam Slonim, Yoav Katz, Roy Bar-Haim, and Elad Venezian5 minute read
Publications
InspectorRAGet: An Introspection Platform for RAG Evaluation
- Benjamin Sznajder
- Kshitij Fadnis
- et al.
- 2025
- NAACL 2025
Selective Self-to-Supervised Fine-Tuning for Generalization in Large Language Models
- Sonam Mishra
- Yatin Nandwani
- et al.
- 2025
- NAACL 2025
Diagnosing and Prioritizing Issues in Automated Order-Taking Systems: A Machine-Assisted Error Discovery Approach
- Maeda Hanafi
- Frederick Reiss
- et al.
- 2025
- CHI 2025
Creating Conversational Datasets for Retrieval-Augmented Generation Applications is Hard: Challenges & Research Opportunities
- Maeda Hanafi
- Kshitij Fadnis
- et al.
- 2025
- CHI 2025
DELIFT: DATA EFFICIENT LANGUAGE MODEL INSTRUCTION FINE-TUNING
- Ishika Agarwal
- Krishnateja Killamsetty
- et al.
- 2025
- ICLR 2025
Controlling AI Agent Participation in Group Conversations: A Human-Centered Approach
- Stephanie Houde
- Krissy Brimijoin
- et al.
- 2025
- IUI 2025
IBM Solution: AI for Customer Service
Innovations from our team were developed into Watson Assistant, IBM’s AI for customer service. It’s the industry’s leading conversational AI platform that customers, human agents, and employees actually want to use.
Learn more