IBM Granite Vision tops the chart for small models in document understanding
News
Mike Murphy
Generative AI
Generative AI is revolutionizing how the world computes, but there are still roadblocks for scaling to solve real business problems. At IBM Research, we’re working on trustworthy models, software, and infrastructure to allow any business to harness the power of generative AI.
Our work
IBM’s Dmitry Krotov wants to crack the ‘physics’ of memory
ResearchKim MartineauA faster way to screen supply chains for harmful chemicals
ResearchKim MartineauA 360 review of AI agent benchmarks
ResearchKim MartineauThe simplest protocol for AI agents to work together
ResearchKim MartineauAn invisible watermark to keep tabs on tabular data
ResearchKim Martineau- See more of our work on Generative AI
Publications
ConCodeEval: Evaluating Large Language Models for Code Constraints in Domain-Specific Languages
- Mehant Kammakomati
- Sameer Pimparkhede
- et al.
- 2025
- ACL 2025
Protecting Users From Themselves: Safeguarding Contextual Privacy in Interactions with Conversational Agents
- Ivoline Ngong
- Swanand Ravindra Kadhe
- et al.
- 2025
- ACL 2025
Multi-Level Explanations for Generative Language Models
- Lucas Monteiro Paes
- Dennis Wei
- et al.
- 2025
- ACL 2025
AUTOCIRCUIT-RL: Reinforcement Learning-Driven LLM for Automated Circuit Topology Generation
- Prashanth Vijayaraghavan
- Luyao Shi
- et al.
- 2025
- ICML 2025
Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead
- Rickard Gabrielsson
- Jiacheng Zhu
- et al.
- 2025
- ICML 2025
Regress, Don't Guess: A Regression-like Loss on Number Tokens for Language Models
- Jonas Zausinger
- Lars Pennig
- et al.
- 2025
- ICML 2025