INSAIT researchers with 8 papers accepted at the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026), a leading conference in natural language processing (NLP), taking place this upcoming week in Hungary.
This year marks the largest EMNLP in the conference’s history, with over 17,000 submissions and an acceptance rate of approximately 15%.
INSAIT’s accepted papers address key challenges in AI and language technologies across 4 research areas:
- Trustworthy AI and factuality – Advancing evidence-based verification and improving the reliability of AI-generated content.
- AI for healthcare – Evaluating AI systems in medical consultations and developing methods for verifying medical claims.
- AI safety and robustness – Investigating harmful content and identifying vulnerabilities in large language models.
- Mathematical reasoning and retrieval – new benchmarks and evaluation methods to assess AI systems’ mathematical reasoning capabilities.
Congratulations to all INSAIT researchers and their collaborators for their contributions to advancing AI research!