The largest ML research conference globally. Where the year's most-cited AI research lands.
NeurIPS (Conference on Neural Information Processing Systems) is the most prestigious and most broadly attended AI research conference, alternating between Montreal and Vancouver with satellite workshops across the globe. The 2026 Vancouver edition is expected to draw 20,000+ registered participants and feature 3,000+ paper acceptances. NeurIPS proceedings define the state of the art across the full AI stack: foundation model training methods, reinforcement learning from human feedback, alignment techniques, AI safety, interpretability, and ML applications in biology, climate, and economics.
For enterprise teams with in-house AI research capability — typically those running 10+ person ML or data science teams — NeurIPS is the most important annual event for recruiting, research partnerships, and staying current on techniques that will reach production in 24–36 months. The associated workshops (running over two days before the main conference) are often more valuable than main-track sessions for practitioners: they allow deeper engagement with specific research communities at the frontier of enterprise-relevant topics.
AIMenta publishes a post-NeurIPS digest for clients, highlighting papers directly relevant to APAC enterprise AI workloads and translating technical findings into product and strategy implications. We also facilitate peer discussion groups for client teams who send delegates — ensuring conference learnings are shared rather than siloed.