R
Published on

ICLR 2026 — Highlights

ICLR 2026 — Highlights

14th International Conference on Learning Representations Singapore • Apr 24–28, 2026

5,355 papers accepted (27.4% — 19,525 submissions) Source: ICLR 2026 Blog


🏆 Outstanding Paper Awards

Outstanding Papers

  1. Transformers are Inherently Succinct — Bergsträßer, Cotterell, Lin
  2. LLMs Get Lost In Multi-Turn Conversation — Laban, Hayashi, Zhou, Neville

Honorable Mention

  • The Polar Express: Optimal Matrix Sign Methods — Amsel, Persson, Musco, Gower

Test of Time Awards

  • DCGAN — Radford, Metz, Chintala
  • DDPG / Continuous Control — Lillicrap et al.

Keynote Speakers

  • Maja Matarić
  • Max Welling
  • Percy Liang
  • Katherine Bouman
  • Karen Adolph
  • Pablo Arbeláez

Topics

#TopicPapersOrals
Agents & Tool Use4260
Audio & Speech990
Computer Vision8160
Data-centric & Curation3580
Diffusion Models & Generative AI6720
Efficiency & Compression6100
Graph Neural Networks2420
Graphs & Combinatorial2550
Graphs & Structured Data1100
Interpretability & Mechanistic Interpretability1770
LLMs & Foundation Models15150
Medical & Healthcare1080
Multi-modal & Vision-Language3900
Natural Language Processing590
Optimization for ML450
Optimization & Training Methods4060
Other / Unclassified11010
Reinforcement Learning6570
Robotics & Control2290
Theory & Deep Learning Theory2560
Trust & Safety4420

Browse papers by topic in the folder above.