ICLR 2026 · Workshop

Latent & Implicit Thinking

Going Beyond CoT Reasoning
April 27, 2026  ·  Rio de Janeiro

Welcome

Beyond chain-of-thought:
continuous, recurrent, and diffusion-based reasoning.

Schedule — Morning

09:00–09:10Opening Remarks
09:10–09:40Keynote 1: Lingpeng Kong
09:40–10:10Keynote 2: Tom Goldstein
10:10–10:30Coffee Break
10:30–11:10Oral Presentations
11:10–11:40Keynote 3: Samy Bengio
11:40–12:30Poster Session 1

Schedule — Afternoon

12:30–13:30Lunch
13:30–14:00Keynote 4: Beidi Chen
14:00–14:30Keynote 5: Lisa Li
14:30–15:00Keynote 6: Yuandong Tian
15:00–15:30Coffee Break
15:30–16:00Panel Discussion
16:00–16:50Poster Session 2
16:50–17:00Award & Closing

Organizers

Xinyi Wang
Nikunj Saunshi
Rui-Jie Zhu
Liu Yang
Yuntian Deng
Nishanth Dikkala
Jiaheng Liu
Zhiyuan Li
Wenhao Huang
Sashank J. Reddi
Chongxuan Li
Sanjiv Kumar
Keynote 1 · 09:10–09:40
Lingpeng Kong

Lingpeng Kong

The University of Hong Kong
Keynote 2 · 09:40–10:10
Tom Goldstein

Tom Goldstein

University of Maryland

Coffee Break

10:10 – 10:30

Oral Presentations 10:30 – 11:10

  1. 10:30 Dynamic Large Concept Models: Latent Reasoning in an Adaptive Semantic Space
  2. 10:40 The Power of Power Law: Asymmetry Enables Compositional Reasoning
  3. 10:50 Think-at-Hard: Selective Latent Iterations to Improve Reasoning Language Models
  4. 11:00 When does Chain-of-Thought Help: A Markovian Perspective
Keynote 3 · 11:10–11:40
Samy Bengio

Samy Bengio

Apple

Poster Session 1

11:40 – 12:30

Lunch

12:30 – 13:30

Welcome Back — This Afternoon

13:30–14:00Keynote 4: Beidi Chen
14:00–14:30Keynote 5: Lisa Li
14:30–15:00Keynote 6: Yuandong Tian
15:00–15:30Coffee Break
15:30–16:00Panel Discussion
16:00–16:50Poster Session 2
16:50–17:00Award & Closing
Keynote 4 · 13:30–14:00
Beidi Chen

Beidi Chen

Carnegie Mellon University
Keynote 5 · 14:00–14:30
Lisa Xiang Li

Lisa Xiang Li

Stanford University
Keynote 6 · 14:30–15:00
Yuandong Tian

Yuandong Tian

Coffee Break

15:00 – 15:30
Panel · 15:30–16:00

Panel Discussion

What are the key challenges in enabling large language models to reason effectively in latent space?

Poster Session 2

16:00 – 16:50
Best Paper Award
Dynamic Large Concept Models:
Latent Reasoning in an Adaptive Semantic Space
Xingwei Qu, Shaowen Wang, Zihao Huang, Kai Hua, Fan Yin, Rui-Jie Zhu, Jundong Zhou, Qiyang Min, Zihao Wang, Yizhi Li, Tianyu Zhang, He Xing, Zheng Zhang, Yuxuan Song, Tianyu Zheng, Zhiyuan Zeng, Chenghua Lin, Ge Zhang, Wenhao Huang

Thank you