Latent & Implicit Thinking Workshop

🏆 LIT Workshop @ ICLR 2026 — Community Choice Award! Vote for your favorite paper from our Best Paper finalists 👇 Details on each paper in the thread 🧵

4️⃣ When does CoT Help? (#discussion" target="_blank" rel="nofollow noopener">openreview.net/forum?id=fz5BC…) @YijunDong1 Models chain-of-thought as a Markov chain and identifies "transition alignment" as the key factor determining when CoT actually helps. Theory + synthetic benchmarks to validate predictions.

3️⃣ Think-at-Hard (#discussion" target="_blank" rel="nofollow noopener">openreview.net/forum?id=3dOB7…) Tackles latent overthinking in looped transformers — a neural decider triggers extra iterations only on hard tokens. Skips 93% of tokens while outperforming always-iterate baselines by 3.8–4.4%.

🏆 LIT Workshop @ ICLR 2026 — Community Choice Award! Vote for your favorite paper from our Best Paper finalists 👇 Details on each paper in the thread 🧵

Organizing Team: @XinyiWang98 @NSaunshi @RidgerZhu @Yang_Liuu @yuntiandeng @NishanthDikkala @liujiaheng2 @zhiyuanli_ Wenhao Huang, Sashank J. Reddi, @LiChongxuan Sanjiv Kumar Speakers: @BeidiChen Samy Bengio @ikekong @tomgoldsteincs @XiangLisaLi2 @tydsh

📢 Call for Papers & Reviewers | Latent & Implicit Thinking (LIT) @ ICLR 2026 🧠 We’re excited to announce the Latent & Implicit Thinking (LIT) Workshop @iclr_conf 2026 in Rio de Janeiro, Brazil! This workshop goes beyond Chain-of-Thought, exploring hidden, efficient, and non-autoregressive reasoning in neural models. 📄 We invite submissions on topics including: • Implicit reasoning training strategies • Looped architectures & special thinking tokens • KV-cache & hybrid stateful reasoning • Inference beyond autoregression • Theory, benchmarks, interpretability, safety & limitations 🗓️ Important dates: • Submission deadline: Feb 5, 2026 • Reviewing period: Feb 6–26, 2026 • Acceptance notification: Mar 1, 2026 • Camera-ready: Mar 10, 2026 • Workshop: Apr 26–27, 2026 @ #ICLR2026 🙌 Reviewer sign-up: forms.gle/ZevZZXi29WKSuA… 📤 Paper submission: openreview.net/group?id=ICLR.… 🔗 Website: latent-implicit-thinking.github.io Join us to shape new paradigms of reasoning in AI 🚀 #ICLR2026 #reasoning