NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
116 results for “keyword:"latent reasoning"” in ICLR 2027. 22 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Not All Thinking is Created Equal: Latent Reasoning Discovers a Recurrent Search Algorithm for Depth Generalization▲6.1k2.00
[PDF]What Limits Recursive Reasoning Models: Optimization, Architecture and Test-Time Scaling▼4.3k2.00
[PDF]Adaptive Look, Flexible Thought: Learning How and How Long to Think in Multimodal Reasoning▲14k3.00
[PDF]TTE-Flash: Accelerating Reasoning-based Multimodal Representations via Think-Then-Embed Tokens▲5103.00
[PDF]Structured Latent Reasoning: Refactoring Reasoning with Reusable Primitives for Multimodal LLMs▼4.5k2.67
[PDF]Semantic Interface Distillation: Turning Native Runtimes into Reusable Latent Interfaces▼1.3k1.75
[PDF]LatentHarness: Learning Latent Actions for Memory and Reasoning via Counterfactual Policy Distillation▼1.2k1.67
[PDF]Right Direction, Wrong Step: Geometric Analysis of Finite-Step Failure in Looped Transformers▲3.0k2.33
[PDF]Visual Thought Matters: Learning to Ground and Refine Continuous Reasoning for Generalizable Deepfake Detection▲3952.67
[PDF]Scaffolding Minds: Optimizing Latent Visual Target Representations for Multimodal Reasoning▲5103.00
[PDF]Reason Forward, Reflect When Needed: Predictive Anchoring with Suffix-Evidence for Revisable LLM Reasoning▲2843.00
[PDF]Beyond Uniform Soft Thinking: Adaptive Gating and Calibration of Soft Tokens for Latent Reasoning▲143.33
[PDF]ForeSense: Learning Latent Visual Intuition for Vision-Language-Action Models via Future Experience Distillation▲5.4k3.00
[PDF]Attention Routing Stabilizes Early: Working-Set Inference for Recurrent-Depth Language Models▼13k2.00
[PDF]GradCuit: Credit-Assigned Gradient Flow Enables Robust and Interpretable Test-Time Latent Reasoning▲14k3.00
Withdrawn or desk rejected (22)
[PDF]Same Answer, Different Futures: Judging Latent States by Their Continuations in Looped TransformersDesk rejected
[PDF]Not All Recurrence Is Created Equal: Depth Sensitivity Differs Between Recurrence Dimensions of a Single CheckpointWithdrawn1.33
[PDF]CopT: Contrastive On-Policy Thinking with Continuous Spaces for General and Agentic ReasoningDesk rejected