Think-at-Hard: Dynamic Looped Transformers for Improved Reasoning
cs.CL, cs.AI, cs.LG, cs.PF
Submitted: 2025-11-11
Updated: 2026-08-28
Comments: Accepted by ICML'26
Code: https://github.com/thu-nics/TaH
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- EE-LLM: Large-Scale Training and Inference of Early-Exit Large Language Models with 3D Parallelism
- Compressed Chain of Thought: Efficient Reasoning Through Dense Representations
- Training Verifiers to Solve Math Word Problems
- Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
- Program Synthesis with Large Language Models
- SkipDecode: Autoregressive Skip Decoding with Batching and Caching for Efficient LLM Inference
- Fast and Robust Early-Exiting Framework for Autoregressive Language Models with Synchronized Parallel Decoding
- R2R: Efficiently Navigating Divergent Reasoning Paths with Small-Large Model Token Routing
- Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation
- Eliciting Latent Predictions from Transformers with the Tuned Lens
- Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
- Learning to Insert [PAUSE] Tokens for Better Reasoning
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Implicit Reasoning in Large Language Models: A Comprehensive Survey
- Training Large Language Models to Reason in a Continuous Latent Space
- Rho-1: Not All Tokens Are What You Need
- OlympiadBench: A Challenging Benchmark for Promoting AGI with Olympiad-Level Bilingual Multimodal Scientific Problems
- Expediting and Elevating Large Language Model Reasoning via Hidden Chain-of-Thought Decoding
- Adaptive Layer-skipping in Pre-trained LLMs
- Measuring Mathematical Problem Solving With the MATH Dataset
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering