Improving Test-Time Scaling with Adaptive Looped Transformers
cs.CL, cs.LG
Submitted: 2026-09-28
Updated: 2026-09-28
Terminology
Sources
- L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning
- Unlocking Out-of-Distribution Generalization in Transformers via Recursive Latent Space Reasoning
- Program Synthesis with Large Language Models
- Relaxed Recursive Transformers: Effective Parameter Sharing with Layer-wise LoRA
- Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation
- Generative Recursive Reasoning
- PonderNet: Learning to Ponder
- A Mechanistic Analysis of Looped Reasoning Language Models
- Large Language Monkeys: Scaling Inference Compute with Repeated Sampling
- Loop as a Bridge: Can Looped Transformers Truly Link Representation Space and Natural Language Outputs?
- Are More LLM Calls All You Need? Towards Scaling Laws of Compound Inference Systems
- Training-Free Looped Transformers
- Evaluating Large Language Models Trained on Code
- Parallel Scaling Law for Language Models
- DND: Boosting Large Language Models with Dynamic Nested Depth
- Think Deep, Not Just Long: Measuring LLM Reasoning Effort via Deep-Thinking Tokens
- LT2: Linear-Time Looped Transformers
- LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding
- Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers
- Adaptive Loops and Memory in Transformers: Think Harder or Know More?
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