Shallow Queries, Mature Values: Depth-Asynchronous Self-Speculation for Looped Transformers
cs.LG, cs.AI
Submitted: 2026-09-28
Updated: 2026-09-28
Code: https://github.com/apoorvumang/prompt-lookup-decoding
Terminology
Sources
- Program Synthesis with Large Language Models
- Accelerating Large Language Model Decoding with Speculative Sampling
- Evaluating Large Language Models Trained on Code
- LoopSpec: Pipelined Self-Speculative Decoding for Looped Transformers
- Training Verifiers to Solve Math Word Problems
- Universal Transformers
- Break the Sequential Dependency of LLM Inference Using Lookahead Decoding
- Efficient Parallel Samplers for Recurrent-Depth Models and Their Connection to Diffusion Language Models
- Measuring Mathematical Problem Solving With the MATH Dataset
- Teaching Pretrained Language Models to Think Deeper with Retrofitted Recurrence
- Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks