Data Efficient Any Transformer-to-Mamba Distillation via Attention Bridge
cs.LG
Submitted: 2025-10-22
Updated: 2026-09-16
Code: https://github.com/wph6/CAB
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Transferring Inductive Biases through Knowledge Distillation
- The Hidden Attention of Mamba Models
- Llamba: Scaling Distilled Recurrent Models for Efficient Language Processing
- TransMamba: Fast Universal Architecture Adaption from Transformers to Mamba
- Mamba: Linear-Time Sequence Modeling with Selective State Spaces
- Distilling the Knowledge in a Neural Network
- Scaling Laws for Neural Language Models
- Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence
- RWKV-7 "Goose" with Expressive Dynamic State Evolution
- FitNets: Hints for Thin Deep Nets
- REPA Works Until It Doesn't: Early-Stopped, Holistic Alignment Supercharges Diffusion Training
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: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks