# Block-Based Double Decoders

arXiv: https://arxiv.org/abs/2605.18807 · HTML: https://mlsanity.com/paper/2605.18807/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 18 references that are in the map.
- First posted on arXiv 11 May 2026.

## Builds on (most cited first, 15 of 15)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Big Bird: Transformers for Longer Sequences](https://mlsanity.com/paper/2007.14062/) (2020)
- [Training Large Language Models to Reason in a Continuous Latent Space](https://mlsanity.com/paper/2412.06769/) (2024)
- [Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach](https://mlsanity.com/paper/2502.05171/) (2025)
- [Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer](https://mlsanity.com/paper/2203.03466/) (2022)
- [Reasoning with Latent Thoughts: On the Power of Looped Transformers](https://mlsanity.com/paper/2502.17416/) (2025)
- [You Only Cache Once: Decoder-Decoder Architectures for Language Models](https://mlsanity.com/paper/2405.05254/) (2024)
- [Parcae: Scaling Laws For Stable Looped Language Models](https://mlsanity.com/paper/2604.12946/) (2026)
- [Stability and Generalization in Looped Transformers](https://mlsanity.com/paper/2604.15259/) (2026)
- [Return of the Encoder: Maximizing Parameter Efficiency for SLMs](https://mlsanity.com/paper/2501.16273/) (2025)

## Abstract (arXiv)

Encoder-decoder models offer substantial inference-time savings over decoder-only models, but their pretraining objectives suffer from sparse supervision and dynamic sequence lengths, keeping them out of practice at scale. We propose block-based double decoders, a novel transformer architecture that utilizes doubly-causal block-based attention masks to train with full loss supervision and static sequence packing, combining decoder-only training efficiency with encoder-decoder inference efficiency. In scaling law experiments, block-based double decoders strongly outperform encoder-decoders and closely track decoder-only models across scales. At inference time, they cut KV-cache memory and per-token compute by at least 2/3 without sacrificing prefill caching or other existing inference optimizations available to decoder-only models.