# Brainformers: Trading Simplicity for Efficiency

arXiv: https://arxiv.org/abs/2306.00008 · HTML: https://mlsanity.com/paper/2306.00008/

## Key facts

- Cited by 31 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 50 references that are in the map.
- First posted on arXiv 29 May 2023.

## Builds on (most cited first, 30 of 43)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [Identity Mappings in Deep Residual Networks](https://mlsanity.com/paper/1603.05027/) (2016)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems](https://mlsanity.com/paper/1905.00537/) (2019)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Rethinking Attention with Performers](https://mlsanity.com/paper/2009.14794/) (2020)
- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [Linformer: Self-Attention with Linear Complexity](https://mlsanity.com/paper/2006.04768/) (2020)
- [MLP-Mixer: An all-MLP Architecture for Vision](https://mlsanity.com/paper/2105.01601/) (2021)
- [Scaling Language Models: Methods, Analysis & Insights from Training Gopher](https://mlsanity.com/paper/2112.11446/) (2021)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [Adafactor: Adaptive Learning Rates with Sublinear Memory Cost](https://mlsanity.com/paper/1804.04235/) (2018)
- [GLaM: Efficient Scaling of Language Models with Mixture-of-Experts](https://mlsanity.com/paper/2112.06905/) (2021)
- [Mixture-of-Experts with Expert Choice Routing](https://mlsanity.com/paper/2202.09368/) (2022)
- [CoAtNet: Marrying Convolution and Attention for All Data Sizes](https://mlsanity.com/paper/2106.04803/) (2021)
- [Pay Attention to MLPs](https://mlsanity.com/paper/2105.08050/) (2021)
- [BASE Layers: Simplifying Training of Large, Sparse Models](https://mlsanity.com/paper/2103.16716/) (2021)
- [NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection](https://mlsanity.com/paper/1904.07392/) (2019)
- [Hash Layers For Large Sparse Models](https://mlsanity.com/paper/2106.04426/) (2021)

## Built on by (most cited first, 30 of 31)

- [OLMoE: Open Mixture-of-Experts Language Models](https://mlsanity.com/paper/2409.02060/) (2024)
- [OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models](https://mlsanity.com/paper/2402.01739/) (2024)
- [FuseMoE: Mixture-of-Experts Transformers for Fleximodal Fusion](https://mlsanity.com/paper/2402.03226/) (2024)
- [Model Compression and Efficient Inference for Large Language Models: A Survey](https://mlsanity.com/paper/2402.09748/) (2024)
- [Statistical Perspective of Top-K Sparse Softmax Gating Mixture of Experts](https://mlsanity.com/paper/2309.13850/) (2023)
- [Sigmoid Gating is More Sample Efficient than Softmax Gating in Mixture of Experts](https://mlsanity.com/paper/2405.13997/) (2024)
- [A General Theory for Softmax Gating Multinomial Logistic Mixture of Experts](https://mlsanity.com/paper/2310.14188/) (2023)
- [CompeteSMoE -- Effective Training of Sparse Mixture of Experts via Competition](https://mlsanity.com/paper/2402.02526/) (2024)
- [On Least Square Estimation in Softmax Gating Mixture of Experts](https://mlsanity.com/paper/2402.02952/) (2024)
- [Bigger is not Always Better: Scaling Properties of Latent Diffusion Models](https://mlsanity.com/paper/2404.01367/) (2024)
- [Statistical Advantages of Perturbing Cosine Router in Mixture of Experts](https://mlsanity.com/paper/2405.14131/) (2024)
- [Is Temperature Sample Efficient for Softmax Gaussian Mixture of Experts?](https://mlsanity.com/paper/2401.13875/) (2024)
- [Revisiting Prefix-tuning: Statistical Benefits of Reparameterization among Prompts](https://mlsanity.com/paper/2410.02200/) (2024)
- [CHAI: Clustered Head Attention for Efficient LLM Inference](https://mlsanity.com/paper/2403.08058/) (2024)
- [eMoE: Task-aware Memory Efficient Mixture-of-Experts-Based (MoE) Model Inference](https://mlsanity.com/paper/2503.06823/) (2025)
- [Prediction Is All MoE Needs: Expert Load Distribution Goes from Fluctuating to Stabilizing](https://mlsanity.com/paper/2404.16914/) (2024)
- [MoLAE: Mixture of Latent Experts for Parameter-Efficient Language Models](https://mlsanity.com/paper/2503.23100/) (2025)
- [MomentumSMoE: Integrating Momentum into Sparse Mixture of Experts](https://mlsanity.com/paper/2410.14574/) (2024)
- [RepLoRA: Reparameterizing Low-Rank Adaptation via the Perspective of Mixture of Experts](https://mlsanity.com/paper/2502.03044/) (2025)
- [Revisit Visual Prompt Tuning: The Expressiveness of Prompt Experts](https://mlsanity.com/paper/2501.18936/) (2025)
- [MoLEx: Mixture of Layer Experts for Finetuning with Sparse Upcycling](https://mlsanity.com/paper/2503.11144/) (2025)
- [SYMI: Efficient Mixture-of-Experts Training via Model and Optimizer State Decoupling](https://mlsanity.com/paper/2504.19925/) (2025)
- [Sigmoid Self-Attention has Lower Sample Complexity than Softmax Self-Attention: A Mixture-of-Experts Perspective](https://mlsanity.com/paper/2502.00281/) (2025)
- [Mix-of-Language-Experts Architecture for Multilingual Programming](https://mlsanity.com/paper/2506.18923/) (2025)
- [Modeling Expert Interactions in Sparse Mixture of Experts via Graph Structures](https://mlsanity.com/paper/2510.16411/) (2025)
- [WAVE++: Capturing Within-Task Variance for Continual Relation Extraction with Adaptive Prompting](https://mlsanity.com/paper/2505.13944/) (2025)
- [MoME: Mixture of Visual Language Medical Experts for Medical Imaging Segmentation](https://mlsanity.com/paper/2510.26996/) (2025)
- [Life-IQA: Boosting Blind Image Quality Assessment through GCN-enhanced Layer Interaction and MoE-based Feature Decoupling](https://mlsanity.com/paper/2511.19024/) (2025)
- [Compositional Reasoning Depth Predicts Clinical AI Failure: Empirical Evidence Consistent with Transformer Compositionality Limits in Electronic Health Record Question Answering](https://mlsanity.com/paper/2606.16890/) (2026)
- [Mixture-of-Control: State-Aware Fine-Tuning for Transformer-based Models](https://mlsanity.com/paper/2606.31397/) (2026)

## Abstract (arXiv)

Transformers are central to recent successes in natural language processing and computer vision. Transformers have a mostly uniform backbone where layers alternate between feed-forward and self-attention in order to build a deep network. Here we investigate this design choice and find that more complex blocks that have different permutations of layer primitives can be more efficient. Using this insight, we develop a complex block, named Brainformer, that consists of a diverse sets of layers such as sparsely gated feed-forward layers, dense feed-forward layers, attention layers, and various forms of layer normalization and activation functions. Brainformer consistently outperforms the state-of-the-art dense and sparse Transformers, in terms of both quality and efficiency. A Brainformer model with 8 billion activated parameters per token demonstrates 2x faster training convergence and 5x faster step time compared to its GLaM counterpart. In downstream task evaluation, Brainformer also demonstrates a 3% higher SuperGLUE score with fine-tuning compared to GLaM with a similar number of activated parameters. Finally, Brainformer largely outperforms a Primer dense model derived with NAS with similar computation per token on fewshot evaluations.