# BiPFT: Binary Pre-trained Foundation Transformer with Low-rank Estimation of Binarization Residual Polynomials

arXiv: https://arxiv.org/abs/2312.08937 · HTML: https://mlsanity.com/paper/2312.08937/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 36 references that are in the map.
- First posted on arXiv 14 Dec 2023.

## Builds on (most cited first, 20 of 20)

- [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)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers](https://mlsanity.com/paper/2002.10957/) (2020)
- [DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing](https://mlsanity.com/paper/2111.09543/) (2021)
- [Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books](https://mlsanity.com/paper/1506.06724/) (2015)
- [MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices](https://mlsanity.com/paper/2004.02984/) (2020)
- [I-BERT: Integer-only BERT Quantization](https://mlsanity.com/paper/2101.01321/) (2021)
- [Bi-Real Net: Enhancing the Performance of 1-bit CNNs With Improved Representational Capability and Advanced Training Algorithm](https://mlsanity.com/paper/1808.00278/) (2018)
- [Compressing BERT: Studying the Effects of Weight Pruning on Transfer Learning](https://mlsanity.com/paper/2002.08307/) (2020)
- [ReActNet: Towards Precise Binary Neural Network with Generalized Activation Functions](https://mlsanity.com/paper/2003.03488/) (2020)
- [Training Binary Neural Networks with Real-to-Binary Convolutions](https://mlsanity.com/paper/2003.11535/) (2020)
- [Deep Transformers without Shortcuts: Modifying Self-attention for Faithful Signal Propagation](https://mlsanity.com/paper/2302.10322/) (2023)
- [DQ-BART: Efficient Sequence-to-Sequence Model via Joint Distillation and Quantization](https://mlsanity.com/paper/2203.11239/) (2022)
- [SiMaN: Sign-to-Magnitude Network Binarization](https://mlsanity.com/paper/2102.07981/) (2021)
- [Gradient Estimation for Binary Latent Variables via Gradient Variance Clipping](https://mlsanity.com/paper/2208.06124/) (2022)
- [Towards Accurate Binary Neural Networks via Modeling Contextual Dependencies](https://mlsanity.com/paper/2209.01404/) (2022)

## Built on by (most cited first, 3 of 3)

- [Not All Layers of LLMs Are Necessary During Inference](https://mlsanity.com/paper/2403.02181/) (2024)
- [SpikeLM: Towards General Spike-Driven Language Modeling via Elastic Bi-Spiking Mechanisms](https://mlsanity.com/paper/2406.03287/) (2024)
- [SpikeLLM: Scaling up Spiking Neural Network to Large Language Models via Saliency-based Spiking](https://mlsanity.com/paper/2407.04752/) (2024)

## Abstract (arXiv)

Pretrained foundation models offer substantial benefits for a wide range of downstream tasks, which can be one of the most potential techniques to access artificial general intelligence. However, scaling up foundation transformers for maximal task-agnostic knowledge has brought about computational challenges, especially on resource-limited devices such as mobiles. This work proposes the first Binary Pretrained Foundation Transformer (BiPFT) for natural language understanding (NLU) tasks, which remarkably saves 56 times operations and 28 times memory. In contrast to previous task-specific binary transformers, BiPFT exhibits a substantial enhancement in the learning capabilities of binary neural networks (BNNs), promoting BNNs into the era of pre-training. Benefiting from extensive pretraining data, we further propose a data-driven binarization method. Specifically, we first analyze the binarization error in self-attention operations and derive the polynomials of binarization error. To simulate full-precision self-attention, we define binarization error as binarization residual polynomials, and then introduce low-rank estimators to model these polynomials. Extensive experiments validate the effectiveness of BiPFTs, surpassing task-specific baseline by 15.4% average performance on the GLUE benchmark. BiPFT also demonstrates improved robustness to hyperparameter changes, improved optimization efficiency, and reduced reliance on downstream distillation, which consequently generalize on various NLU tasks and simplify the downstream pipeline of BNNs. Our code and pretrained models are publicly available at https://github.com/Xingrun-Xing/BiPFT.