# BitSkip: An Empirical Analysis of Quantization and Early Exit Composition in Transformers

arXiv: https://arxiv.org/abs/2510.23766 · HTML: https://mlsanity.com/paper/2510.23766/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Early exit and edge inference, one of 12 directions in Efficient LLM inference and serving.
- Builds on 27 of its 30 references that are in the map.
- First posted on arXiv 27 Oct 2025.

## Placed in

- Efficient LLM inference and serving / [Early exit and edge inference](https://mlsanity.com/inference/early-exit-and-edge-inference/)

## Builds on (most cited first, 27 of 27)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference](https://mlsanity.com/paper/1712.05877/) (2017)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints](https://mlsanity.com/paper/2305.13245/) (2023)
- [BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks](https://mlsanity.com/paper/1709.01686/) (2017)
- [BinaryConnect: Training Deep Neural Networks with binary weights during propagations](https://mlsanity.com/paper/1511.00363/) (2015)
- [Adaptive Computation Time for Recurrent Neural Networks](https://mlsanity.com/paper/1603.08983/) (2016)
- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [TinyStories: How Small Can Language Models Be and Still Speak Coherent English?](https://mlsanity.com/paper/2305.07759/) (2023)
- [QuIP#: Even Better LLM Quantization with Hadamard Incoherence and Lattice Codebooks](https://mlsanity.com/paper/2402.04396/) (2024)
- [DeeBERT: Dynamic Early Exiting for Accelerating BERT Inference](https://mlsanity.com/paper/2004.12993/) (2020)
- [Confident Adaptive Language Modeling](https://mlsanity.com/paper/2207.07061/) (2022)
- [The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits](https://mlsanity.com/paper/2402.17764/) (2024)
- [LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding](https://mlsanity.com/paper/2404.16710/) (2024)
- [Depth-Adaptive Transformer](https://mlsanity.com/paper/1910.10073/) (2019)
- [BitNet: Scaling 1-bit Transformers for Large Language Models](https://mlsanity.com/paper/2310.11453/) (2023)
- [Efficient Large Language Models: A Survey](https://mlsanity.com/paper/2312.03863/) (2023)
- [Primer: Searching for Efficient Transformers for Language Modeling](https://mlsanity.com/paper/2109.08668/) (2021)
- [Compute-Efficient Deep Learning: Algorithmic Trends and Opportunities](https://mlsanity.com/paper/2210.06640/) (2022)
- [BitNet b1.58 2B4T Technical Report](https://mlsanity.com/paper/2504.12285/) (2025)
- [BitNet v2: Native 4-bit Activations with Hadamard Transformation for 1-bit LLMs](https://mlsanity.com/paper/2504.18415/) (2025)
- [DEL: Context-Aware Dynamic Exit Layer for Efficient Self-Speculative Decoding](https://mlsanity.com/paper/2504.05598/) (2025)

## Built on by (most cited first, 1 of 1)

- [Quantizing Recursive Reasoning Models](https://mlsanity.com/paper/2607.16237/) (2026)

## Abstract (arXiv)

The pursuit of efficient Large Language Models (LLMs) has led to increasingly complex techniques like extreme quantization and dynamic routing. While individual benefits of these methods are well-documented, their compositional effects remain poorly understood. This paper introduces BitSkip, a hybrid architectural framework for systematically exploring these interactions. Counter-intuitively, our findings reveal that a simple 8-bit quantized model without Hadamard transform (BitSkip-V1) not only outperforms its more complex 4-bit and Hadamard-enhanced counterparts but also competes the full-precision baseline in quality (perplexity of 1.13 vs 1.19) . The introduction of Hadamard transforms, even at 8-bit precision, catastrophically degraded performance by over 37,000%, tracing fundamental training instability. Our BitSkip-V1 recipe demonstrates superior early-exit characteristics, with layer 18 providing optimal 32.5% speed gain for minimal 4% quality loss.