# Dynamic Context Pruning for Efficient and Interpretable Autoregressive Transformers

arXiv: https://arxiv.org/abs/2305.15805 · HTML: https://mlsanity.com/paper/2305.15805/

## Key facts

- Cited by 67 papers in the map (data through 14 Oct 2026).
- Builds on 52 of its 66 references that are in the map.
- First posted on arXiv 25 May 2023.

## Builds on (most cited first, 30 of 52)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention](https://mlsanity.com/paper/2006.16236/) (2020)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [Big Bird: Transformers for Longer Sequences](https://mlsanity.com/paper/2007.14062/) (2020)
- [Reformer: The Efficient Transformer](https://mlsanity.com/paper/2001.04451/) (2020)
- [Rethinking Attention with Performers](https://mlsanity.com/paper/2009.14794/) (2020)
- [Generating Long Sequences with Sparse Transformers](https://mlsanity.com/paper/1904.10509/) (2019)
- [Linformer: Self-Attention with Linear Complexity](https://mlsanity.com/paper/2006.04768/) (2020)
- [fairseq: A Fast, Extensible Toolkit for Sequence Modeling](https://mlsanity.com/paper/1904.01038/) (2019)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [Energy and Policy Considerations for Deep Learning in NLP](https://mlsanity.com/paper/1906.02243/) (2019)
- [SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot](https://mlsanity.com/paper/2301.00774/) (2023)
- [Perceiver: General Perception with Iterative Attention](https://mlsanity.com/paper/2103.03206/) (2021)
- [Token Merging: Your ViT But Faster](https://mlsanity.com/paper/2210.09461/) (2022)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [On Layer Normalization in the Transformer Architecture](https://mlsanity.com/paper/2002.04745/) (2020)
- [LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale](https://mlsanity.com/paper/2208.07339/) (2022)
- [Efficient Transformers: A Survey](https://mlsanity.com/paper/2009.06732/) (2020)

## Built on by (most cited first, 30 of 67)

- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention](https://mlsanity.com/paper/2407.02490/) (2024)
- [Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Security](https://mlsanity.com/paper/2401.05459/) (2024)
- [A Survey on Efficient Inference for Large Language Models](https://mlsanity.com/paper/2404.14294/) (2024)
- [Efficient Large Language Models: A Survey](https://mlsanity.com/paper/2312.03863/) (2023)
- [Keyformer: KV Cache Reduction through Key Tokens Selection for Efficient Generative Inference](https://mlsanity.com/paper/2403.09054/) (2024)
- [Transformers are Multi-State RNNs](https://mlsanity.com/paper/2401.06104/) (2024)
- [Dynamic Memory Compression: Retrofitting LLMs for Accelerated Inference](https://mlsanity.com/paper/2403.09636/) (2024)
- [Towards Efficient Generative Large Language Model Serving: A Survey from Algorithms to Systems](https://mlsanity.com/paper/2312.15234/) (2023)
- [KVzip: Query-Agnostic KV Cache Compression with Context Reconstruction](https://mlsanity.com/paper/2505.23416/) (2025)
- [A Survey of Resource-efficient LLM and Multimodal Foundation Models](https://mlsanity.com/paper/2401.08092/) (2024)
- [LazyLLM: Dynamic Token Pruning for Efficient Long Context LLM Inference](https://mlsanity.com/paper/2407.14057/) (2024)
- [Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers](https://mlsanity.com/paper/2406.16747/) (2024)
- [Token Reduction Should Go Beyond Efficiency in Generative Models -- From Vision, Language to Multimodality](https://mlsanity.com/paper/2505.18227/) (2025)
- [A Survey on Transformer Compression](https://mlsanity.com/paper/2402.05964/) (2024)
- [Mixture of Attention Spans: Optimizing LLM Inference Efficiency with Heterogeneous Sliding-Window Lengths](https://mlsanity.com/paper/2406.14909/) (2024)
- [Post-Training Sparse Attention with Double Sparsity](https://mlsanity.com/paper/2408.07092/) (2024)
- [NACL: A General and Effective KV Cache Eviction Framework for LLMs at Inference Time](https://mlsanity.com/paper/2408.03675/) (2024)
- [Cognitive Memory in Large Language Models](https://mlsanity.com/paper/2504.02441/) (2025)
- [MrT5: Dynamic Token Merging for Efficient Byte-level Language Models](https://mlsanity.com/paper/2410.20771/) (2024)
- [CITER: Collaborative Inference for Efficient Large Language Model Decoding with Token-Level Routing](https://mlsanity.com/paper/2502.01976/) (2025)
- [Selective Attention Improves Transformer](https://mlsanity.com/paper/2410.02703/) (2024)
- [SqueezeAttention: 2D Management of KV-Cache in LLM Inference via Layer-wise Optimal Budget](https://mlsanity.com/paper/2404.04793/) (2024)
- [HSR-Enhanced Sparse Attention Acceleration](https://mlsanity.com/paper/2410.10165/) (2024)
- [ReAttention: Training-Free Infinite Context with Finite Attention Scope](https://mlsanity.com/paper/2407.15176/) (2024)
- [E2LLM: Encoder Elongated Large Language Models for Long-Context Understanding and Reasoning](https://mlsanity.com/paper/2409.06679/) (2024)
- [In-context KV-Cache Eviction for LLMs via Attention-Gate](https://mlsanity.com/paper/2410.12876/) (2024)
- [FlexiDiT: Your Diffusion Transformer Can Easily Generate High-Quality Samples with Less Compute](https://mlsanity.com/paper/2502.20126/) (2025)
- [Fine-Grained Modeling of Narrative Context: A Coherence Perspective via Retrospective Questions](https://mlsanity.com/paper/2402.13551/) (2024)

## Abstract (arXiv)

Autoregressive Transformers adopted in Large Language Models (LLMs) are hard to scale to long sequences. Despite several works trying to reduce their computational cost, most of LLMs still adopt attention layers between all pairs of tokens in the sequence, thus incurring a quadratic cost. In this study, we present a novel approach that dynamically prunes contextual information while preserving the model's expressiveness, resulting in reduced memory and computational requirements during inference. Our method employs a learnable mechanism that determines which uninformative tokens can be dropped from the context at any point across the generation process. By doing so, our approach not only addresses performance concerns but also enhances interpretability, providing valuable insight into the model's decision-making process. Our technique can be applied to existing pre-trained models through a straightforward fine-tuning process, and the pruning strength can be specified by a sparsity parameter. Notably, our empirical findings demonstrate that we can effectively prune up to 80% of the context without significant performance degradation on downstream tasks, offering a valuable tool for mitigating inference costs. Our reference implementation achieves up to 2× increase in inference throughput and even greater memory savings.