# Shallow Prefill, Deep Decoding: Efficient Long-Context Inference via Layer-Asymmetric KV Visibility

arXiv: https://arxiv.org/abs/2605.06105 · HTML: https://mlsanity.com/paper/2605.06105/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 36 of its 37 references that are in the map.
- First posted on arXiv 7 May 2026.

## Builds on (most cited first, 30 of 36)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [Tulu 3: Pushing Frontiers in Open Language Model Post-Training](https://mlsanity.com/paper/2411.15124/) (2024)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [Splitwise: Efficient generative LLM inference using phase splitting](https://mlsanity.com/paper/2311.18677/) (2023)
- [Efficiently Scaling Transformer Inference](https://mlsanity.com/paper/2211.05102/) (2022)
- [Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference](https://mlsanity.com/paper/2406.10774/) (2024)
- [Reducing Transformer Depth on Demand with Structured Dropout](https://mlsanity.com/paper/1909.11556/) (2019)
- [PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling](https://mlsanity.com/paper/2406.02069/) (2024)
- [Learning to Compress Prompts with Gist Tokens](https://mlsanity.com/paper/2304.08467/) (2023)
- [MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention](https://mlsanity.com/paper/2407.02490/) (2024)
- [Adapting Language Models to Compress Contexts](https://mlsanity.com/paper/2305.14788/) (2023)
- [Confident Adaptive Language Modeling](https://mlsanity.com/paper/2207.07061/) (2022)
- [LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding](https://mlsanity.com/paper/2404.16710/) (2024)
- [DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads](https://mlsanity.com/paper/2410.10819/) (2024)
- [In-context Autoencoder for Context Compression in a Large Language Model](https://mlsanity.com/paper/2307.06945/) (2023)
- [You Only Cache Once: Decoder-Decoder Architectures for Language Models](https://mlsanity.com/paper/2405.05254/) (2024)
- [Reducing Transformer Key-Value Cache Size with Cross-Layer Attention](https://mlsanity.com/paper/2405.12981/) (2024)
- [MiniCache: KV Cache Compression in Depth Dimension for Large Language Models](https://mlsanity.com/paper/2405.14366/) (2024)
- [OLMES: A Standard for Language Model Evaluations](https://mlsanity.com/paper/2406.08446/) (2024)
- [vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention](https://mlsanity.com/paper/2405.04437/) (2024)
- [What Matters in Transformers? Not All Attention is Needed](https://mlsanity.com/paper/2406.15786/) (2024)
- [Layer-Condensed KV Cache for Efficient Inference of Large Language Models](https://mlsanity.com/paper/2405.10637/) (2024)
- [Large language models implicitly learn to straighten neural sentence trajectories to construct a predictive representation of natural language](https://mlsanity.com/paper/2311.04930/) (2023)
- [Attend First, Consolidate Later: On the Importance of Attention in Different LLM Layers](https://mlsanity.com/paper/2409.03621/) (2024)

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

- [ReCache: Efficient KV Cache Reuse and Compression for Tool-Augmented LLM Agents](https://mlsanity.com/paper/2608.19662/) (2026)

## Abstract (arXiv)

Long-context inference in decoder-only language models is costly because long prompts are processed during Prefill, cached at every layer, and repeatedly attended to during autoregressive Decode. We introduce Shallow Prefill, dEEp Decode (SPEED), a phase-asymmetric KV-visibility policy that materializes non-anchor prompt-token KV states only in lower layers while keeping Decode-phase tokens full-depth. Unlike previous approaches that make upper-layer prompt KV states cheaper to store or construct, SPEED removes prefill tokens from the upper-layer Decode visibility set altogether. With a minimal BoS anchor, this simple change preserves broad benchmark quality while reducing long-context cost. In a controlled Llama-3.1-8B instruction-tuning study, SPEED using only 75% of layers for prefill tokens reaches 51.2 average score on OLMES-style benchmarks, compared with 51.4 for the full-depth baseline, while improving TTFT by 33%, TPOT by 22%, and reducing active KV memory by 25.0% at 128K context. Layer-wise diagnostics suggest that this cutoff retains the main prompt-selection and representation-stabilization regions of the full-depth model. These results show that long-context prompt tokens need not always persist as full-depth KV-cache objects when Decode-phase tokens remain full-depth.