# LookaheadKV: Fast and Accurate KV Cache Eviction by Glimpsing into the Future without Generation

arXiv: https://arxiv.org/abs/2603.10899 · HTML: https://mlsanity.com/paper/2603.10899/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Placed in KV cache optimization, one of 12 directions in Efficient LLM inference and serving.
- Builds on 35 of its 39 references that are in the map.
- First posted on arXiv 11 Mar 2026.

## Placed in

- Efficient LLM inference and serving / [KV cache optimization](https://mlsanity.com/inference/kv-cache-optimization/)

## Builds on (most cited first, 30 of 35)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers](https://mlsanity.com/paper/2002.10957/) (2020)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024)
- [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)
- [A Dataset of Information-Seeking Questions and Answers Anchored in Research Papers](https://mlsanity.com/paper/2105.03011/) (2021)
- [Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time](https://mlsanity.com/paper/2305.17118/) (2023)
- [PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling](https://mlsanity.com/paper/2406.02069/) (2024)
- [The Stack: 3 TB of permissively licensed source code](https://mlsanity.com/paper/2211.15533/) (2022)
- [RepoBench: Benchmarking Repository-Level Code Auto-Completion Systems](https://mlsanity.com/paper/2306.03091/) (2023)
- [Long-context LLMs Struggle with Long In-context Learning](https://mlsanity.com/paper/2404.02060/) (2024)
- [Smaug: Fixing Failure Modes of Preference Optimisation with DPO-Positive](https://mlsanity.com/paper/2402.13228/) (2024)
- [Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference](https://mlsanity.com/paper/2407.11550/) (2024)
- [Distilling Knowledge from Reader to Retriever for Question Answering](https://mlsanity.com/paper/2012.04584/) (2020)
- [Many-Shot In-Context Learning](https://mlsanity.com/paper/2404.11018/) (2024)
- [Leave No Document Behind: Benchmarking Long-Context LLMs with Extended Multi-Doc QA](https://mlsanity.com/paper/2406.17419/) (2024)
- [Transformers are Multi-State RNNs](https://mlsanity.com/paper/2401.06104/) (2024)
- [KVzip: Query-Agnostic KV Cache Compression with Context Reconstruction](https://mlsanity.com/paper/2505.23416/) (2025)
- [Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters](https://mlsanity.com/paper/2406.12335/) (2024)
- [RepoAgent: An LLM-Powered Open-Source Framework for Repository-level Code Documentation Generation](https://mlsanity.com/paper/2402.16667/) (2024)
- [ChatQA 2: Bridging the Gap to Proprietary LLMs in Long Context and RAG Capabilities](https://mlsanity.com/paper/2407.14482/) (2024)
- [KeyDiff: Key Similarity-Based KV Cache Eviction for Long-Context LLM Inference in Resource-Constrained Environments](https://mlsanity.com/paper/2504.15364/) (2025)
- [Your LLM Knows the Future: Uncovering Its Multi-Token Prediction Potential](https://mlsanity.com/paper/2507.11851/) (2025)

## Built on by (most cited first, 10 of 10)

- [IntentKV: Cross-Turn Intent-Aware KV Cache Pruning for Agent Inference](https://mlsanity.com/paper/2606.09916/) (2026)
- [KVEraser: Learning to Steer KV Cache for Efficient Localized Context Erasing](https://mlsanity.com/paper/2606.17034/) (2026)
- [When Does Value-Aware KV Eviction Help? A Fixed-Contract Diagnostic for Non-Monotone Cache Compression](https://mlsanity.com/paper/2605.08234/) (2026)
- [RestoreKV: Recovering Full-Cache Behavior Under Aggressive Query-Agnostic KV Cache Eviction](https://mlsanity.com/paper/2608.01247/) (2026)
- [Monte Carlo Estimation for KV Cache Eviction](https://mlsanity.com/paper/2610.07643/) (2026)
- [Discovering KV Cache Eviction Policies via LLM-Guided Program Evolution](https://mlsanity.com/paper/2608.14555/) (2026)
- [WnW: Waxing-and-Waning KV Cache for Long-Form Speech LLMs](https://mlsanity.com/paper/2608.22704/) (2026)
- [Divergence Timing and Cumulative Disagreement under KV-Cache Eviction](https://mlsanity.com/paper/2609.16617/) (2026)
- [When to Evict, Not What to Keep: Draft-Guided Eviction for Training-Free KV-Cache Compression](https://mlsanity.com/paper/2609.33334/) (2026)
- [Mask-Guided KV Cache Eviction in Block Diffusion Language Models](https://mlsanity.com/paper/2610.06996/) (2026)

## Abstract (arXiv)

Transformer-based large language models (LLMs) rely on key-value (KV) caching to avoid redundant computation during autoregressive inference. While this mechanism greatly improves efficiency, the cache size grows linearly with the input sequence length, quickly becoming a bottleneck for long-context tasks. Existing solutions mitigate this problem by evicting prompt KV that are deemed unimportant, guided by estimated importance scores. Notably, a recent line of work proposes to improve eviction quality by "glimpsing into the future", in which a draft generator produces a surrogate future response approximating the target model's true response, and this surrogate is subsequently used to estimate the importance of cached KV more accurately. However, these approaches rely on computationally expensive draft generation, which introduces substantial prefilling overhead and limits their practicality in real-world deployment. To address this challenge, we propose LookaheadKV, a lightweight eviction framework that leverages the strength of surrogate future response without requiring explicit draft generation. LookaheadKV augments transformer layers with parameter-efficient modules trained to predict true importance scores with high accuracy. Our design ensures negligible runtime overhead comparable to existing inexpensive heuristics, while achieving accuracy superior to more costly approximation methods. Extensive experiments on long-context understanding benchmarks, across a wide range of models, demonstrate that our method not only outperforms recent competitive baselines in various long-context understanding tasks, but also reduces the eviction cost by up to 14.5x, leading to significantly faster time-to-first-token. Our code is available at https://github.com/SamsungLabs/LookaheadKV.