% Key papers for KV cache compression (scime atlas, https://atlas.scime.ai). Authors are not included: add them from the paper pages.

@inproceedings{cvpr2024_artrackv2_22689,
  title = {{ARTrackV2: Prompting Autoregressive Tracker Where to Look and How to Describe}},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2024},
  url = {https://cvpr.thecvf.com/virtual/2024/poster/31769},
}

@inproceedings{neurips2024_snapkv_73986,
  title = {{SnapKV: LLM Knows What You are Looking for Before Generation}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/93531},
}

@inproceedings{neurips2025_kvlink_79329,
  title = {{KVLink: Accelerating Large Language Models via Efficient KV Cache Reuse}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=oDcAGSXZZP},
}

@inproceedings{icml2025_epic_48696,
  title = {{EPIC: Efficient Position-Independent Caching for Serving Large Language Models}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/43926},
}

@inproceedings{icml2024_d_43692,
  title = {{DéjàVu: KV-cache Streaming for Fast, Fault-tolerant Generative LLM Serving}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2024},
  url = {https://icml.cc/virtual/2024/poster/34760},
}

@inproceedings{icml2025_compute_47576,
  title = {{Compute or Load KV Cache? Why Not Both?}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/45020},
}

@inproceedings{neurips2025_r_74942,
  title = {{R-KV: Redundancy-aware KV Cache Compression for Reasoning Models}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=2jwAjomEDB},
}

@inproceedings{neurips2025_tail_75445,
  title = {{Tail-Optimized Caching for LLM Inference}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=8C8F4NmHfz},
}

@inproceedings{neurips2025_learned_77613,
  title = {{Learned Prefix Caching for Efficient LLM Inference}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=Vj48eXaQDM},
}

@inproceedings{icml2024_kv_44277,
  title = {{KV-Runahead: Scalable Causal LLM Inference by Parallel Key-Value Cache Generation}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2024},
  url = {https://icml.cc/virtual/2024/poster/34175},
}

@inproceedings{icml2025_lacache_47356,
  title = {{LaCache: Ladder-Shaped KV Caching for Efficient Long-Context Modeling of Large Language Models}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/45236},
}

@inproceedings{icml2026_triattention_49266,
  title = {{TriAttention: Efficient Long Reasoning with Trigonometric KV Compression}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=0tgzJK50Jz},
}

@inproceedings{icml2026_semantic_49350,
  title = {{Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=1fGD4sKs8j},
}

@inproceedings{icml2026_relaycaching_49368,
  title = {{RelayCaching: Accelerating LLM Collaboration via Decoding KV Cache Reuse}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=1tbhBSXcyX},
}

@inproceedings{icml2026_mvr_52455,
  title = {{MVR-cache: Optimizing Semantic Caching via Multi-Vector Retrieval and Learned Prompt Segmentation}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=WALBUqiamf},
}

@inproceedings{icml2026_arborkv_52461,
  title = {{ArborKV: Structure-Aware KV Cache Management for Scaling Tree-based LLM Reasoning}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=WIk0aJfcbK},
}

@inproceedings{icml2026_beaconkv_53058,
  title = {{BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=bgAbffQCrL},
}

@inproceedings{icml2026_which_53400,
  title = {{Which Heads Matter for Reasoning? RL-Guided KV Cache Compression}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=fBRIgbTvId},
}

@inproceedings{icml2026_redvisor_53642,
  title = {{RedVisor: Reasoning-Aware Prompt Injection Defense via Zero-Copy KV Cache Reuse}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=hLLdM61qen},
}

@inproceedings{icml2026_foresightkv_55510,
  title = {{ForesightKV: Optimizing KV Cache Eviction for Reasoning Models by Learning Long-Term Contribution}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2026},
  url = {https://openreview.net/forum?id=znV8JHv8b8},
}