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

@inproceedings{neurips2023_s_70085,
  title = {{\$S^3\$: Increasing GPU Utilization during Generative Inference for Higher Throughput}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2023},
  url = {https://nips.cc/virtual/2023/poster/69891},
}

@inproceedings{neurips2024_reducing_71738,
  title = {{Reducing Transformer Key-Value Cache Size with Cross-Layer Attention}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/95548},
}

@inproceedings{iclr2024_model_35498,
  title = {{Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2024},
  url = {https://iclr.cc/virtual/2024/poster/17575},
}

@inproceedings{neurips2024_you_70256,
  title = {{You Only Cache Once: Decoder-Decoder Architectures for Language Models}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/96833},
}

@inproceedings{neurips2025_chunkkv_74864,
  title = {{ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=20JDhbJqn3},
}

@inproceedings{neurips2025_retrievalattention_75515,
  title = {{RetrievalAttention: Accelerating Long-Context LLM Inference via Vector Retrieval}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=8z3cOVER4z},
}

@inproceedings{neurips2025_ada_79854,
  title = {{Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=tcisuhGsQZ},
}

@inproceedings{neurips2025_dkv_76212,
  title = {{dKV-Cache: The Cache for Diffusion Language Models}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=Gppo2JImHs},
}

@inproceedings{iclr2025_duoattention_82736,
  title = {{DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2025},
  url = {https://iclr.cc/virtual/2025/poster/29062},
}

@inproceedings{icml2024_dynamic_45578,
  title = {{Dynamic Memory Compression: Retrofitting LLMs for Accelerated Inference}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2024},
  url = {https://icml.cc/virtual/2024/poster/32874},
}

@inproceedings{icml2025_shadowkv_48568,
  title = {{ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/44053},
}

@inproceedings{neurips2024_block_70622,
  title = {{Block Transformer: Global-to-Local Language Modeling for Fast Inference}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2024},
  url = {https://neurips.cc/virtual/2024/poster/96498},
}

@inproceedings{neurips2025_tensor_75975,
  title = {{Tensor Product Attention Is All You Need}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=ECTxVRFhUa},
}

@inproceedings{neurips2025_prefixkv_79817,
  title = {{PrefixKV: Adaptive Prefix KV Cache is What Vision Instruction-Following Models Need for Efficient Generation}},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  url = {https://openreview.net/forum?id=tDG6bY48ch},
}

@inproceedings{iclr2025_squeezeattention_81061,
  title = {{SqueezeAttention: 2D Management of KV-Cache in LLM Inference via Layer-wise Optimal Budget}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2025},
  url = {https://iclr.cc/virtual/2025/poster/30707},
}

@inproceedings{iclr2025_scbench_83001,
  title = {{SCBench: A KV Cache-Centric Analysis of Long-Context Methods}},
  booktitle = {International Conference on Learning Representations (ICLR)},
  year = {2025},
  url = {https://iclr.cc/virtual/2025/poster/28803},
}

@inproceedings{aaai2026_head_107334,
  title = {{Head-Aware KV Cache Compression for Efficient Visual Autoregressive Modeling}},
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2026},
  url = {https://ojs.aaai.org/index.php/AAAI/article/view/39686},
}

@inproceedings{icml2024_lococo_44250,
  title = {{LoCoCo: Dropping In Convolutions for Long Context Compression}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2024},
  url = {https://icml.cc/virtual/2024/poster/34202},
}

@inproceedings{icml2025_quantspec_46234,
  title = {{QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/46326},
}

@inproceedings{icml2025_specache_47201,
  title = {{SpeCache: Speculative Key-Value Caching for Efficient Generation of LLMs}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/45386},
}

@inproceedings{icml2025_rocketkv_47339,
  title = {{RocketKV: Accelerating Long-Context LLM Inference via Two-Stage KV Cache Compression}},
  booktitle = {International Conference on Machine Learning (ICML)},
  year = {2025},
  url = {https://icml.cc/virtual/2025/poster/45253},
}