Research graph
References from Task-KV: Task-Aware KV Cache Optimization via Semantic Differentiation of Attention Heads. Local targets link to admitted publications; unresolved targets remain external evidence.
10.18653/v1/2024.emnlp-main.64
10.18653/v1/2024.emnlp-main.64 · External reference
10.1145/3704435
10.1145/3704435 · External reference
10.18653/v1/p19-1102
10.18653/v1/p19-1102 · External reference
10.52202/079017-2733
10.52202/079017-2733 · External reference
Retrieval-augmented generation for large language models: A survey
2023 · External reference
BioRAG: A RAG-LLM framework for biological question reasoning
2024 · External reference
H$_{2}$2O: Heavy-hitter oracle for efficient generative inference of large language models
External reference
10.18653/v1/2024.findings-emnlp.417
10.18653/v1/2024.findings-emnlp.417 · External reference
Efficient streaming language models with attention sinks
2023 · External reference
PyramidKV: Dynamic KV cache compression based on pyramidal information funneling
2024 · External reference
10.18653/v1/2024.findings-acl.195
10.18653/v1/2024.findings-acl.195 · External reference
10.52202/079017-0722
10.52202/079017-0722 · External reference
10.52202/079017-4443
10.52202/079017-4443 · External reference
RazorAttention: Efficient KV cache compression through retrieval heads
2024 · External reference
10.52202/085713-3773
10.52202/085713-3773 · External reference
DuoAttention: Efficient long-context LLM inference with retrieval and streaming heads
2024 · External reference
Not all heads matter: A head-level KV cache compression method with integrated retrieval and reasoning
2024 · External reference
Retrieval head mechanistically explains long-context factuality
2024 · External reference
10.18653/v1/2021.acl-long.152
10.18653/v1/2021.acl-long.152 · External reference
10.18653/v1/2023.wmt-1.45
10.18653/v1/2023.wmt-1.45 · External reference
10.18653/v1/2024.acl-long.172
10.18653/v1/2024.acl-long.172 · External reference
10.18653/v1/2024.acl-long.859
10.18653/v1/2024.acl-long.859 · External reference
10.18653/v1/2023.emnlp-main.298
10.18653/v1/2023.emnlp-main.298 · External reference
Llama 2: Open foundation and fine-tuned chat models
2023 · External reference
Mistral 7B
2023 · External reference
InfLLM: Unveiling the intrinsic capacity of LLMs for understanding extremely long sequences with training-free memory
2024 · External reference
Model tells you what to discard: Adaptive KV cache compression for LLMs
2023 · External reference
10.1016/0098-3004(93)90090-r
10.1016/0098-3004(93)90090-r · External reference
10.48550/arxiv.1706.03762
10.48550/arxiv.1706.03762 · External reference
The Lipschitz constant of self-attention
External reference
LM-infinite: Simple on-the-fly length generalization for large language models
2023 · External reference
10.18653/v1/2021.naacl-main.365
10.18653/v1/2021.naacl-main.365 · External reference
10.18653/v1/d18-1259
10.18653/v1/d18-1259 · External reference
10.18653/v1/2020.coling-main.580
10.18653/v1/2020.coling-main.580 · External reference
10.18653/v1/2021.naacl-main.472
10.18653/v1/2021.naacl-main.472 · External reference
10.18653/v1/2021.naacl-main.112
10.18653/v1/2021.naacl-main.112 · External reference
10.18653/v1/p17-1147
10.18653/v1/p17-1147 · External reference
10.3115/1072228.1072378
10.3115/1072228.1072378 · External reference
10.1007/s10844-012-0211-x
10.1007/s10844-012-0211-x · External reference
RepoBench: Benchmarking repository-level code auto-completion systems
2023 · External reference
LongCoder: A long-range pre-trained language model for code completion
External reference
