MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models | Paper Citation Record

Observation 4c9fa41f-ceba-4ddf-bf19-fd591d23e4c7 · outbound

This paper cites GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:19.894867Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 610b6963-6a14-426d-9b6a-0b323fa6235a · outbound

This paper cites Nikolopoulos, Hans Vandierendonck, Deepu John, and Bo Ji.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Nikolopoulos, Hans Vandierendonck, Deepu John, and Bo Ji

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:20.054978Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d5560da7-6839-4c85-a888-927562fbf26b · outbound

This paper cites Qwen2.5-VL Technical Report.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Qwen2.5-VL Technical Report

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:20.222339Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1403ce93-0877-46f5-8cd4-6d4cd083be51 · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:20.585480Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1ab748ed-05c8-4f8d-8d04-4e7a709034e8 · outbound

This paper cites FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:20.913521Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4ed5212b-863e-4cd8-b443-04f25aa6187c · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:21.026569Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 04a2b01f-4983-4b17-80b9-b58b4dbf6f4c · outbound

This paper cites See the Forest for the Trees: Loosely Speculative Decoding via Visual-Semantic Guidance for Efficient Inference of Video LLMs.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models See the Forest for the Trees: Loosely Speculative Decoding via Visual-Semantic Guidance for Efficient Inference of Video LLMs

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:21.450728Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ad466646-dfd4-4c2e-bfbf-33543f930347 · outbound

This paper cites Abdi, Dongsheng Li, Chin-Yew Lin, Yuqing Yang, and Lili Qiu.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Abdi, Dongsheng Li, Chin-Yew Lin, Yuqing Yang, and Lili Qiu

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:21.569466Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ea860bbb-30e8-454f-b31c-de52ed70f3d1 · outbound

This paper cites LLaVA-OneVision: Easy Visual Task Transfer.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models LLaVA-OneVision: Easy Visual Task Transfer

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:21.770610Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3df26648-73ab-4607-b379-44f33c18f53f · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:21.891492Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 672026b0-2428-4a96-852a-a19364078c1e · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:22.063778Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 40a1932f-c375-43cf-892d-0df718559623 · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:22.181795Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 713147ac-3da5-4830-9770-8dfa6e57943e · outbound

This paper cites Training-Free Activation Sparsity in Large Language Models.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Training-Free Activation Sparsity in Large Language Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:22.293241Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b2be34b0-341b-4c68-b7b5-37c4b0c39e49 · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:22.698273Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7fe856aa-0215-44ef-9a08-e0b5b3992028 · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:22.839506Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 50b5f7a1-d2c3-4162-a318-e3c84d2dc512 · outbound

This paper cites Morse, Raghavv Goel, Mingu Lee, and Chris Lott.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Morse, Raghavv Goel, Mingu Lee, and Chris Lott

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:22.959713Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b00a8e43-9164-48e3-8c1a-d9f65f3bad6d · outbound

This paper cites ANLS* -- A Universal Document Processing Metric for Generative Large Language Models.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models ANLS* -- A Universal Document Processing Metric for Generative Large Language Models

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:23.108518Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9c0e047-deeb-4640-adc3-d986e28eeb9e · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:23.423299Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b76d9ec2-e6a8-4c75-bb4b-7485e2118a70 · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:23.571629Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 30e33678-916c-476e-9e8a-73626e926fc7 · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:23.810534Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20c69d2f-748e-4b1f-8a59-577e5f5e0335 · outbound

This paper cites SparseMM: Head Sparsity Emerges from Visual Concept Responses in MLLMs.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models SparseMM: Head Sparsity Emerges from Visual Concept Responses in MLLMs

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:24.192980Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d24cf86a-d86d-4779-af46-ecac5407352a · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:24.335787Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f2bd8cea-7ddb-4931-ba17-f052837b67fb · outbound

This paper cites Efficient Streaming Language Models with Attention Sinks.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Efficient Streaming Language Models with Attention Sinks

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:24.448018Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3459f733-da4e-49d3-b874-c26b79a2e42f · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:24.611574Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7c186425-9c29-498c-8f4c-603e8ba516ab · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:24.737749Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 90b9b16c-7779-4e73-94e4-1064694c1459 · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:24.851839Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5dda8ded-2289-44d8-a403-4508c5cb6fa6 · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:24.963171Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f811e027-85bb-4c6a-be33-3662a4401d3c · outbound

This paper cites HybridKV: Hybrid KV Cache Compression for Efficient Multimodal Large Language Model Inference.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models HybridKV: Hybrid KV Cache Compression for Efficient Multimodal Large Language Model Inference

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:25.085915Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 82653632-2db3-4d9f-a680-d6a8e1c94ad2 · outbound

This paper cites Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:25.244892Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 36df42ca-8cab-4db5-bc92-6a6657fc4cb7 · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:25.477725Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6922c32b-0f07-417d-a96a-34fa06eff11b · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:25.572692Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fc4a7285-a518-4125-97db-8435ee05097a · outbound

This paper cites an unresolved cited work.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Unresolved cited work

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:25.702041Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5d211f02-76b7-4335-b5a0-4b146a7d472a · outbound

This paper cites 2025 , eprint =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2025 , eprint =

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:25.791801Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4e6a33a1-3ea4-49e2-9cc8-d27a4e75c50e · outbound

This paper cites 2025 , eprint =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2025 , eprint =

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:25.922021Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a6d8d12c-f06b-40f5-bcf6-2c953812cad8 · outbound

This paper cites Advances in Neural Information Processing Systems , year =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Advances in Neural Information Processing Systems , year =

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:26.082011Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f830f334-e753-4852-be3c-3b67bbe3877e · outbound

This paper cites 2025 , eprint =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2025 , eprint =

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:26.168724Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 246fab55-4d51-422b-b25b-f32212782cfa · outbound

This paper cites 2023 , eprint =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2023 , eprint =

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:26.270260Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f0cb8b69-e774-4754-89e0-325ae7162404 · outbound

This paper cites 2023 , eprint =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2023 , eprint =

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:26.385348Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ad1ef3bc-18b5-42d1-98d5-ea8223b8c366 · outbound

This paper cites 2025 , eprint =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2025 , eprint =

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:26.512769Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 973b3d56-208c-4b95-ae63-42e351e05ff4 · outbound

This paper cites 2024 , eprint =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2024 , eprint =

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:26.635446Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ba523d12-1836-4ff9-8495-eec731bfa467 · outbound

This paper cites 2024 , eprint =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2024 , eprint =

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:26.768096Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cfb5dd04-0ff9-4297-ad0c-19a70376f45d · outbound

This paper cites OCRBench: On the Hidden Mystery of OCR in Large Multimodal Models.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models OCRBench: On the Hidden Mystery of OCR in Large Multimodal Models

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:26.902599Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 70e8483f-bc4e-4b6e-afd4-007a2ece93d4 · outbound

This paper cites TextCaps : A Dataset for Image Captioning with Reading Comprehension.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models TextCaps : A Dataset for Image Captioning with Reading Comprehension

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:27.039296Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1029a303-26d1-49e1-a0cf-5863cb5e4817 · outbound

This paper cites MMMU : A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models MMMU : A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:27.150104Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 956b0500-334f-4eb6-995b-dd6c947af03b · outbound

This paper cites Findings of the Association for Computational Linguistics:. ChartQA : A Benchmark for Question Answering about Charts with Visual and Logical Reasoning. 2022 , address =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Findings of the Association for Computational Linguistics:. ChartQA : A Benchmark for Question Answering about Charts with Visual and Logical Reasoning. 2022 , address =

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:27.289634Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 636132c9-4b37-4837-8a89-f7b94ec292b3 · outbound

This paper cites Towards VQA Models That Can Read.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Towards VQA Models That Can Read

Reference 60

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:27.422148Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb64e5c9-fa21-4552-bf88-666ffc84c5f9 · outbound

This paper cites , booktitle = "Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models , booktitle = "Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (

Reference 61

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:27.549270Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8a43d895-8e4e-49a8-a5fd-6d3a8be1cee5 · outbound

This paper cites KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache

Reference 62

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:27.683696Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d18f3edf-a0dc-4c1a-bff2-65ae682dd6fb · outbound

This paper cites AWQ : Activation-aware Weight Quantization for On-Device LLM Compression and Acceleration.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models AWQ : Activation-aware Weight Quantization for On-Device LLM Compression and Acceleration

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:27.842372Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1a51ac52-0af4-4c70-a11a-e24e0bd6f9c3 · outbound

This paper cites Computer Vision --. An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models. 2025 , publisher =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Computer Vision --. An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models. 2025 , publisher =

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:27.962962Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 618513c7-b8a9-49ed-a6c3-48315be267e7 · outbound

This paper cites Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:28.105753Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 94e7f057-f949-40b9-897e-1162f22cc9e3 · outbound

This paper cites Association for Computing Machinery.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Association for Computing Machinery

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:28.185604Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9027bd55-13a7-4f06-9b86-e558141dad8e · outbound

This paper cites InfLLM : Training-free Long-context Extrapolation for LLMs with an Efficient Context Memory.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models InfLLM : Training-free Long-context Extrapolation for LLMs with an Efficient Context Memory

Reference 67

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:28.334271Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ae356680-5cb6-4441-89e3-60a5f58b5277 · outbound

This paper cites H2O : Heavy-hitter Oracle for Efficient Generative Inference of Large Language Models.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models H2O : Heavy-hitter Oracle for Efficient Generative Inference of Large Language Models

Reference 68

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:28.466982Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 15fbe337-d973-4b09-b47e-e986514b5c09 · outbound

This paper cites RotateKV: Accurate and Robust 2-Bit KV Cache Quantization for LLMs via Outlier-Aware Adaptive Rotations.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models RotateKV: Accurate and Robust 2-Bit KV Cache Quantization for LLMs via Outlier-Aware Adaptive Rotations

Reference 69

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:28.560458Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3a451ae3-ee0e-4995-b1fe-ff143af95e43 · outbound

This paper cites Generating Long Sequences with Sparse Transformers.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Generating Long Sequences with Sparse Transformers

Reference 70

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:28.674400Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1429e6f6-af11-466b-b455-4c7253c6d26b · outbound

This paper cites Training-free and Adaptive Sparse Attention for Efficient Long Video Generation.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Training-free and Adaptive Sparse Attention for Efficient Long Video Generation

Reference 71

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:28.800958Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5d9ccd38-9620-42e3-8271-7f82ef51b2a7 · outbound

This paper cites A Simple and Effective $L_2$ Norm-Based Strategy for KV Cache Compression.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models A Simple and Effective $L_2$ Norm-Based Strategy for KV Cache Compression

Reference 72

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:28.898360Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8ff85135-0005-4cd1-a816-5f5b2afa60ba · outbound

This paper cites , journal =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models , journal =

Reference 73

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:28.992900Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 01fccef6-38ff-457f-ba8f-bfd7f0bb4477 · outbound

This paper cites PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:29.142425Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f69adfc5-62c1-4796-8bab-c8adb5afc3a1 · outbound

This paper cites VisionZip : Longer is Better but Not Necessary in Vision-Language Models.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models VisionZip : Longer is Better but Not Necessary in Vision-Language Models

Reference 75

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:29.256668Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bac81452-c74b-4091-9a4b-3ade95250148 · outbound

This paper cites DyCoke : Dynamic Compression of Tokens for Fast Video Large Language Models.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models DyCoke : Dynamic Compression of Tokens for Fast Video Large Language Models

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:29.361640Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da5de966-860a-4fa6-8ea5-f2d30ba177a6 · outbound

This paper cites LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding

Reference 77

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:29.497400Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eedee17c-53af-4b77-81d6-c8aafc2a1471 · outbound

This paper cites TopV : Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision-Language Model.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models TopV : Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision-Language Model

Reference 78

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:29.609514Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f4c52329-4ebd-4876-957d-ec47b3ddc319 · outbound

This paper cites LOOK-M: Look-Once Optimization in KV Cache for Efficient Multimodal Long-Context Inference.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models LOOK-M: Look-Once Optimization in KV Cache for Efficient Multimodal Long-Context Inference

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:29.746228Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 95ed4cdf-9601-4f9c-a19a-5b5fe7b61733 · outbound

This paper cites Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

Reference 80

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:29.845572Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5348cdde-7e76-4659-8e7e-6d942dd84429 · outbound

This paper cites DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:29.935195Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0e0d93f9-2ca8-4c7a-8d0c-75e38cda5bc0 · outbound

This paper cites Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference

Reference 82

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:30.056509Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9ddf2ab0-fcee-4955-aa19-8f1645883cd3 · outbound

This paper cites and Li, Dongsheng and Lin, Chin-Yew and Yang, Yuqing and Qiu, Lili , booktitle = "Advances in Neural Information Processing Systems (.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models and Li, Dongsheng and Lin, Chin-Yew and Yang, Yuqing and Qiu, Lili , booktitle = "Advances in Neural Information Processing Systems (

Reference 83

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:30.166806Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c6446cc8-25ce-48e3-b393-e3363a4e9e7e · outbound

This paper cites SparseVLM : Visual Token Sparsification for Efficient Vision-Language Models Inference.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models SparseVLM : Visual Token Sparsification for Efficient Vision-Language Models Inference

Reference 84

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:30.314903Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cf8bd247-429e-4259-a8fa-97fc5e1019eb · outbound

This paper cites LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding

Reference 85

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:30.462516Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7840632-cd37-4724-a617-bf44c1d6517b · outbound

This paper cites GQA : Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models GQA : Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints

Reference 86

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:30.589649Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7d7e243-73ff-4194-a052-a263e3d8d0a1 · outbound

This paper cites Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints

Reference 87

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:30.699815Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4b9599c-978b-4803-9284-d698d4c5d12c · outbound

This paper cites Sparse VideoGen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Sparse VideoGen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity

Reference 88

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:30.811138Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 41e47fe2-a1c4-48a4-b890-0c33ada6060d · outbound

This paper cites Proceedings of the Thirty-Ninth AAAI Conference on Artificial Intelligence.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Proceedings of the Thirty-Ninth AAAI Conference on Artificial Intelligence

Reference 89

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:30.936965Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4ae0194c-339d-47ff-bdd1-b83036f7ba99 · outbound

This paper cites Multi-Layer Visual Feature Fusion in Multimodal LLMs : Methods, Analysis, and Best Practices.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Multi-Layer Visual Feature Fusion in Multimodal LLMs : Methods, Analysis, and Best Practices

Reference 90

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:31.077641Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 40bbf5e9-7530-4ceb-950b-f498f214554d · outbound

This paper cites LMMs-Eval: Reality Check on the Evaluation of Large Multimodal Models.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models LMMs-Eval: Reality Check on the Evaluation of Large Multimodal Models

Reference 91

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:31.210800Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c389fac6-4e4d-4d9e-944c-cea45b386618 · outbound

This paper cites Exact Matching: Algorithms and Related Problems.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Exact Matching: Algorithms and Related Problems

Reference 92

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:31.358219Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7eee086-72ee-43e4-b1d7-6ec22f130b18 · outbound

This paper cites CIDEr: Consensus-based Image Description Evaluation.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models CIDEr: Consensus-based Image Description Evaluation

Reference 93

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:31.477570Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 71d68937-354d-4e03-811b-8792b63dca2a · outbound

This paper cites ANLS* -- A Universal Document Processing Metric for Generative Large Language Models.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models ANLS* -- A Universal Document Processing Metric for Generative Large Language Models

Reference 94

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:31.594074Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6631fa7b-dd24-4a0c-9655-bf93d45333d1 · outbound

This paper cites 2026 , eprint=.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2026 , eprint=

Reference 95

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:31.690860Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bcc1a2f5-baa6-4f6d-bce4-865a6f4ac7a9 · outbound

This paper cites 2026 , eprint=.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2026 , eprint=

Reference 96

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:31.854901Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation adebac80-f026-4807-85a5-c17826c9a6a9 · outbound

This paper cites 2026 , eprint=.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models 2026 , eprint=

Reference 97

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:31.995756Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eade1ae3-54f1-4bb1-b807-0b5d172e1d6e · outbound

This paper cites Proceedings of the 42nd International Conference on Machine Learning , series =.

MM-ShiftKV: Decode-Aware Prefill-Stage KV Selection for Multimodal Large Language Models Proceedings of the 42nd International Conference on Machine Learning , series =

Reference 98

Resolution: unresolved; no resolver link, observed 2026-08-02T11:58:32.095426Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers