conftrace_

kv cache

74 papers

Explore in graph

Also known as

KVCACHE KV CACHE KV

Co-occurring keywords

large language model (13587) model compression (3302) inference optimization (219) inference efficiency (248) inference acceleration (244) memory efficiency (261) attention mechanism (4001) long context (121) memory optimization (114) key-value cache (55)

Papers

TailorKV: A Hybrid Framework for Long-Context Inference via Tailored KV Cache Optimization ACL 2025
SCOPE: Optimizing Key-Value Cache Compression in Long-context Generation ACL 2025
MiniKV: Pushing the Limits of 2-Bit KV Cache via Compression and System Co-Design for Efficient Long Context Inference ACL 2025
UNComp: Can Matrix Entropy Uncover Sparsity? — A Compressor Design from an Uncertainty-Aware Perspective EMNLP 2025
DynamicKV: Task-Aware Adaptive KV Cache Compression for Long Context LLMs EMNLP 2025
Align Attention Heads Before Merging Them: An Effective Way for Converting MHA to GQA EMNLP 2025
Source-primed Multi-turn Conversation Helps Large Language Models Translate Documents EMNLP 2025
IAM: Efficient Inference through Attention Mapping between Different-scale LLMs ACL 2025
MadaKV: Adaptive Modality-Perception KV Cache Eviction for Efficient Multimodal Long-Context Inference ACL 2025
MobiLoRA: Accelerating LoRA-based LLM Inference on Mobile Devices via Context-aware KV Cache Optimization ACL 2025
Towards Economical Inference: Enabling DeepSeek’s Multi-Head Latent Attention in Any Transformer-based LLMs ACL 2025
FIER: Fine-Grained and Efficient KV Cache Retrieval for Long-context LLM Inference EMNLP 2025
Unifying Mixture of Experts and Multi-Head Latent Attention for Efficient Language Models EMNLP 2025
SwiftKV: Fast Prefill-Optimized Inference with Knowledge-Preserving Model Transformation EMNLP 2025
Multi-matrix Factorization Attention ACL 2025
RefreshKV: Updating Small KV Cache During Long-form Generation ACL 2025
SpindleKV: A Novel KV Cache Reduction Method Balancing Both Shallow and Deep Layers ACL 2025
Smarter, Not Harder: Training-Free Adaptive Computation for Transformers ACL 2025
MEDA: Dynamic KV Cache Allocation for Efficient Multimodal Long-Context Inference NAACL 2025
XQuant: Achieving Ultra-Low Bit KV Cache Quantization with Cross-Layer Compression EMNLP 2025
Efficient Beam Search for Large Language Models Using Trie-Based Decoding EMNLP 2025
TokenSelect: Efficient Long-Context Inference and Length Extrapolation for LLMs via Dynamic Token-Level KV Cache Selection EMNLP 2025
LeanK: Learnable K Cache Channel Pruning for Efficient Decoding EMNLP 2025
Lookahead Q-Cache: Achieving More Consistent KV Cache Eviction via Pseudo Query EMNLP 2025
KV-Edit: Training-Free Image Editing for Precise Background Preservation ICCV 2025