MOIRA: Mass-Oriented Indexing with Ragged Attention for Long-Context Decoding

cs.AI

Submitted: 2026-10-03

Updated: 2026-10-03

Comments: 19 pages, 7 figures, 7 tables

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Sources

• SpotAttention: Plug-In Block-Sparse Routing for Pretrained Long-Context Transformers
• LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding
• RetroInfer: A Vector Storage Engine for Scalable Long-Context LLM Inference
• MagicPIG: LSH Sampling for Efficient LLM Generation
• FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning
• FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
• DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
• DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
• vAttention: Verified Sparse Attention
• Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference
• Not All Heads Matter: A Head-Level KV Cache Compression Method with Integrated Retrieval and Reasoning
• SeerAttention-R: Sparse Attention Adaptation for Long Reasoning
• The Llama 3 Herd of Models
• Elastic Threshold Attention: Learned Contextual Sparsity for Long-Context Decoding
• FlashDecoding++: Faster Large Language Model Inference on GPUs
• KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization
• RULER: What's the Real Context Size of Your Long-Context Language Models?
• CompKV: Compensation-Aware KV Selection for Long-Context LLM Inference
• Efficient Memory Management for Large Language Model Serving with PagedAttention
• MiniMax Sparse Attention

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems