🪟 Context Windows - bloknayrb · Scour

Tangram: Unlocking Non-Uniform KV Cache for Efficient Multi-turn LLM Serving

🤖LLM Academic

arxiv.org··Hacker News

SparDA: Sparse Decoupled Attention for Efficient Long-Context LLM Inference

🤖LLM Academic

You Only Index Once: Cross-Layer Sparse Attention with Shared Routing

✍️Prompt Engineering Academic

Signed Dual Attention: Capturing Signed Dependencies in Time Series Forecasting

📝NLP Academic

Rethinking LoRA Memory Through the Lens of KV Cache Compression

🦙Ollama Academic

Cartridges at Scale: Training Modular KV Caches over Large Document Collections

💬LLMs Academic

Where does Absolute Position come from in decoder-only Transformers?

🛡️AI Security Academic

GRAMformer: Any-Order Modality Interactions via Volumetric Multimodal Cross-Attention

💬LLMs Academic

Transformer-Enhanced Reinforcement Learning: Fundamentals and Applications in Communication Networks

💬LLMs Academic

Selective Coupling of Decoupled Informative Regions: Masked Attention Alignment for Data-Free Quantization of Vision Transformers

💬LLMs Academic

Do Transformers Need Three Projections? Systematic Study of QKV Variants

📱Edge AI Academic

arxiv.org··Hacker News

Dense Contexts Are Hard Contexts: Lexical Density Limits Effective Context in LLMs

🤖LLM Academic

arxiv.org··Hacker News

Log in to enable infinite scrolling