BasisKV: Value Latents for Communication and Key Routing
cs.AI
Submitted: 2026-11-02
Updated: 2026-11-02
Code: https://github.com/alexz949/BasisKV
Terminology
Sources
- STAR-KV: Low-Rank KV Cache Compression via Soft Thresholding for Adaptive Rank Control
- PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling
- Communication Compression for Tensor Parallel LLM Inference
- RULER: What's the Real Context Size of Your Long-Context Language Models?
- DeInfer: Efficient Parallel Inferencing for Decomposed Large Language Models
- GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM
- Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- 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
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents