AI RESEARCH
[R] TriAttention: Efficient KV Cache Compression for Long-Context Reasoning
r/MachineLearning
•
Submitted by /u/Benlus [link] [comments]