TriAttention compresses KV cache by exploiting stable pre-RoPE Q/K concentration and trigonometric distance preferences to match full-attention reasoning accuracy with far lower memory and higher speed.
Transformers are multi-state rnns
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
TriAttention: Efficient Long Reasoning with Trigonometric KV Compression
TriAttention compresses KV cache by exploiting stable pre-RoPE Q/K concentration and trigonometric distance preferences to match full-attention reasoning accuracy with far lower memory and higher speed.