Thin Keys, Full Values: Reducing KV Cache via Low-Dimensional Attention Selection
Hengshuai Yao, Xing Chen, Ahmed Murtadha, Guan Wang · Feb 16, 2026 · Citations: 0
Data freshness
Extraction: FreshCheck recency before relying on this page for active eval decisions. Use stale pages as context and verify against current hub results.
Metadata refreshed
Mar 18, 2026, 3:05 PM
StaleExtraction refreshed
Apr 13, 2026, 6:38 AM
FreshExtraction source
Persisted extraction
Confidence 0.20
Abstract
Standard transformer attention uses identical dimensionality for queries, keys, and values, yet these components serve different roles: queries and keys produce scalar attention weights (selection), while values carry rich representations (value transfer). We show that selection requires only $O(\log N)$ dimensions to distinguish among $N$ relevant token categories (e.g., syntactic roles, semantic clusters, positional patterns) -- far fewer than value transfer needs. We introduce factored keys, which exploit this asymmetry to physically shrink the KV cache of any pretrained model without retraining from scratch -- unlike GQA and MLA, which must be designed into the architecture before pretraining. We factorize each key projection $W_K \approx A_{d \times r} B_{r \times d}$ via truncated SVD (where $r = d_{\text{select}}$), set $W_K' = A$ as the new key projection producing compact $r$-dimensional keys for the cache, and absorb $B^\top$ into the query projection ($W_Q' = W_Q B^\top$) at zero cost -- since queries are never cached. At 7B scale, training from scratch with $r = d_{\text{model}}/4$ matches full-attention perplexity (9.2 vs 9.3 PPL after 20B tokens) while using 12% fewer parameters and training 8% faster. For existing models, SVD + QK fine-tuning (3 epochs, less than 1% of pretraining data) achieves 75% key cache savings at approximately 2% quality cost on both GPT-2 and Mistral-7B. The approach composes with GQA and quantization for up to $16\times$ combined key cache compression. For a 7B model serving 128K context, factored keys save 25 GB of KV cache per user, enabling approximately 60% more concurrent users on identical hardware.