- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
QREAM does query-agnostic KV-cache compression. It drops AM's per-key bias by directly matching the normalized attention output, then solves the retained values by ridge toward the originals, and picks keys via pivoted QR on value-weighted attention columns. Tested on Qwen3-0.6B/4B/8B over QuALITY/LongHealth/QASPER at 5×–100×.
Critical strengths
- The bias-free reformulation is the paper's strongest selling point and it is well-argued: by matching the normalized output directly and re-solving the values, QREAM keeps a standard KV cache representation, so it drops into existing inference stacks without the kernel modification AM requires.
- The pivoted-QR key selection is well-motivated by the method's own math: since the re-solved values can only live in the row space spanned by the retained attention columns, picking keys for linear independence rather than raw attention magnitude is the correct criterion.
- The 100× result is a clean and useful empirical finding: every eviction-based baseline collapses to near-zero F1 while the attention-matched caches retain most performance. This sharply illustrates that "keeping high-scoring tokens" is not the same as "keeping the information needed to answer," and it gives the paper a real headline.
Critical weaknesses
- In your method, every retained value is re-solved, so it can overfit or break when the reference continuation is long. The benchmarks here (LongBench/HealthBench-type) have short outputs, so it needs a long-generation test like LongGenBench.
- No overhead analysis against traditional cheap methods. Your method must compare against StreamingLLM/SnapKV, and show the gain is worth the added compression-time cost.
- SnapKV scores are implausibly low; it is unclear if the question is dropped during compression, so this setup must be stated.
- H2O, the canonical attention-eviction baseline, is absent and should be included. The paper also lacks both a reproducibility statement and an ethics statement, which is a straightforward completeness gap for a submission.
Questions for authors
The paper states that QREAM and AM "use the same reference queries". However, for Qwen3-0.6B and 8B, QREAM uses the full ss+repeat probe (five reference-query families; qream_press.py L470-537) while AM is run as AM-fast with repeat-only (evaluate_registry.py L134/143, am_press.py L67). Is this design reasonable or a mistake?