M. Laurent and A. McLeod, “Query-Aware Context Compression for Efficient Long-Context Language Model Inference”, CPL, vol. 15, no. 2, pp. 48–54, Nov. 2026, doi: 10.54097/zhed3474.