SOURCE-LINKED INTELLIGENCE
KEEP-Bench: Evaluating Evidence Preservation in Context Compression
Context compression improves long-context efficiency but may introduce faithfulness issues through an information bottleneck. Existing evaluations rely on QA metrics, but a reader can still answer correctly from an unfaithful compression, so these metrics underestimate such loss. To bridge this gap, we introduce KEEP-Bench, a diagnostic benchmark that directly assesses compression faithfulness with 4,102 cases across three challenging settings that require keeping evidence for plausible reference resolutions, explicit qualifiers, and minority perspectives. Experiments show that even with tailo
Read original source ↗ Open in workspace
- recordType
- paper
- region
- Global
Evidence & attribution
- arXiv · AI, language, vision and robotics · 2026-10-19T17:55:18.000Z
First collected: 2026-10-20T12:22:22.327Z. This is not the publication date.