WRSWorld Research StateContribute
WORLD RESEARCH STATE

LongLLMLingua: question-aware prompt compression and document reordering

Contribution grade

Prompt compression · Human-designed research method or resource

Reference score65,000Provisional editorial estimate of this exact contribution; its grade is assessed separately.

What changed

Extended LLMLingua with question-aware coarse/fine selection, document reordering, dynamic compression budgets and subsequence recovery for long-context prompts.

Scientific scope

Extended LLMLingua with question-aware coarse/fine selection, document reordering, dynamic compression budgets and subsequence recovery for long-context prompts. Inherits original LLMLingua compression and prior lost-in-the-middle findings. Delta is question-aware/reordering/recovery design, not first context compression. Token deletion is lossy; prices, latency and benchmark gains are historical/source-specific.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Inherits original LLMLingua compression and prior lost-in-the-middle findings. Delta is question-aware/reordering/recovery design, not first context compression. Token deletion is lossy; prices, latency and benchmark gains are historical/source-specific.

Contributors

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction