What does research say about: LLMs struggle to use information placed in the middle of long contexts?

Model performance degrades by up to 20% when key information is placed in the middle vs the beginning or end of a 4K-token context window. (Source: Liu et al., 'Lost in the Middle: How Language Models Use Long Contexts', TACL 2024). This 'Lost in the Middle' effect means prompt engineers must strategically place critical instructions at the start and end of their prompts.

LLMs struggle to use information placed in the middle of…

Context & Methodology

This 'Lost in the Middle' effect means prompt engineers must strategically place critical instructions at the start and end of their prompts.

Applies To

openaianthropicgoogle

Confidence Level

High

Implementation Effort

low

Recommendation

Execution Priority

Put This Evidence to Work

Use the STCO framework to implement findings like this in structured, testable prompts.

Start Building Free Browse All 141 Citations

ROI Calculator Token Calculator Prompt Templates