Peer-reviewed research
Lost in the Middle: How Language Models Use Long Contexts
Nelson F. Liu et al., arXiv (accepted to Transactions of the Association for Computational Linguistics), 2023
Read the originalRead on October 17, 2026
- What it is
- A study of how language models use information placed at different positions in a long input. The authors test multi-document question answering and key-value retrieval, and move the one relevant passage to the start, middle or end of the context.
- How much weight it carries
- Peer-reviewed at TACL and widely cited. The tasks are retrieval tasks rather than chat prompts, and the models are from 2023, so the exact size of the effect will differ in newer models. The direction of the finding is the part to rely on.
Facts taken from this source
- In tests on long inputs, language model accuracy was highest when the relevant information sat at the start or end of the input, and dropped significantly when it sat in the middle.
“performance is often highest when relevant information occurs at the beginning or end of the input context, and significantly degrades when models must access relevant information in the middle of long contexts” (Abstract)