liu-2023-models-evaluated

IN premise — summaries/2026/08/24/liu-2023-lost-in-middle-s1-introduction.md

Created 2026-08-25T02:58:08+00:00

The lost-in-the-middle study evaluates GPT-3.5-Turbo, Claude-1.3, MPT-30B-Instruct, and LongChat-13B 16K, with extended-context variants GPT-3.5-Turbo (16K) and Claude-1.3 (100K).

Summary

This pins down exactly which models the lost-in-the-middle findings apply to: four main systems (GPT-3.5-Turbo, Claude-1.3, MPT-30B-Instruct, LongChat-13B 16K) plus two longer-context versions of GPT-3.5 and Claude. That scope matters because any conclusion about positional bias or mid-context degradation you cite from this work is only guaranteed for those specific model families and context lengths, not for every LLM you might deploy.