hendel-knowledge-data-meng-2022
IN premise — summaries/2026/08/24/hendel-2023-icl-task-vectors-sR-references.md
Created 2026-08-25T02:58:05+00:00
Knowledge tasks in Hendel et al. use the counterfactual dataset from Meng et al. (2022), linking the work to ROME-style knowledge editing.
Summary
Hendel et al. evaluate their knowledge-editing work using the same counterfactual fact dataset that Meng et al. built for ROME, so their results live within the ROME editing lineage and are directly comparable to those baselines. This also means any assumptions or limitations baked into that dataset carry over into Hendel et al.'s conclusions.