hendel-scope-single-token-output

IN premise — summaries/2026/08/24/hendel-2023-icl-task-vectors-s8-conclusions.md

Created 2026-08-25T02:58:04+00:00

The single-task-vector model in Hendel et al. applies to simple, single-token-output tasks; complex multi-step reasoning (e.g., arithmetic) may require multiple or higher-dimensional representations.

Summary

The result from Hendel et al. only covers the simplest kind of prediction — tasks where the model just picks one word or symbol as output. Anything requiring a chain of intermediate steps, like working through an arithmetic problem, sits outside what that finding can support, so the system should not treat it as evidence for multi-step reasoning capabilities.