---
title: "Distractor task"
type: "entity"
entity_kind: "concept"
status: "hub"
canonical_name: "Distractor task"
aliases: []
first_seen: "2026-10-23T00:00:00.000Z"
writer_model: "claude-sonnet-5"
connects_to: ["inverse scaling","Inverse Scaling Prize","U-shaped scaling","task-taxonomy"]
seek_code_commit: "d40ee39"
---


A proposed mechanism for scale-dependent performance drops in language models: a benchmark task contains an easier, superficially plausible sub-task that a model can latch onto instead of the harder behavior the task actually intends to measure. Mid-capability models get good enough at the easy distractor to actively prefer it over the real task, producing a drop in measured performance; only sufficiently capable models learn to recognize and bypass the distractor.

Matters to this vault as a mechanism that does double duty across two independent papers: [[claim-mckenzie-2023-inverse-scaling-four-causal-categories|McKenzie et al. (2023)]] name "Distractor Task" as one of four causal categories of inverse scaling, and [[claim-wei-2023-inverse-scaling-becomes-u-shaped-at-larger-scale|Wei, Kim, Tay & Le (2023)]] separately invoke the same mechanism to explain why many of those same tasks later recover at larger scale — the one explanation covers both the onset and the resolution of the divergence.

## References
- [[claim-mckenzie-2023-inverse-scaling-four-causal-categories]]
- [[claim-wei-2023-inverse-scaling-becomes-u-shaped-at-larger-scale]]
- Captures: 10-inbox/raw/2026-10-23-does-published-ml-research-go-further-than-hu.md