Skip to content
AI News HubLIVE
Original source2 min read

Forcing a Letter Suppresses Refusal: Mechanistic Evidence from WMDP

Summary

A study accepted to BlackboxNLP 2026 provides the first mechanistic explanation for why multiple-choice question format reduces refusal rates in LLMs: models commit quickly to answer letters, activations lie farther from the refusal direction, and answer selection competes with refusal.

SourcearXiv Computational LinguisticsAuthor: Stepan Epifantsev, Anna Antipova, Oleg Chernikov, Leon Strelkov, Alexey Dontsov
Forcing a Letter Suppresses Refusal: Mechanistic Evidence from WMDP
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 5 Nov 2026]

Title:Forcing a Letter Suppresses Refusal: Mechanistic Evidence from WMDP

View a PDF of the paper titled Forcing a Letter Suppresses Refusal: Mechanistic Evidence from WMDP, by Stepan Epifantsev and 4 other authors

View PDF HTML (experimental)

Abstract:Safety benchmarks for large language models (LLMs) commonly use multiple-choice question (MCQ) format. However, prior studies documented that MCQ format reduces refusal rates, which may significantly affect the safety behavior of the model. We provide a first mechanistic explanation of this effect on the Weapons of Mass Destruction Proxy (WMDP) dataset across two models - Llama-3.1-8B and Qwen3-8B. First, we use the tuned lens technique to show that when processing MCQ the model commits to an answer-letter token very quickly, which is in contrast to the gradual refusal build-up in open-ended formats. Additionally, we observe that activations on the MCQ format lie roughly 3x further from the refusal direction, compared with their open-ended counterparts. Moreover, ablating the attention heads that carry the MCQ format answer increases the refusal, which is a direct evidence that the MCQ answer-selection competes with refusal. These findings indicate that MCQ safety scores primarily measure knowledge capabilities and do not assess the safety-mechanisms.

Comments: Accepted to BlackboxNLP 2026 (EMNLP 2026 workshop). 17 pages, 15 figures

Subjects:

Computation and Language (cs.CL); Machine Learning (cs.LG)

Cite as: arXiv:2611.08639 [cs.CL]

(or arXiv:2611.08639v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2611.08639

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Stepan Epifantsev [view email] [v1] Thu, 5 Nov 2026 19:36:44 UTC (4,997 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Forcing a Letter Suppresses Refusal: Mechanistic Evidence from WMDP, by Stepan Epifantsev and 4 other authors

View PDF

HTML (experimental)

TeX Source

view license

Additional Features

Audio Summary

Current browse context:

cs.CL

new | recent | 2026-11

Change to browse by:

cs cs.LG

References & Citations

SciX

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • Models commit to an answer letter very quickly in MCQ format, unlike the gradual refusal build-up in open-ended formats.
  • MCQ activations lie roughly 3x farther from the refusal direction.
  • Ablating attention heads carrying the MCQ answer increases refusal, directly showing competition between answer selection and refusal.
  • MCQ safety scores primarily measure knowledge capabilities, not safety mechanisms.

Highlights and analysis are generated automatically and may contain errors. Check the original source.

Forcing a Letter Suppresses Refusal: Mechanistic Evidence from WMDP | AI News Hub