Mind the Gap: How Elicitation Protocols Shape the Stated-Revealed Preference Gap in Language Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914563568435200 |
|---|---|
| author | Mahajan, Pranav Kendiukhov, Ihor Hussain, Syed Nottingham, Lydia |
| author_facet | Mahajan, Pranav Kendiukhov, Ihor Hussain, Syed Nottingham, Lydia |
| contents | Recent work identifies a stated-revealed (SvR) preference gap in language models (LMs): a mismatch between the values models endorse and the choices they make in context. Existing evaluations rely heavily on binary forced-choice prompting, which entangles genuine preferences with artifacts of the elicitation protocol. We systematically study how elicitation protocols affect SvR correlation across 24 LMs. Allowing neutrality and abstention during stated preference elicitation allows us to exclude weak signals, substantially improving Spearman's rank correlation ($ρ$) between volunteered stated preferences and forced-choice revealed preferences. However, further allowing abstention in revealed preferences drives $ρ$ to near-zero or negative values due to high neutrality rates. Finally, we find that system prompt steering using stated preferences during revealed preference elicitation does not reliably improve SvR correlation on AIRiskDilemmas. Together, our results show that SvR correlation is highly protocol-dependent and that preference elicitation requires methods that account for indeterminate preferences. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_21975 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Mind the Gap: How Elicitation Protocols Shape the Stated-Revealed Preference Gap in Language Models Mahajan, Pranav Kendiukhov, Ihor Hussain, Syed Nottingham, Lydia Artificial Intelligence Emerging Technologies Recent work identifies a stated-revealed (SvR) preference gap in language models (LMs): a mismatch between the values models endorse and the choices they make in context. Existing evaluations rely heavily on binary forced-choice prompting, which entangles genuine preferences with artifacts of the elicitation protocol. We systematically study how elicitation protocols affect SvR correlation across 24 LMs. Allowing neutrality and abstention during stated preference elicitation allows us to exclude weak signals, substantially improving Spearman's rank correlation ($ρ$) between volunteered stated preferences and forced-choice revealed preferences. However, further allowing abstention in revealed preferences drives $ρ$ to near-zero or negative values due to high neutrality rates. Finally, we find that system prompt steering using stated preferences during revealed preference elicitation does not reliably improve SvR correlation on AIRiskDilemmas. Together, our results show that SvR correlation is highly protocol-dependent and that preference elicitation requires methods that account for indeterminate preferences. |
| title | Mind the Gap: How Elicitation Protocols Shape the Stated-Revealed Preference Gap in Language Models |
| topic | Artificial Intelligence Emerging Technologies |
| url | https://arxiv.org/abs/2601.21975 |