How Overconfidence in Initial Choices and Underconfidence Under Criticism Modulate Change of Mind in Large Language Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909675636654080 |
|---|---|
| author | Kumaran, Dharshan Fleming, Stephen M Markeeva, Larisa Heyward, Joe Banino, Andrea Mathur, Mrinal Pascanu, Razvan Osindero, Simon de Martino, Benedetto Velickovic, Petar Patraucean, Viorica |
| author_facet | Kumaran, Dharshan Fleming, Stephen M Markeeva, Larisa Heyward, Joe Banino, Andrea Mathur, Mrinal Pascanu, Razvan Osindero, Simon de Martino, Benedetto Velickovic, Petar Patraucean, Viorica |
| contents | Large language models (LLMs) exhibit strikingly conflicting behaviors: they can appear steadfastly overconfident in their initial answers whilst at the same time being prone to excessive doubt when challenged. To investigate this apparent paradox, we developed a novel experimental paradigm, exploiting the unique ability to obtain confidence estimates from LLMs without creating memory of their initial judgments -- something impossible in human participants. We show that LLMs -- Gemma 3, GPT4o and o1-preview -- exhibit a pronounced choice-supportive bias that reinforces and boosts their estimate of confidence in their answer, resulting in a marked resistance to change their mind. We further demonstrate that LLMs markedly overweight inconsistent compared to consistent advice, in a fashion that deviates qualitatively from normative Bayesian updating. Finally, we demonstrate that these two mechanisms -- a drive to maintain consistency with prior commitments and hypersensitivity to contradictory feedback -- parsimoniously capture LLM behavior in a different domain. Together, these findings furnish a mechanistic account of LLM confidence that explains both their stubbornness and excessive sensitivity to criticism. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_03120 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | How Overconfidence in Initial Choices and Underconfidence Under Criticism Modulate Change of Mind in Large Language Models Kumaran, Dharshan Fleming, Stephen M Markeeva, Larisa Heyward, Joe Banino, Andrea Mathur, Mrinal Pascanu, Razvan Osindero, Simon de Martino, Benedetto Velickovic, Petar Patraucean, Viorica Machine Learning Artificial Intelligence Large language models (LLMs) exhibit strikingly conflicting behaviors: they can appear steadfastly overconfident in their initial answers whilst at the same time being prone to excessive doubt when challenged. To investigate this apparent paradox, we developed a novel experimental paradigm, exploiting the unique ability to obtain confidence estimates from LLMs without creating memory of their initial judgments -- something impossible in human participants. We show that LLMs -- Gemma 3, GPT4o and o1-preview -- exhibit a pronounced choice-supportive bias that reinforces and boosts their estimate of confidence in their answer, resulting in a marked resistance to change their mind. We further demonstrate that LLMs markedly overweight inconsistent compared to consistent advice, in a fashion that deviates qualitatively from normative Bayesian updating. Finally, we demonstrate that these two mechanisms -- a drive to maintain consistency with prior commitments and hypersensitivity to contradictory feedback -- parsimoniously capture LLM behavior in a different domain. Together, these findings furnish a mechanistic account of LLM confidence that explains both their stubbornness and excessive sensitivity to criticism. |
| title | How Overconfidence in Initial Choices and Underconfidence Under Criticism Modulate Change of Mind in Large Language Models |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2507.03120 |