Cover Image

How Overconfidence in Initial Choices and Underconfidence Under Criticism Modulate Change of Mind in Large Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Kumaran, Dharshan, Fleming, Stephen M, Markeeva, Larisa, Heyward, Joe, Banino, Andrea, Mathur, Mrinal, Pascanu, Razvan, Osindero, Simon, de Martino, Benedetto, Velickovic, Petar, Patraucean, Viorica
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

How LLMs Detect and Correct Their Own Errors: The Role of Internal Confidence Signals
by: Kumaran, Dharshan, et al.
Published: (2026)

Causal Evidence that Language Models use Confidence to Drive Behavior
by: Kumaran, Dharshan, et al.
Published: (2026)

How do LLMs Compute Verbal Confidence
by: Kumaran, Dharshan, et al.
Published: (2026)

Understanding Performance Gap Between Parallel and Sequential Sampling in Large Reasoning Models
by: Gu, Xiangming, et al.
Published: (2026)

The Illusion of Stochasticity in LLMs
by: Gu, Xiangming, et al.
Published: (2026)

Transformers need glasses! Information over-squashing in language tasks
by: Barbero, Federico, et al.
Published: (2024)

Perplexity Cannot Always Tell Right from Wrong
by: Veličković, Petar, et al.
Published: (2026)

Transformers meet Neural Algorithmic Reasoners
by: Bounsi, Wilfried, et al.
Published: (2024)

Latent Space Representations of Neural Algorithmic Reasoners
by: Mirjanić, Vladimir V., et al.
Published: (2023)

Softmax is not Enough (for Sharp Size Generalisation)
by: Veličković, Petar, et al.
Published: (2024)

Asynchronous Algorithmic Alignment with Cocycles
by: Dudzik, Andrew, et al.
Published: (2023)

Perception Test 2024: Challenge Summary and a Novel Hour-Long VideoQA Benchmark
by: Heyward, Joseph, et al.
Published: (2024)

Round and Round We Go! What makes Rotary Positional Encodings useful?
by: Barbero, Federico, et al.
Published: (2024)

TRecViT: A Recurrent Video Transformer
by: Pătrăucean, Viorica, et al.
Published: (2024)

Amplifying human performance in combinatorial competitive programming
by: Veličković, Petar, et al.
Published: (2024)

Mining Generalizable Activation Functions
by: Vitvitskyi, Alex, et al.
Published: (2026)

Filter Equivariant Functions: A symmetric account of length-general extrapolation on lists
by: Lewis, Owen, et al.
Published: (2025)

AI Failure Loops in Devalued Work: The Confluence of Overconfidence in AI and Underconfidence in Worker Expertise
by: Kawakami, Anna, et al.
Published: (2025)

Why do LLMs attend to the first token?
by: Barbero, Federico, et al.
Published: (2025)

Synth$^2$: Boosting Visual-Language Models with Synthetic Captions and Image Embeddings
by: Sharifzadeh, Sahand, et al.
Published: (2024)

Perception Test 2025: Challenge Summary and a Unified VQA Extension
by: Heyward, Joseph, et al.
Published: (2026)

A Simple Recipe for Contrastively Pre-training Video-First Encoders Beyond 16 Frames
by: Papalampidi, Pinelopi, et al.
Published: (2023)

Learning from Streaming Video with Orthogonal Gradients
by: Han, Tengda, et al.
Published: (2025)

The CLRS-Text Algorithmic Reasoning Language Benchmark
by: Markeeva, Larisa, et al.
Published: (2024)

Recurrent Aggregators in Neural Algorithmic Reasoning
by: Xu, Kaijia, et al.
Published: (2024)

Leveraging Classical Algorithms for Graph Neural Networks
by: Wu, Jason, et al.
Published: (2025)

Dynamic Reflections: Probing Video Representations with Text Alignment
by: Zhu, Tyler, et al.
Published: (2025)

Optimizers Qualitatively Alter Solutions And We Should Leverage This
by: Pascanu, Razvan, et al.
Published: (2025)

Lattice: Learning to Efficiently Compress the Memory
by: Karami, Mahdi, et al.
Published: (2025)

Deep Grokking: Would Deep Neural Networks Generalize Better?
by: Fan, Simin, et al.
Published: (2024)

Learning from One Continuous Video Stream
by: Carreira, João, et al.
Published: (2023)

Parallel Algorithms Align with Neural Execution
by: Engelmayer, Valerie, et al.
Published: (2023)

Commute-Time-Optimised Graphs for GNNs
by: Sterner, Igor, et al.
Published: (2024)

NoProp: Training Neural Networks without Full Back-propagation or Full Forward-propagation
by: Li, Qinyu, et al.
Published: (2025)

Meta-learning how to Share Credit among Macro-Actions
by: Hosu, Ionel-Alexandru, et al.
Published: (2025)

Revisiting Adam for Streaming Reinforcement Learning
by: Gogianu, Florin, et al.
Published: (2026)

Are Generative Models Underconfident? Better Quality Estimation with Boosted Model Probability
by: Dinh, Tu Anh, et al.
Published: (2025)

Cayley Graph Propagation
by: Wilson, JJ, et al.
Published: (2024)

Building on Efficient Foundations: Effectively Training LLMs with Structured Feedforward Layers
by: Wei, Xiuying, et al.
Published: (2024)

What Can Grokking Teach Us About Learning Under Nonstationarity?
by: Lyle, Clare, et al.
Published: (2025)