Image de couverture de livre

Perplexity Cannot Always Tell Right from Wrong

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Veličković, Petar, Barbero, Federico, Perivolaropoulos, Christos, Osindero, Simon, Pascanu, Razvan
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning Artificial Intelligence Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Softmax is not Enough (for Sharp Size Generalisation)
par: Veličković, Petar, et autres
Publié: (2024)

Round and Round We Go! What makes Rotary Positional Encodings useful?
par: Barbero, Federico, et autres
Publié: (2024)

How do LLMs Compute Verbal Confidence
par: Kumaran, Dharshan, et autres
Publié: (2026)

Filter Equivariant Functions: A symmetric account of length-general extrapolation on lists
par: Lewis, Owen, et autres
Publié: (2025)

Why do LLMs attend to the first token?
par: Barbero, Federico, et autres
Publié: (2025)

Transformers need glasses! Information over-squashing in language tasks
par: Barbero, Federico, et autres
Publié: (2024)

Asynchronous Algorithmic Alignment with Cocycles
par: Dudzik, Andrew, et autres
Publié: (2023)

Mining Generalizable Activation Functions
par: Vitvitskyi, Alex, et autres
Publié: (2026)

The Illusion of Stochasticity in LLMs
par: Gu, Xiangming, et autres
Publié: (2026)

Wanting to be Understood
par: Fernando, Chrisantha, et autres
Publié: (2025)

Always Tell Me The Odds: Fine-grained Conditional Probability Estimation
par: Wang, Liaoyaqi, et autres
Publié: (2025)

How Overconfidence in Initial Choices and Underconfidence Under Criticism Modulate Change of Mind in Large Language Models
par: Kumaran, Dharshan, et autres
Publié: (2025)

Rethinking GSPO: The Perplexity-Entropy Equivalence
par: Liu, Chi
Publié: (2025)

Latent Space Representations of Neural Algorithmic Reasoners
par: Mirjanić, Vladimir V., et autres
Publié: (2023)

Transformers meet Neural Algorithmic Reasoners
par: Bounsi, Wilfried, et autres
Publié: (2024)

Momentum Point-Perplexity Mechanics in Large Language Models
par: Tomaz, Lorenzo, et autres
Publié: (2025)

Inconsistent Tokenizations Cause Language Models to be Perplexed by Japanese Grammar
par: Gambardella, Andrew, et autres
Publié: (2025)

Alzheimer's Dementia Detection Using Perplexity from Paired Large Language Models
par: Xiao, Yao, et autres
Publié: (2025)

Always Learning, Always Mixing: Efficient and Simple Data Mixing All The Time
par: Hu, Michael Y., et autres
Publié: (2026)

Recurrent Aggregators in Neural Algorithmic Reasoning
par: Xu, Kaijia, et autres
Publié: (2024)

Beyond Perplexity: A Geometric and Spectral Study of Low-Rank Pre-Training
par: Shivagunde, Namrata, et autres
Publié: (2026)

SePer: Measure Retrieval Utility Through The Lens Of Semantic Perplexity Reduction
par: Dai, Lu, et autres
Publié: (2025)

Easy Problems That LLMs Get Wrong
par: Williams, Sean, et autres
Publié: (2024)

Stepwise Perplexity-Guided Refinement for Efficient Chain-of-Thought Reasoning in Large Language Models
par: Cui, Yingqian, et autres
Publié: (2025)

On the generalization of language models from in-context learning and finetuning: a controlled study
par: Lampinen, Andrew K., et autres
Publié: (2025)

What is Wrong with Perplexity for Long-context Language Modeling?
par: Fang, Lizhe, et autres
Publié: (2024)

Chain-of-Thought Reasoning In The Wild Is Not Always Faithful
par: Arcuschin, Iván, et autres
Publié: (2025)

Weak-to-Strong Elicitation via Mismatched Wrong Drafts
par: Deng, Wei
Publié: (2026)

Reasoning Models Don't Always Say What They Think
par: Chen, Yanda, et autres
Publié: (2025)

Amplifying human performance in combinatorial competitive programming
par: Veličković, Petar, et autres
Publié: (2024)

Semantic Anchors in In-Context Learning: Why Small LLMs Cannot Flip Their Labels
par: Kumar, Anantha Padmanaban Krishna
Publié: (2025)

Lattice: Learning to Efficiently Compress the Memory
par: Karami, Mahdi, et autres
Publié: (2025)

Rethinking Reward Model Evaluation: Are We Barking up the Wrong Tree?
par: Wen, Xueru, et autres
Publié: (2024)

The Strongest Teacher Is Not Always the Best Teacher: Student-Centric Answer Selection
par: Hu, Zhengyu, et autres
Publié: (2026)

It's Not Always Sycophancy: Measuring LLM Conformity as a Function of Epistemic Uncertainty
par: Guo, Kevin H., et autres
Publié: (2026)

Sparse but Wrong: Incorrect L0 Leads to Incorrect Features in Sparse Autoencoders
par: Chanin, David, et autres
Publié: (2025)

ELLEN: Extremely Lightly Supervised Learning For Efficient Named Entity Recognition
par: Riaz, Haris, et autres
Publié: (2024)

An Interpretable N-gram Perplexity Threat Model for Large Language Model Jailbreaks
par: Boreiko, Valentyn, et autres
Publié: (2024)

Revisiting Adam for Streaming Reinforcement Learning
par: Gogianu, Florin, et autres
Publié: (2026)

Meta-learning how to Share Credit among Macro-Actions
par: Hosu, Ionel-Alexandru, et autres
Publié: (2025)