Cover Image

Sample Complexity of Autoregressive Reasoning: Chain-of-Thought vs. End-to-End

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Hanneke, Steve, Mehalel, Idan, Moran, Shay
Format:	Preprint
Published:	2026
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Optimal Prediction Using Expert Advice and Randomized Littlestone Dimension
by: Filmus, Yuval, et al.
Published: (2023)

Bandit-Feedback Online Multiclass Classification: Variants and Tradeoffs
by: Filmus, Yuval, et al.
Published: (2024)

A Theory of Online Learning with Autoregressive Chain-of-Thought Reasoning
by: Doron-Arad, Ilan, et al.
Published: (2026)

List Sample Compression and Uniform Convergence
by: Hanneke, Steve, et al.
Published: (2024)

A Theory of Universal Agnostic Learning
by: Hanneke, Steve, et al.
Published: (2026)

PAC Learning with Bandit Feedback: Sharp Sample Complexity in the Realizable Setting
by: Hanneke, Steve, et al.
Published: (2026)

Private List Learnability vs. Online List Learnability
by: Hanneke, Steve, et al.
Published: (2025)

Sample Compression Scheme Reductions
by: Attias, Idan, et al.
Published: (2024)

Regret-Oracle Complexity Tradeoffs in Agnostic Online Learning
by: Attias, Idan, et al.
Published: (2026)

Adversarially Robust PAC Learnability of Real-Valued Functions
by: Attias, Idan, et al.
Published: (2022)

Sample Complexity of Agnostic Multiclass Classification: Natarajan Dimension Strikes Back
by: Cohen, Alon, et al.
Published: (2025)

Most Convolutional Networks Suffer from Small Adversarial Perturbations
by: Daniely, Amit, et al.
Published: (2026)

Deterministic Apple Tasting
by: Chase, Zachary, et al.
Published: (2024)

Data Selection for ERMs
by: Hanneke, Steve, et al.
Published: (2025)

On the ERM Principle in Meta-Learning
by: Alon, Yannay, et al.
Published: (2024)

An Optimal Sauer Lemma Over $k$-ary Alphabets
by: Hanneke, Steve, et al.
Published: (2026)

Optimal Mistake Bounds for Transductive Online Learning
by: Chase, Zachary, et al.
Published: (2025)

Adversarial Resilience in Sequential Prediction via Abstention
by: Goel, Surbhi, et al.
Published: (2023)

Dual VC Dimension Obstructs Sample Compression by Embeddings
by: Chase, Zachary, et al.
Published: (2024)

A Characterization of Semi-Supervised Adversarially-Robust PAC Learnability
by: Attias, Idan, et al.
Published: (2022)

Agnostic Sample Compression Schemes for Regression
by: Attias, Idan, et al.
Published: (2018)

Online Learning of Neural Networks
by: Daniely, Amit, et al.
Published: (2025)

A Tight Lower Bound for Non-stochastic Multi-armed Bandits with Expert Advice
by: Chase, Zachary, et al.
Published: (2025)

On the Learning Curves of Revenue Maximization
by: Hanneke, Steve, et al.
Published: (2026)

Tradeoffs between Mistakes and ERM Oracle Calls in Online and Transductive Online Learning
by: Attias, Idan, et al.
Published: (2025)

Online Realizable Regression and Applications for ReLU Networks
by: Doron-Arad, Ilan, et al.
Published: (2026)

Ramsey Theorems for Trees and a General 'Private Learning Implies Online Learning' Theorem
by: Fioravanti, Simone, et al.
Published: (2024)

Adaptive Sample Aggregation In Transfer Learning
by: Hanneke, Steve, et al.
Published: (2024)

Multiclass Online Learnability under Bandit Feedback
by: Raman, Ananth, et al.
Published: (2023)

Optimal Learners for Realizable Regression: PAC Learning and Online Learning
by: Attias, Idan, et al.
Published: (2023)

End-to-End Autoregressive Image Generation with 1D Semantic Tokenizer
by: Chu, Wenda, et al.
Published: (2026)

On Characterizing Learnability for Adversarial Noisy Bandits
by: Hanneke, Steve, et al.
Published: (2026)

Universal Multiclass Transductive Online Learning
by: Hanneke, Steve, et al.
Published: (2026)

When More Data Doesn't Help: Limits of Adaptation in Multitask Learning
by: Hanneke, Steve, et al.
Published: (2026)

The Dimension of Self-Directed Learning
by: Devulapalli, Pramith, et al.
Published: (2024)

Universal rates of ERM for agnostic learning
by: Hanneke, Steve, et al.
Published: (2025)

Learning from Snapshots of Discrete and Continuous Data Streams
by: Devulapalli, Pramith, et al.
Published: (2024)

Universal Rates of Empirical Risk Minimization
by: Hanneke, Steve, et al.
Published: (2024)

A Theory of Optimistically Universal Online Learnability for General Concept Classes
by: Hanneke, Steve, et al.
Published: (2025)

CoMAT: Chain of Mathematically Annotated Thought Improves Mathematical Reasoning
by: Leang, Joshua Ong Jun, et al.
Published: (2024)