Cover Image

Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Otth, Matthias, Hübotter, Jonas, Hakimi, Ido, Krause, Andreas
Format:	Preprint
Published:	2025
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Efficiently Learning at Test-Time: Active Fine-Tuning of LLMs
by: Hübotter, Jonas, et al.
Published: (2024)

Local Mixtures of Experts: Essentially Free Test-Time Training via Model Merging
by: Bertolissi, Ryo, et al.
Published: (2025)

Learning on the Job: Test-Time Curricula for Targeted Reinforcement Learning
by: Hübotter, Jonas, et al.
Published: (2025)

Majority Voting for Code Generation
by: Launer, Tim, et al.
Published: (2026)

LITE: Efficiently Estimating Gaussian Probability of Maximality
by: Menet, Nicolas, et al.
Published: (2025)

Probabilistic Artificial Intelligence
by: Krause, Andreas, et al.
Published: (2025)

Test-time Offline Reinforcement Learning on Goal-related Experience
by: Bagatella, Marco, et al.
Published: (2025)

Specialization after Generalization: Towards Understanding Test-Time Training in Foundation Models
by: Hübotter, Jonas, et al.
Published: (2025)

Active Fine-Tuning of Multi-Task Policies
by: Bagatella, Marco, et al.
Published: (2024)

Active Few-Shot Fine-Tuning
by: Hübotter, Jonas, et al.
Published: (2024)

Transductive Active Learning: Theory and Applications
by: Hübotter, Jonas, et al.
Published: (2024)

DISCOVER: Automated Curricula for Sparse-Reward Reinforcement Learning
by: Diaz-Bone, Leander, et al.
Published: (2025)

Reinforcement Learning via Self-Distillation
by: Hübotter, Jonas, et al.
Published: (2026)

Maximizing Confidence Alone Improves Reasoning
by: Prabhudesai, Mihir, et al.
Published: (2025)

Rethinking Fine-Tuning when Scaling Test-Time Compute: Limiting Confidence Improves Mathematical Reasoning
by: Chen, Feng, et al.
Published: (2025)

ActiveUltraFeedback: Efficient Preference Data Generation using Active Learning
by: Melikidze, Davit, et al.
Published: (2026)

Bias-Restrained Prefix Representation Finetuning for Mathematical Reasoning
by: Liang, Sirui, et al.
Published: (2025)

Aligning Language Models from User Interactions
by: Buening, Thomas Kleine, et al.
Published: (2026)

Self-Distillation Enables Continual Learning
by: Shenfeld, Idan, et al.
Published: (2026)

Beyond Pairwise Correlations: Higher-Order Redundancies in Self-Supervised Representation Learning
by: Zollikofer, David, et al.
Published: (2024)

Confidence Estimation via Sequential Likelihood Mixing
by: Kirschner, Johannes, et al.
Published: (2025)

RewardUQ: A Unified Framework for Uncertainty-Aware Reward Models
by: Yang, Daniel, et al.
Published: (2026)

Prefix Grouper: Efficient GRPO Training through Shared-Prefix Forward
by: Liu, Zikang, et al.
Published: (2025)

Fast and Effective On-policy Distillation from Reasoning Prefixes
by: Zhang, Dongxu, et al.
Published: (2026)

Sequential-Parallel Duality in Prefix Scannable Models
by: Yau, Morris, et al.
Published: (2025)

Confidence over Time: Confidence Calibration with Temporal Logic for Large Language Model Reasoning
by: Mao, Zhenjiang, et al.
Published: (2026)

Tool-R0: Self-Evolving LLM Agents for Tool-Learning from Zero Data
by: Acikgoz, Emre Can, et al.
Published: (2026)

Simulation Priors for Data-Efficient Deep Learning
by: Treven, Lenart, et al.
Published: (2025)

Test-Time Tuned Language Models Enable End-to-end De Novo Molecular Structure Generation from MS/MS Spectra
by: Mismetti, Laura, et al.
Published: (2025)

FOReCAst: The Future Outcome Reasoning and Confidence Assessment Benchmark
by: Yuan, Zhangdie, et al.
Published: (2025)

Data-Efficient Task Generalization via Probabilistic Model-based Meta Reinforcement Learning
by: Bhardwaj, Arjun, et al.
Published: (2023)

Efficient Test-Time Inference via Deterministic Exploration of Truncated Decoding Trees
by: Li, Xueyan, et al.
Published: (2026)

When to Sense and Control? A Time-adaptive Approach for Continuous-Time RL
by: Treven, Lenart, et al.
Published: (2024)

MathTutorBench: A Benchmark for Measuring Open-ended Pedagogical Capabilities of LLM Tutors
by: Macina, Jakub, et al.
Published: (2025)

CER: Confidence Enhanced Reasoning in LLMs
by: Razghandi, Ali, et al.
Published: (2025)

Form Follows Function: Recursive Stem Model
by: Hakimi, Navid
Published: (2026)

Plan and Budget: Effective and Efficient Test-Time Scaling on Reasoning Large Language Models
by: Lin, Junhong, et al.
Published: (2025)

ECHO: Entropy-Confidence Hybrid Optimization for Test-Time Reinforcement Learning
by: Zhao, Chu, et al.
Published: (2026)

PrefixQuant: Eliminating Outliers by Prefixed Tokens for Large Language Models Quantization
by: Chen, Mengzhao, et al.
Published: (2024)

PERK: Long-Context Reasoning as Parameter-Efficient Test-Time Learning
by: Chen, Zeming, et al.
Published: (2025)