Cover Image

Maximizing Confidence Alone Improves Reasoning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Prabhudesai, Mihir, Chen, Lili, Ippoliti, Alex, Fragkiadaki, Katerina, Liu, Hao, Pathak, Deepak
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

Self-Questioning Language Models
by: Chen, Lili, et al.
Published: (2025)

Aligning Text-to-Image Diffusion Models with Reward Backpropagation
by: Prabhudesai, Mihir, et al.
Published: (2023)

Diffusion Beats Autoregressive in Data-Constrained Settings
by: Prabhudesai, Mihir, et al.
Published: (2025)

Unified Multimodal Discrete Diffusion
by: Swerdlow, Alexander, et al.
Published: (2025)

Video Diffusion Alignment via Reward Gradients
by: Prabhudesai, Mihir, et al.
Published: (2024)

Iterative Refinement Improves Compositional Image Generation
by: Jaiswal, Shantanu, et al.
Published: (2026)

Solving Physics Olympiad via Reinforcement Learning on Physics Simulators
by: Prabhudesai, Mihir, et al.
Published: (2026)

Can LLMs Lie? Investigation beyond Hallucination
by: Huan, Haoran, et al.
Published: (2025)

3D Diffuser Actor: Policy Diffusion with 3D Scene Representations
by: Ke, Tsung-Wei, et al.
Published: (2024)

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

Intrinsic Explainability of Multimodal Learning for Crop Yield Prediction
by: Najjar, Hiba, et al.
Published: (2025)

Marginals Before Conditionals
by: Sahasrabudhe, Mihir
Published: (2026)

Pearls from Pebbles: Improved Confidence Functions for Auto-labeling
by: Vishwakarma, Harit, et al.
Published: (2024)

OT Score: An OT based Confidence Score for Prototype-Assisted Source Free Unsupervised Domain Adaptation
by: Zhang, Yiming, et al.
Published: (2025)

Meta-Evolve: Continuous Robot Evolution for One-to-many Policy Transfer
by: Liu, Xingyu, et al.
Published: (2024)

Evolutionary Policy Optimization
by: Wang, Jianren, et al.
Published: (2025)

HELPER-X: A Unified Instructable Embodied Agent to Tackle Four Interactive Vision-Language Domains with Memory-Augmented Language Models
by: Sarch, Gabriel, et al.
Published: (2024)

Diffusion-ES: Gradient-free Planning with Diffusion for Autonomous Driving and Zero-Shot Instruction Following
by: Yang, Brian, et al.
Published: (2024)

AdaptiveStep: Automatically Dividing Reasoning Step through Model Confidence
by: Liu, Yuliang, et al.
Published: (2025)

Temporalizing Confidence: Evaluation of Chain-of-Thought Reasoning with Signal Temporal Logic
by: Mao, Zhenjiang, et al.
Published: (2025)

VLM Agents Generate Their Own Memories: Distilling Experience into Embodied Programs of Thought
by: Sarch, Gabriel, et al.
Published: (2024)

Standard Neural Computation Alone Is Insufficient for Logical Intelligence
by: Kim, Youngsung
Published: (2025)

HEARTS: Benchmarking LLM Reasoning on Health Time Series
by: Li, Sirui, et al.
Published: (2026)

Reasoning-Aware Training for Time Series Forecasting
by: Ahamed, Md Atik, et al.
Published: (2026)

Think Just Enough: Sequence-Level Entropy as a Confidence Signal for LLM Reasoning
by: Sharma, Aman, et al.
Published: (2025)

Beyond Correctness: Confidence-Aware Reward Modeling for Enhancing Large Language Model Reasoning
by: He, Qianxi, et al.
Published: (2025)

Certified Robustness via Dynamic Margin Maximization and Improved Lipschitz Regularization
by: Fazlyab, Mahyar, et al.
Published: (2023)

GradientSpace: Unsupervised Data Clustering for Improved Instruction Tuning
by: Sridharan, Shrihari, et al.
Published: (2025)

Sanity Checks for Explanation Uncertainty
by: Valdenegro-Toro, Matias, et al.
Published: (2024)

Uncertainty Quantification for Gradient-based Explanations in Neural Networks
by: Mulye, Mihir, et al.
Published: (2024)

LYNX: Learning Dynamic Exits for Confidence-Controlled Reasoning
by: Akgül, Ömer Faruk, et al.
Published: (2025)

Flowing with Confidence
by: de Kruiff, Friso, et al.
Published: (2026)

Energy-based Models are Zero-Shot Planners for Compositional Scene Rearrangement
by: Gkanatsios, Nikolaos, et al.
Published: (2023)

Improving Reproducibility in Evaluation through Multi-Level Annotator Modeling
by: Pandita, Deepak, et al.
Published: (2026)

Flora: Low-Rank Adapters Are Secretly Gradient Compressors
by: Hao, Yongchang, et al.
Published: (2024)

NeuZip: Memory-Efficient Training and Inference with Dynamic Compression of Neural Networks
by: Hao, Yongchang, et al.
Published: (2024)

VI-CuRL: Stabilizing Verifier-Independent RL Reasoning via Confidence-Guided Variance Reduction
by: Cai, Xin-Qiang, et al.
Published: (2026)

Not Just RLHF: Why Alignment Alone Won't Fix Multi-Agent Sycophancy
by: Kumarappan, Adarsh, et al.
Published: (2026)

Early Stopping for Large Reasoning Models via Confidence Dynamics
by: Hosseini, Parsa, et al.
Published: (2026)

Theoretical Insights in Model Inversion Robustness and Conditional Entropy Maximization for Collaborative Inference Systems
by: Xia, Song, et al.
Published: (2025)