Copertina

Honesty over Accuracy: Trustworthy Language Models through Reinforced Hesitation

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Mohamadi, Mohamad Amin, Wang, Tianhao, Li, Zhiyuan
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Adam Exploits $\ell_\infty$-geometry of Loss Landscape via Coordinate-wise Adaptivity
di: Xie, Shuo, et al.
Pubblicazione: (2024)

Why Do You Grok? A Theoretical Analysis of Grokking Modular Addition
di: Mohamadi, Mohamad Amin, et al.
Pubblicazione: (2024)

The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systems
di: Ren, Richard, et al.
Pubblicazione: (2025)

Honesty to Subterfuge: In-Context Reinforcement Learning Can Make Honest Models Reward Hack
di: McKee-Reid, Leo, et al.
Pubblicazione: (2024)

Training LLMs for Honesty via Confessions
di: Joglekar, Manas, et al.
Pubblicazione: (2025)

EvidenceRL: Reinforcing Evidence Consistency for Trustworthy Language Models
di: Tamo, J. Ben, et al.
Pubblicazione: (2026)

How do Large Language Models Navigate Conflicts between Honesty and Helpfulness?
di: Liu, Ryan, et al.
Pubblicazione: (2024)

Honesty in Causal Forests: When It Helps and When It Hurts
di: Hou, Yanfang, et al.
Pubblicazione: (2025)

Mix Data or Merge Models? Balancing the Helpfulness, Honesty, and Harmlessness of Large Language Model via Model Merging
di: Yang, Jinluan, et al.
Pubblicazione: (2025)

Provable Benefit of Sign Descent: A Minimal Model Under Heavy-Tailed Class Imbalance
di: Yadav, Robin, et al.
Pubblicazione: (2025)

A Tale of Two Geometries: Adaptive Optimizers and Non-Euclidean Descent
di: Xie, Shuo, et al.
Pubblicazione: (2025)

Relative Kinetic Utility for Reasoning-Aware Structural Pruning in Large Language Models
di: Qian, Tianhao
Pubblicazione: (2026)

Know your Trajectory -- Trustworthy Reinforcement Learning deployment through Importance-Based Trajectory Analysis
di: F, Clifford, et al.
Pubblicazione: (2025)

Preference Learning with Lie Detectors can Induce Honesty or Evasion
di: Cundy, Chris, et al.
Pubblicazione: (2025)

Unlearners Can Lie: Evaluating and Improving Honesty in LLM Unlearning
di: Gu, Renjie, et al.
Pubblicazione: (2026)

Structured Preconditioners in Adaptive Optimization: A Unified Analysis
di: Xie, Shuo, et al.
Pubblicazione: (2025)

Incentivizing Honesty among Competitors in Collaborative Learning and Optimization
di: Dorner, Florian E., et al.
Pubblicazione: (2023)

AntiPaSTO: Self-Supervised Honesty Steering via Anti-Parallel Representations
di: Clark, Michael J.
Pubblicazione: (2026)

The Obfuscation Atlas: Mapping Where Honesty Emerges in RLVR with Deception Probes
di: Taufeeque, Mohammad, et al.
Pubblicazione: (2026)

TrustLDM: Benchmarking Trustworthiness in Language Diffusion Models
di: Mo, Yichuan, et al.
Pubblicazione: (2026)

The Marginal Value of Momentum for Small Learning Rate SGD
di: Wang, Runzhe, et al.
Pubblicazione: (2023)

Enhanced High-Dimensional Data Visualization through Adaptive Multi-Scale Manifold Embedding
di: Ni, Tianhao, et al.
Pubblicazione: (2025)

Think Before You Lie: How Reasoning Leads to Honesty
di: Yuan, Ann, et al.
Pubblicazione: (2026)

Your Offline Policy is Not Trustworthy: Bilevel Reinforcement Learning for Sequential Portfolio Optimization
di: Yuan, Haochen, et al.
Pubblicazione: (2025)

PrivORL: Differentially Private Synthetic Dataset for Offline Reinforcement Learning
di: Gong, Chen, et al.
Pubblicazione: (2025)

TrajDeleter: Enabling Trajectory Forgetting in Offline Reinforcement Learning Agents
di: Gong, Chen, et al.
Pubblicazione: (2024)

Generalizing Trust: Weak-to-Strong Trustworthiness in Language Models
di: Pawelczyk, Martin, et al.
Pubblicazione: (2024)

Beyond Accuracy: On the Effects of Fine-tuning Towards Vision-Language Model's Prediction Rationality
di: Wang, Qitong, et al.
Pubblicazione: (2024)

Trustworthy Classification through Rank-Based Conformal Prediction Sets
di: Luo, Rui, et al.
Pubblicazione: (2024)

HARP: Hesitation-Aware Reframing in Transformer Inference Pass
di: Storaï, Romain, et al.
Pubblicazione: (2024)

Strong Transitivity Relations and Graph Neural Networks
di: Mohamadi, Yassin, et al.
Pubblicazione: (2024)

Beyond Benchmarks: Dynamic, Automatic And Systematic Red-Teaming Agents For Trustworthy Medical Language Models
di: Pan, Jiazhen, et al.
Pubblicazione: (2025)

CARES: A Comprehensive Benchmark of Trustworthiness in Medical Vision Language Models
di: Xia, Peng, et al.
Pubblicazione: (2024)

Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignment
di: Liu, Yang, et al.
Pubblicazione: (2023)

Unlearning Imperative: Securing Trustworthy and Responsible LLMs through Engineered Forgetting
di: Kang, James Jin, et al.
Pubblicazione: (2025)

Evaluating Reinforcement Learning Safety and Trustworthiness in Cyber-Physical Systems
di: Dearstyne, Katherine, et al.
Pubblicazione: (2025)

T1: Advancing Language Model Reasoning through Reinforcement Learning and Inference Scaling
di: Hou, Zhenyu, et al.
Pubblicazione: (2025)

A Survey on Medical Large Language Models: Technology, Application, Trustworthiness, and Future Directions
di: Liu, Lei, et al.
Pubblicazione: (2024)

AutoTrust: Benchmarking Trustworthiness in Large Vision Language Models for Autonomous Driving
di: Xing, Shuo, et al.
Pubblicazione: (2024)

Open-World Reinforcement Learning over Long Short-Term Imagination
di: Li, Jiajian, et al.
Pubblicazione: (2024)