Cover Image

Stable Reasoning, Unstable Responses: Mitigating LLM Deception via Stability Asymmetry

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhang, Guoxi, Chen, Jiawei, Yang, Tianzhuo, Qin, Lang, Dai, Juntao, Yang, Yaodong, Yi, Jingwei
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

VISA: Value Injection via Shielded Adaptation for Personalized LLM Alignment
by: Chen, Jiawei, et al.
Published: (2026)

A Game-Theoretic Negotiation Framework for Cross-Cultural Consensus in LLMs
by: Zhang, Guoxi, et al.
Published: (2025)

Mitigating Reward Over-Optimization in RLHF via Behavior-Supported Regularization
by: Dai, Juntao, et al.
Published: (2025)

SafeMCP: Proactive Power Regulation for LLM Agent Defense via Environment-Grounded Look-Ahead Reasoning
by: Wang, Lichao, et al.
Published: (2026)

Mitigating Deceptive Alignment via Self-Monitoring
by: Ji, Jiaming, et al.
Published: (2025)

Safe Reinforcement Learning using Finite-Horizon Gradient-based Estimation
by: Dai, Juntao, et al.
Published: (2024)

MiraBench: Evaluating Action-Conditioned Reliability in Robotic World Models
by: Yang, Tianzhuo, et al.
Published: (2026)

Align Once, Benefit Multilingually: Enforcing Multilingual Consistency for LLM Safety Alignment
by: Bu, Yuyan, et al.
Published: (2026)

SPADE-Bench: Evaluating Spontaneous Strategic Deception in Agents via Plan-Action Divergence
by: Bu, Yuyan, et al.
Published: (2026)

MVR: Multi-view Video Reward Shaping for Reinforcement Learning
by: Luo, Lirui, et al.
Published: (2026)

Act Only When It Pays: Efficient Reinforcement Learning for LLM Reasoning via Selective Rollouts
by: Zheng, Haizhong, et al.
Published: (2025)

Towards Mitigating Excessive Forgetting in LLM Unlearning via Entanglement-Guidance with Proxy Constraint
by: Liu, Zhihao, et al.
Published: (2025)

Aligner: Efficient Alignment by Learning to Correct
by: Ji, Jiaming, et al.
Published: (2024)

Deception at Scale: Deceptive Designs in 1K LLM-Generated Ecommerce Components
by: Chen, Ziwei, et al.
Published: (2025)

Safety-Gymnasium: A Unified Safe Reinforcement Learning Benchmark
by: Ji, Jiaming, et al.
Published: (2023)

UDora: A Unified Red Teaming Framework against LLM Agents by Dynamically Hijacking Their Own Reasoning
by: Zhang, Jiawei, et al.
Published: (2025)

Explore Briefly, Then Decide: Mitigating LLM Overthinking via Cumulative Entropy Regulation
by: Bin, Yi, et al.
Published: (2025)

S-GRPO: Early Exit via Reinforcement Learning in Reasoning Models
by: Dai, Muzhi, et al.
Published: (2025)

MathBode: Measuring the Stability of LLM Reasoning using Frequency Response
by: Wang, Charles L.
Published: (2025)

SARAD: LLM-Based Safety-Aware Hybrid Reinforcement Learning with Collision Prediction for Autonomous Driving
by: Wu, Kangyu, et al.
Published: (2026)

Debate with Images: Detecting Deceptive Behaviors in Multimodal Large Language Models
by: Fang, Sitong, et al.
Published: (2025)

Automating Deception: Scalable Multi-Turn LLM Jailbreaks
by: Kumarappan, Adarsh, et al.
Published: (2025)

Stable Reinforcement Learning for Efficient Reasoning
by: Dai, Muzhi, et al.
Published: (2025)

DecepChain: Inducing Deceptive Reasoning in Large Language Models
by: Shen, Wei, et al.
Published: (2025)

STAS: Spatial-Temporal Return Decomposition for Multi-agent Reinforcement Learning
by: Chen, Sirui, et al.
Published: (2023)

EntroPIC: Towards Stable Long-Term Training of LLMs via Entropy Stabilization with Proportional-Integral Control
by: Yang, Kai, et al.
Published: (2025)

Sequence to Sequence Reward Modeling: Improving RLHF by Language Feedback
by: Zhou, Jiayi, et al.
Published: (2024)

End-to-End Neuro-Symbolic Reinforcement Learning with Textual Explanations
by: Luo, Lirui, et al.
Published: (2024)

AI Deception: Risks, Dynamics, and Controls
by: Chen, Boyuan, et al.
Published: (2025)

Latent-Space Contrastive Reinforcement Learning for Stable and Efficient LLM Reasoning
by: Shan, Lianlei, et al.
Published: (2026)

The Unreasonable Effectiveness of Entropy Minimization in LLM Reasoning
by: Agarwal, Shivam, et al.
Published: (2025)

Boosting LLM Reasoning via Human-Inspired Reward Shaping
by: Lin, Wenze, et al.
Published: (2026)

ReasonBENCH: Benchmarking the (In)Stability of LLM Reasoning
by: Potamitis, Nearchos, et al.
Published: (2025)

Mitigating Non-IID Drift in Zeroth-Order Federated LLM Fine-Tuning with Transferable Sparsity
by: Ran, Yide, et al.
Published: (2025)

Dr. MAS: Stable Reinforcement Learning for Multi-Agent LLM Systems
by: Feng, Lang, et al.
Published: (2026)

Policy Guided Tree Search for Enhanced LLM Reasoning
by: Li, Yang
Published: (2025)

Adversarial Activation Patching: A Framework for Detecting and Mitigating Emergent Deception in Safety-Aligned Transformers
by: Ravindran, Santhosh Kumar
Published: (2025)

Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts
by: Wu, Zhaomin, et al.
Published: (2025)

Quantile Advantage Estimation: Stabilizing RLVR for LLM Reasoning
by: Wu, Junkang, et al.
Published: (2025)

Stream Aligner: Efficient Sentence-Level Alignment via Distribution Induction
by: Lou, Hantao, et al.
Published: (2025)