Imagen de Portada

Finding RELIEF: Shaping Reasoning Behavior without Reasoning Supervision via Belief Engineering

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Leong, Chak Tou, Chen, Dingwei, Xia, Heming, Yin, Qingyu, Lee, Sunbowen, Wang, Jian, Li, Wenjie
Formato:	Preprint
Publicado:	2026
Materias:	Artificial Intelligence Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Why Safeguarded Ships Run Aground? Aligned Large Language Models' Safety Mechanisms Tend to Be Anchored in The Template Region
por: Leong, Chak Tou, et al.
Publicado: (2025)

Seeing Isn't Believing: Mitigating Belief Inertia via Active Intervention in Embodied Agents
por: Wang, Hanlin, et al.
Publicado: (2026)

TokenSkip: Controllable Chain-of-Thought Compression in LLMs
por: Xia, Heming, et al.
Publicado: (2025)

Probing the Difficulty Perception Mechanism of Large Language Models
por: Lee, Sunbowen, et al.
Publicado: (2025)

Scaling over Scaling: Exploring Test-Time Scaling Plateau in Large Reasoning Models
por: Wang, Jian, et al.
Publicado: (2025)

Merlin's Whisper: Enabling Efficient Reasoning in Large Language Models via Black-box Persuasive Prompting
por: Xia, Heming, et al.
Publicado: (2025)

E2CL: Exploration-based Error Correction Learning for Embodied Agents
por: Wang, Hanlin, et al.
Publicado: (2024)

STeCa: Step-level Trajectory Calibration for LLM Agent Learning
por: Wang, Hanlin, et al.
Publicado: (2025)

Induction Head Toxicity Mechanistically Explains Repetition Curse in Large Language Models
por: Wang, Shuxun, et al.
Publicado: (2025)

Subtle Errors in Reasoning: Preference Learning via Error-injected Self-editing
por: Xu, Kaishuai, et al.
Publicado: (2024)

KNN-SSD: Enabling Dynamic Self-Speculative Decoding via Nearest Neighbor Layer Set Optimization
por: Song, Mingbo, et al.
Publicado: (2025)

Instruct Once, Chat Consistently in Multiple Rounds: An Efficient Tuning Framework for Dialogue
por: Wang, Jian, et al.
Publicado: (2024)

Constrain Alignment with Sparse Autoencoders
por: Yin, Qingyu, et al.
Publicado: (2024)

SPA-RL: Reinforcing LLM Agents via Stepwise Progress Attribution
por: Wang, Hanlin, et al.
Publicado: (2025)

Refusal Falls off a Cliff: How Safety Alignment Fails in Reasoning?
por: Yin, Qingyu, et al.
Publicado: (2025)

Understanding Inter-Session Intentions via Complex Logical Reasoning
por: Bai, Jiaxin, et al.
Publicado: (2023)

Reasoning Models Hallucinate More: Factuality-Aware Reinforcement Learning for Large Reasoning Models
por: Li, Junyi, et al.
Publicado: (2025)

Mitigating Unhelpfulness in Emotional Support Conversations with Multifaceted AI Feedback
por: Wang, Jiashuo, et al.
Publicado: (2024)

AutoPal: Autonomous Adaptation to Users for Personal AI Companionship
por: Cheng, Yi, et al.
Publicado: (2024)

Fantastic Reasoning Behaviors and Where to Find Them: Unsupervised Discovery of the Reasoning Process
por: Zhang, Zhenyu, et al.
Publicado: (2025)

No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks
por: Leong, Chak Tou, et al.
Publicado: (2024)

Verbalized Confidence Triggers Self-Verification: Emergent Behavior Without Explicit Reasoning Supervision
por: Jang, Chaeyun, et al.
Publicado: (2025)

K-Level Reasoning: Establishing Higher Order Beliefs in Large Language Models for Strategic Reasoning
por: Zhang, Yadong, et al.
Publicado: (2024)

Enhancing Tool Retrieval with Iterative Feedback from Large Language Models
por: Xu, Qiancheng, et al.
Publicado: (2024)

Towards Robust Temporal Reasoning of Large Language Models via a Multi-Hop QA Dataset and Pseudo-Instruction Tuning
por: Tan, Qingyu, et al.
Publicado: (2023)

From Query to Logic: Ontology-Driven Multi-Hop Reasoning in LLMs
por: Bian, Haonan, et al.
Publicado: (2025)

Reasoning Beyond Language: A Comprehensive Survey on Latent Chain-of-Thought Reasoning
por: Chen, Xinghao, et al.
Publicado: (2025)

Deeper Insights Without Updates: The Power of In-Context Learning Over Fine-Tuning
por: Yin, Qingyu, et al.
Publicado: (2024)

BaRDa: A Belief and Reasoning Dataset that Separates Factual Accuracy and Reasoning Ability
por: Clark, Peter, et al.
Publicado: (2023)

MatchTIR: Fine-Grained Supervision for Tool-Integrated Reasoning via Bipartite Matching
por: Qu, Changle, et al.
Publicado: (2026)

Reasoning Like a Doctor: Improving Medical Dialogue Systems via Diagnostic Reasoning Process Alignment
por: Xu, Kaishuai, et al.
Publicado: (2024)

Finding Answers in Thought Matters: Revisiting Evaluation on Large Language Models with Reasoning
por: Jo, Hwiyeol, et al.
Publicado: (2025)

Process Supervision via Verbal Critique Improves Reasoning in Large Language Models
por: Chen, Hao-Yuan
Publicado: (2026)

Efficient Mathematical Reasoning Models via Dynamic Pruning and Knowledge Distillation
por: Yu, Fengming, et al.
Publicado: (2025)

Expanding before Inferring: Enhancing Factuality in Large Language Models through Premature Layers Interpolation
por: Chen, Dingwei, et al.
Publicado: (2025)

HauntAttack: When Attack Follows Reasoning as a Shadow
por: Ma, Jingyuan, et al.
Publicado: (2025)

Reasoning Pattern Matters: Learning to Reason without Human Rationales
por: Pang, Chaoxu, et al.
Publicado: (2025)

Code Execution as Grounded Supervision for LLM Reasoning
por: Jung, Dongwon, et al.
Publicado: (2025)

Mitigating Overthinking through Reasoning Shaping
por: Song, Feifan, et al.
Publicado: (2025)

BIS Reasoning 1.0: The First Large-Scale Japanese Benchmark for Belief-Inconsistent Syllogistic Reasoning
por: Nguyen, Ha-Thanh, et al.
Publicado: (2025)