Cover Image

Just Enough Shifts: Mitigating Over-Refusal in Aligned Language Models with Targeted Representation Fine-Tuning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Dabas, Mahavir, Chen, Si, Fleming, Charles, Jin, Ming, Jia, Ruoxi
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

Characterizing Model-Native Skills
by: Kang, Feiyang, et al.
Published: (2026)

DiPT: Enhancing LLM reasoning through diversified perspective-taking
by: Just, Hoang Anh, et al.
Published: (2024)

Mitigating Over-Refusal in Aligned Large Language Models via Inference-Time Activation Energy
by: Jiang, Eric Hanchen, et al.
Published: (2025)

Memory-Induced Tool-Drift in LLM Agents
by: Dabas, Mahavir, et al.
Published: (2026)

Probing Knowledge Holes in Unlearned LLMs
by: Ko, Myeongseob, et al.
Published: (2025)

Do Language Models Align with Brains? Prediction Scores Are Not Enough
by: Jia, Xiao
Published: (2026)

Linearly Decoding Refused Knowledge in Aligned Language Models
by: Shrivastava, Aryan, et al.
Published: (2025)

Retracing the Past: LLMs Emit Training Data When They Get Lost
by: Ko, Myeongseob, et al.
Published: (2025)

OR-Bench: An Over-Refusal Benchmark for Large Language Models
by: Cui, Justin, et al.
Published: (2024)

Think Before Refusal : Triggering Safety Reflection in LLMs to Mitigate False Refusal Behavior
by: Si, Shengyun, et al.
Published: (2025)

SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal
by: Xie, Tinghao, et al.
Published: (2024)

EVOREFUSE: Evolutionary Prompt Optimization for Evaluation and Mitigation of LLM Over-Refusal to Pseudo-Malicious Instructions
by: Wu, Xiaorui, et al.
Published: (2025)

Don't Just Fine-tune the Agent, Tune the Environment
by: Lu, Siyuan, et al.
Published: (2025)

Fine-Tuning and Deploying Large Language Models Over Edges: Issues and Approaches
by: Dong, Yanjie, et al.
Published: (2024)

State-Dependent Refusal and Learned Incapacity in RLHF-Aligned Language Models
by: Lee, TK
Published: (2025)

Restoring Calibration for Aligned Large Language Models: A Calibration-Aware Fine-Tuning Approach
by: Xiao, Jiancong, et al.
Published: (2025)

LLMs Can Plan Only If We Tell Them
by: Sel, Bilgehan, et al.
Published: (2025)

Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models
by: Sel, Bilgehan, et al.
Published: (2023)

Accidental Vulnerability: Factors in Fine-Tuning that Shift Model Safeguards
by: Pandey, Punya Syon, et al.
Published: (2025)

Safety at One Shot: Patching Fine-Tuned LLMs with A Single Instance
by: Zhang, Jiawen, et al.
Published: (2026)

Noise Augmented Fine Tuning for Mitigating Hallucinations in Large Language Models
by: Khadangi, Afshin, et al.
Published: (2025)

Surgery: Mitigating Harmful Fine-Tuning for Large Language Models via Attention Sink
by: Liu, Guozhi, et al.
Published: (2026)

Parameter-Efficient Fine-Tuning of Large Language Models via Deconvolution in Subspace
by: Zhang, Jia-Chen, et al.
Published: (2025)

Natural Language Fine-Tuning
by: Liu, Jia, et al.
Published: (2024)

Understanding and Preserving Safety in Fine-Tuned LLMs
by: Zhang, Jiawen, et al.
Published: (2026)

Deactivating Refusal Triggers: Understanding and Mitigating Overrefusal in Safety Alignment
by: Xue, Zhiyu, et al.
Published: (2026)

LFTF: Locating First and Then Fine-Tuning for Mitigating Gender Bias in Large Language Models
by: Qin, Zhanyue, et al.
Published: (2025)

Assessing and Mitigating Data Memorization Risks in Fine-Tuned Large Language Models
by: Ramakrishnan, Badrinath, et al.
Published: (2025)

Improved Supervised Fine-Tuning for Large Language Models to Mitigate Catastrophic Forgetting
by: Ding, Fei, et al.
Published: (2025)

Forget the Data and Fine-Tuning! Just Fold the Network to Compress
by: Wang, Dong, et al.
Published: (2025)

Learning to Refuse: Towards Mitigating Privacy Risks in LLMs
by: Liu, Zhenhua, et al.
Published: (2024)

Learning to Decide with Just Enough: Information-Theoretic Context Summarization for CMDPs
by: Liu, Peidong, et al.
Published: (2025)

Blind Refusal: Language Models Refuse to Help Users Evade Unjust, Absurd, and Illegitimate Rules
by: Pattison, Cameron, et al.
Published: (2026)

Aligning Backchannel and Dialogue Context Representations via Contrastive LLM Fine-Tuning
by: Qian, Livia, et al.
Published: (2026)

FalseReject: A Resource for Improving Contextual Safety and Mitigating Over-Refusals in LLMs via Structured Reasoning
by: Zhang, Zhehao, et al.
Published: (2025)

RaVL: Discovering and Mitigating Spurious Correlations in Fine-Tuned Vision-Language Models
by: Varma, Maya, et al.
Published: (2024)

Align is not Enough: Multimodal Universal Jailbreak Attack against Multimodal Large Language Models
by: Wang, Youze, et al.
Published: (2025)

FASTTRACK: Fast and Accurate Fact Tracing for LLMs
by: Chen, Si, et al.
Published: (2024)

The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence
by: Wollschläger, Tom, et al.
Published: (2025)

Membership Inference Attacks Against Fine-tuned Diffusion Language Models
by: Chen, Yuetian, et al.
Published: (2026)