Mitigating Over-Refusal in Aligned Large Language Models via Inference-Time Activation Energy

Eric Hanchen Jiang, Weixuan Ou, Run Liu, Shengyuan Pang, Guancheng Wan, Ranjie Duan, Wei Dong, Kai-Wei Chang, XiaoFeng Wang, Ying Nian Wu, and Xinfeng Li, in ACL, 2026.

Download the full text

Abstract

Safety alignment of large language models currently faces a central challenge: existing alignment techniques often prioritize mitigating responses to harmful prompts at the expense of overcautious behavior, leading models to incorrectly refuse benign requests. A key goal of safe alignment is therefore to improve safety while simultaneously minimizing false refusals. In this work, we introduce Energy Landscape Steering (ELS), a novel, fine-tuning free framework designed to resolve this challenge through dynamic, inference-time intervention. We train a lightweight external Energy-Based Model (EBM) to assign high energy to undesirable states (false refusal or jailbreak) and low energy to desirable states (helpful response or safe reject). During inference, the EBM maps the LLM’s internal activations to an energy landscape, and we use the gradient of the energy function to steer the hidden states toward low-energy regions in real time. This dynamically guides the model toward desirable behavior without modifying its parameters. By decoupling behavioral control from the model’s core knowledge, ELS provides a flexible and computationally efficient solution. Extensive experiments across diverse models demonstrate its effectiveness, raising compliance on the ORB-H benchmark from 57.3 percent to 82.6 percent while maintaining baseline safety performance. Our work establishes a promising paradigm for building LLMs that simultaneously achieve high safety and low false refusal rates.

Bib Entry

@inproceedings{jiang2026overrefusal,
  title = {Mitigating Over-Refusal in Aligned Large Language Models via Inference-Time Activation Energy},
  author = {Jiang, Eric Hanchen and Ou, Weixuan and Liu, Run and Pang, Shengyuan and Wan, Guancheng and Duan, Ranjie and Dong, Wei and Chang, Kai-Wei and Wang, XiaoFeng and Wu, Ying Nian and Li, Xinfeng},
  booktitle = {ACL},
  year = {2026}
}

Related Publications

MM-PoisonRAG: Disrupting Multimodal RAG with Local and Global Knowledge Poisoning Attacks, ACL, 2026
SWAN: Semantic Watermarking with Abstract Meaning Representation, ACL, 2026
ARES: Adaptive Red-Teaming and End-to-End Repair of Policy-Reward System, ACL, 2026
Open-Domain Safety Policy Construction, EACL-Findings, 2026
Customize Multi-modal RAI Guardrails with Precedent-based predictions, COLM 2025, 2025
X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents, COLM 2025, 2025
Vulnerability of LLMs to Vertically Aligned Text Manipulations, ACL, 2025
Exploring Visual Vulnerabilities via Multi-Loss Adversarial Search for Jailbreaking Vision-Language Models, CVPR, 2025
Vulnerability of Large Language Models to Output Prefix Jailbreaks: Impact of Positions on Safety, NAACL-Finding, 2025
SafeWorld: Geo-Diverse Safety Alignment, NeurIPS, 2024
FLIRT: Feedback Loop In-context Red Teaming, EMNLP, 2024
Data Advisor: Data Curation with Foresight for Safety Alignment of Large Language Models, EMNLP, 2024
Prompt-Driven LLM Safeguarding via Directed Representation Optimization, ICML, 2024