Cover Image

EVOREFUSE: Evolutionary Prompt Optimization for Evaluation and Mitigation of LLM Over-Refusal to Pseudo-Malicious Instructions

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Wu, Xiaorui, Li, Fei, Mao, Xiaofeng, Zhang, Xin, Zheng, Li, Peng, Yuxiang, Teng, Chong, Ji, Donghong, Li, Zhuang
Format:	Preprint
Published:	2025
Subjects:	Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

TRIDENT: Enhancing Large Language Model Safety with Tri-Dimensional Diversified Red-Teaming Data Synthesis
by: Wu, Xiaorui, et al.
Published: (2025)

Are Emotion and Rhetoric Neurons in LLM? Neuron Recognition and Adaptive Masking for Emotion-Rhetoric Prediction Steering
by: Zheng, Li, et al.
Published: (2026)

DiscoSG: Towards Discourse-Level Text Scene Graph Parsing through Iterative Graph Refinement
by: Lin, Shaoqing, et al.
Published: (2025)

Revisiting Structured Sentiment Analysis as Latent Dependency Graph Parsing
by: Zhou, Chengjie, et al.
Published: (2024)

CMNER: A Chinese Multimodal NER Dataset based on Social Media
by: Ji, Yuanze, et al.
Published: (2024)

Enhancing Hyperbole and Metaphor Detection with Their Bidirectional Dynamic Interaction and Emotion Knowledge
by: Zheng, Li, et al.
Published: (2025)

Multi-Granular Multimodal Clue Fusion for Meme Understanding
by: Zheng, Li, et al.
Published: (2025)

PaSE: Prototype-aligned Calibration and Shapley-based Equilibrium for Multimodal Sentiment Analysis
by: He, Kang, et al.
Published: (2025)

Enhance-then-Balance Modality Collaboration for Robust Multimodal Sentiment Analysis
by: He, Kang, et al.
Published: (2026)

DALR: Dual-level Alignment Learning for Multimodal Sentence Representation Learning
by: He, Kang, et al.
Published: (2025)

Modeling Unified Semantic Discourse Structure for High-quality Headline Generation
by: Xu, Minghui, et al.
Published: (2024)

Harvesting Events from Multiple Sources: Towards a Cross-Document Event Extraction Paradigm
by: Gao, Qiang, et al.
Published: (2024)

Dynamic Emotion and Personality Profiling for Multimodal Deception Detection
by: Zheng, Li, et al.
Published: (2026)

Code-MIE: A Code-style Model for Multimodal Information Extraction with Scene Graph and Entity Attribute Knowledge Enhancement
by: Liu, Jiang, et al.
Published: (2026)

Enhancing Cross-Document Event Coreference Resolution by Discourse Structure and Semantic Information
by: Gao, Qiang, et al.
Published: (2024)

Zero-Shot Conversational Stance Detection: Dataset and Approaches
by: Ding, Yuzhe, et al.
Published: (2025)

Automatic Pseudo-Harmful Prompt Generation for Evaluating False Refusals in Large Language Models
by: An, Bang, et al.
Published: (2024)

Enhancing LLM Instruction Following: An Evaluation-Driven Multi-Agentic Workflow for Prompt Instructions Optimization
by: Purpura, Alberto, et al.
Published: (2026)

Beyond Over-Refusal: Scenario-Based Diagnostics and Post-Hoc Mitigation for Exaggerated Refusals in LLMs
by: Yuan, Shuzhou, et al.
Published: (2025)

Mitigating Indirect Prompt Injection via Instruction-Following Intent Analysis
by: Kang, Mintong, et al.
Published: (2025)

Optimizing Instruction Synthesis: Effective Exploration of Evolutionary Space with Tree Search
by: Li, Chenglin, et al.
Published: (2024)

Refusal Evaluation in Coding LLMs and Code Agents: A Systematic Review of Thirteen Malicious-Code Prompt Corpora (2023-2025)
by: Young, Richard J., et al.
Published: (2026)

Cross-Document Cross-Lingual NLI via RST-Enhanced Graph Fusion and Interpretability Prediction
by: Yuan, Mengying, et al.
Published: (2025)

Consistency Assessment of CORDEX Multi‐Domain Simulations Over the Tibetan Plateau Using REMO
by: Ping Li, et al.
Published: (2025)

Evolve the Method, Not the Prompts: Evolutionary Synthesis of Jailbreak Attacks on LLMs
by: Chen, Yunhao, et al.
Published: (2025)

Feature-Aware Malicious Output Detection and Mitigation
by: Dong, Weilong, et al.
Published: (2025)

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

Smoke and Mirrors: Jailbreaking LLM-based Code Generation via Implicit Malicious Prompts
by: Ouyang, Sheng, et al.
Published: (2025)

InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models
by: Li, Hao, et al.
Published: (2024)

Should LLM Safety Be More Than Refusing Harmful Instructions?
by: Maskey, Utsav, et al.
Published: (2025)

GRAIT: Gradient-Driven Refusal-Aware Instruction Tuning for Effective Hallucination Mitigation
by: Zhu, Runchuan, et al.
Published: (2025)

The Landscape of Prompt Injection Threats in LLM Agents: From Taxonomy to Analysis
by: Wang, Peiran, et al.
Published: (2026)

SafeConstellations: Mitigating Over-Refusals in LLMs Through Task-Aware Representation Steering
by: Maskey, Utsav, et al.
Published: (2025)

ORFuzz: Fuzzing the "Other Side" of LLM Safety -- Testing Over-Refusal
by: Zhang, Haonan, et al.
Published: (2025)

Invisible Prompts, Visible Threats: Malicious Font Injection in External Resources for Large Language Models
by: Xiong, Junjie, et al.
Published: (2025)

Analysing Safety Risks in LLMs Fine-Tuned with Pseudo-Malicious Cyber Security Data
by: ElZemity, Adel, et al.
Published: (2025)

OVERT: A Benchmark for Over-Refusal Evaluation on Text-to-Image Models
by: Cheng, Ziheng, et al.
Published: (2025)

Mitigating Malicious Attacks in Federated Learning via Confidence-aware Defense
by: Li, Qilei, et al.
Published: (2024)

Discern Truth from Falsehood: Reducing Over-Refusal via Contrastive Refinement
by: Lu, Yuxiao, et al.
Published: (2026)

Hear No Evil: Detecting Gradient Leakage by Malicious Servers in Federated Learning
by: Wang, Fei, et al.
Published: (2025)