Cover Image

The Struggle Between Continuation and Refusal: A Mechanistic Analysis of the Continuation-Triggered Jailbreak in LLMs

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Deng, Yonghong, Yang, Zhen, Jian, Ping, Zhang, Xinyue, Guo, Zhongbin, Li, Chengzhi
Format:	Preprint
Published:	2026
Subjects:	Artificial Intelligence Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

How Do LLMs and VLMs Understand Viewpoint Rotation Without Vision? An Interpretability Study
by: Yang, Zhen, et al.
Published: (2026)

JailbreakRadar: Comprehensive Assessment of Jailbreak Attacks Against LLMs
by: Chu, Junjie, et al.
Published: (2024)

How Post-Training Reshapes LLMs: A Mechanistic View on Knowledge, Truthfulness, Refusal, and Confidence
by: Du, Hongzhe, et al.
Published: (2025)

Tracing the Dynamics of Refusal: Exploiting Latent Refusal Trajectories for Robust Jailbreak Detection
by: Hu, Xulin, et al.
Published: (2026)

Metis: Learning to Jailbreak LLMs via Self-Evolving Metacognitive Policy Optimization
by: Zhou, Huilin, et al.
Published: (2026)

A Closer Look at Adversarial Suffix Learning for Jailbreaking LLMs: Augmented Adversarial Trigger Learning
by: Wang, Zhe, et al.
Published: (2025)

Can LLMs See Without Pixels? Benchmarking Spatial Intelligence from Textual Descriptions
by: Guo, Zhongbin, et al.
Published: (2026)

TAMMs: Change Understanding and Forecasting in Satellite Image Time Series with Temporal-Aware Multimodal Models
by: Guo, Zhongbin, et al.
Published: (2025)

Furina: Fragmented Uncertainty-Driven Refusal Instability Attack
by: Wu, Tongxi, et al.
Published: (2026)

On the Implicit Adversariality of Catastrophic Forgetting in Deep Continual Learning
by: Peng, Ze, et al.
Published: (2025)

COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability
by: Guo, Xingang, et al.
Published: (2024)

What Drives Representation Steering? A Mechanistic Case Study on Steering Refusal
by: Cheng, Stephen, et al.
Published: (2026)

Jailbreaking LLMs via Calibration
by: Lu, Yuxuan, et al.
Published: (2026)

Federated Continual Instruction Tuning
by: Guo, Haiyang, et al.
Published: (2025)

Neural Predictive Control to Coordinate Discrete- and Continuous-Time Models for Time-Series Analysis with Control-Theoretical Improvements
by: Li, Haoran, et al.
Published: (2025)

A Systematic Investigation of The RL-Jailbreaker in LLMs
by: Mohammedalamen, Montaser, et al.
Published: (2026)

Continued AI Scaling Requires Repeated Efficiency Doublings
by: Lu, Chien-Ping
Published: (2026)

Does Refusal Training in LLMs Generalize to the Past Tense?
by: Andriushchenko, Maksym, et al.
Published: (2024)

Reasoning-targeted Jailbreak Attacks on Large Reasoning Models via Semantic Triggers and Psychological Framing
by: Wang, Zehao, et al.
Published: (2026)

Understanding Temporal Logic Consistency in Video-Language Models through Cross-Modal Attention Discriminability
by: Li, Chengzhi, et al.
Published: (2025)

Injecting Universal Jailbreak Backdoors into LLMs in Minutes
by: Chen, Zhuowei, et al.
Published: (2025)

Lost in Cultural Translation: Do LLMs Struggle with Math Across Cultural Contexts?
by: Karim, Aabid, et al.
Published: (2025)

Trusted Weights, Treacherous Optimizations? Optimization-Triggered Backdoor Attacks on LLMs
by: Wang, Yifei, et al.
Published: (2026)

SELF-[IN]CORRECT: LLMs Struggle with Discriminating Self-Generated Responses
by: Jiang, Dongwei, et al.
Published: (2024)

Not All Tokens Are Created Equal: Query-Efficient Jailbreak Fuzzing for LLMs
by: Chen, Wenyu, et al.
Published: (2026)

Continual Learning for Smart City: A Survey
by: Yang, Li, et al.
Published: (2024)

Self-Evolving LLMs via Continual Instruction Tuning
by: Kang, Jiazheng, et al.
Published: (2025)

CoEvo: Continual Evolution of Symbolic Solutions Using Large Language Models
by: Guo, Ping, et al.
Published: (2024)

Sparse Adapter Fusion for Continual Learning in NLP
by: Zeng, Min, et al.
Published: (2026)

Information-Theoretic Dual Memory System for Continual Learning
by: Wu, RunQing, et al.
Published: (2025)

Frontier LLMs Still Struggle with Simple Reasoning Tasks
by: Malek, Alan, et al.
Published: (2025)

The Struggles of LLMs in Cross-lingual Code Clone Detection
by: Moumoula, Micheline Bénédicte, et al.
Published: (2024)

Merge then Realign: Simple and Effective Modality-Incremental Continual Learning for Multimodal LLMs
by: Zhang, Dingkun, et al.
Published: (2025)

LibContinual: A Comprehensive Library towards Realistic Continual Learning
by: Li, Wenbin, et al.
Published: (2025)

Adaptive Budget Allocation for Orthogonal-Subspace Adapter Tuning in LLMs Continual Learning
by: Wan, Zhiyi, et al.
Published: (2025)

Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapes
by: Hu, Xiaomeng, et al.
Published: (2024)

Functional Homotopy: Smoothing Discrete Optimization via Continuous Parameters for LLM Jailbreak Attacks
by: Wang, Zi, et al.
Published: (2024)

Can LLMs Alleviate Catastrophic Forgetting in Graph Continual Learning? A Systematic Study
by: Cheng, Ziyang, et al.
Published: (2025)

Reflex: Reinforcement Learning with Reflection Symmetry Exploitation in State-Based Continuous Control
by: Zhen, Shuai, et al.
Published: (2026)

Continuous Approximations for Improving Quantization Aware Training of LLMs
by: Li, He, et al.
Published: (2024)