Cover Image

Monitoring the Internal Monologue: Probe Trajectories Reveal Reasoning Dynamics

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Chrabąszcz, Maciej, Szymczyk, Aleksander, Sendera, Marcin, Trzciński, Tomasz, Cygert, Sebastian
Format:	Preprint
Published:	2026
Subjects:	Computation and Language Cryptography and Security
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Efficient LLM Moderation with Multi-Layer Latent Prototypes
by: Chrabąszcz, Maciej, et al.
Published: (2025)

Conditioned Activation Transport for T2I Safety Steering
by: Chrabąszcz, Maciej, et al.
Published: (2026)

Watermarking LLM Agent Trajectories
by: Meng, Wenlong, et al.
Published: (2026)

Conti Inc.: Understanding the Internal Discussions of a large Ransomware-as-a-Service Operator with Machine Learning
by: Ruellan, Estelle, et al.
Published: (2023)

DINA: A Dual Defense Framework Against Internal Noise and External Attacks in Natural Language Processing
by: Chuang, Ko-Wei, et al.
Published: (2025)

Expert Selections In MoE Models Reveal (Almost) As Much As Text
by: Nuriyev, Amir, et al.
Published: (2026)

Segment-Level Coherence for Robust Harmful Intent Probing in LLMs
by: He, Xuanli, et al.
Published: (2026)

garak: A Framework for Security Probing Large Language Models
by: Derczynski, Leon, et al.
Published: (2024)

Jailbreaking Leaves a Trace: Understanding and Detecting Jailbreak Attacks from Internal Representations of Large Language Models
by: Kadali, Sri Durga Sai Sowmya, et al.
Published: (2026)

Factor(T,U): Factored Cognition Strengthens Monitoring of Untrusted AI
by: Sandoval, Aaron, et al.
Published: (2025)

Subversion via Focal Points: Investigating Collusion in LLM Monitoring
by: Järviniemi, Olli
Published: (2025)

Beyond Jailbreaks: Revealing Stealthier and Broader LLM Security Risks Stemming from Alignment Failures
by: Zhou, Yukai, et al.
Published: (2025)

Self-Jailbreaking: Language Models Can Reason Themselves Out of Safety Alignment After Benign Reasoning Training
by: Yong, Zheng-Xin, et al.
Published: (2025)

SecureGate: Learning When to Reveal PII Safely via Token-Gated Dual-Adapters for Federated LLMs
by: Shaaban, Mohamed, et al.
Published: (2026)

Probe before You Talk: Towards Black-box Defense against Backdoor Unalignment for Large Language Models
by: Yi, Biao, et al.
Published: (2025)

Bag of Tricks for Subverting Reasoning-based Safety Guardrails
by: Chen, Shuo, et al.
Published: (2025)

Do Reasoning LLMs Refuse What They Infer in Long Contexts?
by: Fu, Yu, et al.
Published: (2026)

ShadowCoT: Cognitive Hijacking for Stealthy Reasoning Backdoors in LLMs
by: Zhao, Gejian, et al.
Published: (2025)

BAIT: Boundary-Guided Disclosure Escalation via Self-Conditioned Reasoning
by: Luo, Xuan, et al.
Published: (2026)

MIRAGE: Multimodal Immersive Reasoning and Guided Exploration for Red-Team Jailbreak Attacks
by: You, Wenhao, et al.
Published: (2025)

Internal Safety Collapse in Frontier Large Language Models
by: Wu, Yutao, et al.
Published: (2026)

Chain-of-Code Collapse: Reasoning Failures in LLMs via Adversarial Prompting in Code Generation
by: Roh, Jaechul, et al.
Published: (2025)

Safety is Not Only About Refusal: Reasoning-Enhanced Fine-tuning for Interpretable LLM Safety
by: Zhang, Yuyou, et al.
Published: (2025)

From Retrieval to Reasoning: A Framework for Cyber Threat Intelligence NER with Explicit and Adaptive Instructions
by: Peng, Jiaren, et al.
Published: (2025)

Adversarial Robustness through Dynamic Ensemble Learning
by: Waghela, Hetvi, et al.
Published: (2024)

Adversarial Text Generation with Dynamic Contextual Perturbation
by: Waghela, Hetvi, et al.
Published: (2025)

Provable Secure Steganography Based on Adaptive Dynamic Sampling
by: Pang, Kaiyi, et al.
Published: (2025)

LRCTI: A Large Language Model-Based Framework for Multi-Step Evidence Retrieval and Reasoning in Cyber Threat Intelligence Credibility Verification
by: Tang, Fengxiao, et al.
Published: (2025)

SEEP: Training Dynamics Grounds Latent Representation Search for Mitigating Backdoor Poisoning Attacks
by: He, Xuanli, et al.
Published: (2024)

ScaleOT: Privacy-utility-scalable Offsite-tuning with Dynamic LayerReplace and Selective Rank Compression
by: Yao, Kai, et al.
Published: (2024)

TempCharBERT: Keystroke Dynamics for Continuous Access Control Based on Pre-trained Language Models
by: Simão, Matheus, et al.
Published: (2024)

Sentinels of the Stream: Unleashing Large Language Models for Dynamic Packet Classification in Software Defined Networks -- Position Paper
by: Murtuza, Shariq
Published: (2024)

$PD^3F$: A Pluggable and Dynamic DoS-Defense Framework Against Resource Consumption Attacks Targeting Large Language Models
by: Zhang, Yuanhe, et al.
Published: (2025)

T-MAP: Red-Teaming LLM Agents with Trajectory-aware Evolutionary Search
by: Lee, Hyomin, et al.
Published: (2026)

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

Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning
by: Ahmed, Nesreen K., et al.
Published: (2026)

The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs
by: Berezin, Sergey, et al.
Published: (2025)

Measuring Copyright Risks of Large Language Model via Partial Information Probing
by: Zhao, Weijie, et al.
Published: (2024)

Preemptive Answer "Attacks" on Chain-of-Thought Reasoning
by: Xu, Rongwu, et al.
Published: (2024)

Large Reasoning Models Are Autonomous Jailbreak Agents
by: Hagendorff, Thilo, et al.
Published: (2025)