Copertina

From Thinking to Output: Chain-of-Thought and Text Generation Characteristics in Reasoning Language Models

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Liu, Junhao, Xu, Zhenhao, Fang, Yuxin, Chen, Yichuan, Ying, Zuobin, Chang, Wenhan
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Computation and Language Artificial Intelligence Cryptography and Security
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Safety Context Injection: Inference-Time Safety Alignment via Static Filtering and Agentic Analysis
di: Xu, Zhenhao, et al.
Pubblicazione: (2026)

Unreal Thinking: Chain-of-Thought Hijacking via Two-stage Backdoor
di: Chang, Wenhan, et al.
Pubblicazione: (2026)

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

BadThink: Triggered Overthinking Attacks on Chain-of-Thought Reasoning in Large Language Models
di: Liu, Shuaitong, et al.
Pubblicazione: (2025)

Chain-of-Lure: A Universal Jailbreak Attack Framework using Unconstrained Synthetic Narratives
di: Chang, Wenhan, et al.
Pubblicazione: (2025)

Autonomous Chain-of-Thought Distillation for Graph-Based Fraud Detection
di: Li, Yuan, et al.
Pubblicazione: (2026)

SALT: Steering Activations towards Leakage-free Thinking in Chain of Thought
di: Batra, Shourya, et al.
Pubblicazione: (2025)

A Character-based Diffusion Embedding Algorithm for Enhancing the Generation Quality of Generative Linguistic Steganographic Texts
di: Chen, Yingquan, et al.
Pubblicazione: (2025)

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

CoTGuard: Using Chain-of-Thought Triggering for Copyright Protection in Multi-Agent LLM Systems
di: Wen, Yan, et al.
Pubblicazione: (2025)

Leaky Thoughts: Large Reasoning Models Are Not Private Thinkers
di: Green, Tommaso, et al.
Pubblicazione: (2025)

GenBreak: Red Teaming Text-to-Image Generators Using Large Language Models
di: Wang, Zilong, et al.
Pubblicazione: (2025)

From Trade-off to Synergy: A Versatile Symbiotic Watermarking Framework for Large Language Models
di: Wang, Yidan, et al.
Pubblicazione: (2025)

R-CoT: A Reasoning-Layer Watermark via Redundant Chain-of-Thought in Large Language Models
di: Zhang, Ziming, et al.
Pubblicazione: (2026)

Bileve: Securing Text Provenance in Large Language Models Against Spoofing with Bi-level Signature
di: Zhou, Tong, et al.
Pubblicazione: (2024)

Reasoning-Oriented Programming: Chaining Semantic Gadgets to Jailbreak Large Vision Language Models
di: Zou, Quanchen, et al.
Pubblicazione: (2026)

Chain-of-Thought Prompting of Large Language Models for Discovering and Fixing Software Vulnerabilities
di: Nong, Yu, et al.
Pubblicazione: (2024)

Output Supervision Can Obfuscate the Chain of Thought
di: Drori, Jacob, et al.
Pubblicazione: (2025)

Thought-Transfer: Indirect Targeted Poisoning Attacks on Chain-of-Thought Reasoning Models
di: Chaudhari, Harsh, et al.
Pubblicazione: (2026)

Beyond Text: Unveiling Privacy Vulnerabilities in Multi-modal Retrieval-Augmented Generation
di: Zhang, Jiankun, et al.
Pubblicazione: (2025)

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

Lightweight Yet Secure: Secure Scripting Language Generation via Lightweight LLMs
di: Zhang, Keyang, et al.
Pubblicazione: (2026)

Simulate and Eliminate: Revoke Backdoors for Generative Large Language Models
di: Li, Haoran, et al.
Pubblicazione: (2024)

LLM-based Privacy Data Augmentation Guided by Knowledge Distillation with a Distribution Tutor for Medical Text Classification
di: Song, Yiping, et al.
Pubblicazione: (2024)

GuardReasoner-Omni: A Reasoning-based Multi-modal Guardrail for Text, Image, Video, and Audio
di: Zhu, Zhenhao, et al.
Pubblicazione: (2026)

Text Embedding Inversion Security for Multilingual Language Models
di: Chen, Yiyi, et al.
Pubblicazione: (2024)

Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations
di: Wei, Zeming, et al.
Pubblicazione: (2023)

Yet Another Watermark for Large Language Models
di: Bao, Siyuan, et al.
Pubblicazione: (2025)

Can Reasoning Models Obfuscate Reasoning? Stress-Testing Chain-of-Thought Monitorability
di: Zolkowski, Artur, et al.
Pubblicazione: (2025)

A General Pseudonymization Framework for Cloud-Based LLMs: Replacing Privacy Information in Controlled Text Generation
di: Hou, Shilong, et al.
Pubblicazione: (2025)

GradEscape: A Gradient-Based Evader Against AI-Generated Text Detectors
di: Meng, Wenlong, et al.
Pubblicazione: (2025)

ExplainableGuard: Interpretable Adversarial Defense for Large Language Models Using Chain-of-Thought Reasoning
di: Guan, Shaowei, et al.
Pubblicazione: (2025)

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

BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models
di: Xiang, Zhen, et al.
Pubblicazione: (2024)

StructuralSleight: Automated Jailbreak Attacks on Large Language Models Utilizing Uncommon Text-Organization Structures
di: Li, Bangxin, et al.
Pubblicazione: (2024)

DualGuard: Dual-stream Large Language Model Watermarking Defense against Paraphrase and Spoofing Attack
di: Li, Hao, et al.
Pubblicazione: (2025)

Time Will Tell: Timing Side Channels via Output Token Count in Large Language Models
di: Zhang, Tianchen, et al.
Pubblicazione: (2024)

BiAxisAudit: A Novel Framework to Evaluate LLM Bias Across Prompt Sensitivity and Response-Layer Divergence
di: Gan, Jialing, et al.
Pubblicazione: (2026)

The Surprising Universality of LLM Outputs: A Real-Time Verification Primitive
di: Bogdan, Alex, et al.
Pubblicazione: (2026)

InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents
di: Zhan, Qiusi, et al.
Pubblicazione: (2024)