Copertina

Decoding Answers Before Chain-of-Thought: Evidence from Pre-CoT Probes and Activation Steering

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Cox, Kyle, Kianersi, Darius, Garriga-Alonso, Adrià
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

SIM-CoT: Supervised Implicit Chain-of-Thought
di: Wei, Xilin, et al.
Pubblicazione: (2025)

Activation Steering for Chain-of-Thought Compression
di: Azizi, Seyedarmin, et al.
Pubblicazione: (2025)

CoT-Valve: Length-Compressible Chain-of-Thought Tuning
di: Ma, Xinyin, et al.
Pubblicazione: (2025)

SynthSAEBench: Evaluating Sparse Autoencoders on Scalable Realistic Synthetic Data
di: Chanin, David, et al.
Pubblicazione: (2026)

Among Us: A Sandbox for Measuring and Detecting Agentic Deception
di: Golechha, Satvik, et al.
Pubblicazione: (2025)

KAM-CoT: Knowledge Augmented Multimodal Chain-of-Thoughts Reasoning
di: Mondal, Debjyoti, et al.
Pubblicazione: (2024)

CoT Red-Handed: Stress Testing Chain-of-Thought Monitoring
di: Arnav, Benjamin, et al.
Pubblicazione: (2025)

Shorter Thoughts, Same Answers: Difficulty-Scaled Segment-Wise RL for CoT Compression
di: Tian, Ye, et al.
Pubblicazione: (2026)

CAP-CoT: Cycle Adversarial Prompt for Improving Chain of Thoughts in LLM Reasoning
di: Chen, Shuxu, et al.
Pubblicazione: (2026)

Chain-of-Sanitized-Thoughts: Plugging PII Leakage in CoT of Large Reasoning Models
di: Das, Arghyadeep, et al.
Pubblicazione: (2026)

CoT-X: An Adaptive Framework for Cross-Model Chain-of-Thought Transfer and Optimization
di: Bi, Ziqian, et al.
Pubblicazione: (2025)

To CoT or not to CoT? Chain-of-thought helps mainly on math and symbolic reasoning
di: Sprague, Zayne, et al.
Pubblicazione: (2024)

Sparse but Wrong: Incorrect L0 Leads to Incorrect Features in Sparse Autoencoders
di: Chanin, David, et al.
Pubblicazione: (2025)

CoT-BERT: Enhancing Unsupervised Sentence Representation through Chain-of-Thought
di: Zhang, Bowen, et al.
Pubblicazione: (2023)

Analyzing Chain of Thought (CoT) Approaches in Control Flow Code Deobfuscation Tasks
di: Mohseni, Seyedreza, et al.
Pubblicazione: (2026)

GTR-CoT: Graph Traversal as Visual Chain of Thought for Molecular Structure Recognition
di: Wang, Jingchao, et al.
Pubblicazione: (2025)

To CoT or To Loop? A Formal Comparison Between Chain-of-Thought and Looped Transformers
di: Xu, Kevin, et al.
Pubblicazione: (2025)

Deconstructing Long Chain-of-Thought: A Structured Reasoning Optimization Framework for Long CoT Distillation
di: Luo, Yijia, et al.
Pubblicazione: (2025)

Chain-of-Thought in Large Language Models: Decoding, Projection, and Activation
di: Yang, Hao, et al.
Pubblicazione: (2024)

Chain of Thoughtlessness? An Analysis of CoT in Planning
di: Stechly, Kaya, et al.
Pubblicazione: (2024)

VG-CoT: Towards Trustworthy Visual Reasoning via Grounded Chain-of-Thought
di: Lim, Byeonggeuk, et al.
Pubblicazione: (2026)

CoT-VLM4Tar: Chain-of-Thought Guided Vision-Language Models for Traffic Anomaly Resolution
di: Ren, Tianchi, et al.
Pubblicazione: (2025)

Diagnosing Harmful Continuation in Answer-Correct Long-CoT Training Traces
di: He, Chen, et al.
Pubblicazione: (2026)

Diffusion Language Models Know the Answer Before Decoding
di: Li, Pengxiang, et al.
Pubblicazione: (2025)

Cross Domain Evaluation of Multimodal Chain-of-Thought Reasoning of different datasets into the Amazon CoT Framework
di: Tiwari, Nitya, et al.
Pubblicazione: (2025)

TIBSTC-CoT: A Multi-Domain Instruction Dataset for Chain-of-Thought Reasoning in Language Models
di: Gao, Fan, et al.
Pubblicazione: (2025)

Feature Hedging: Correlated Features Break Narrow Sparse Autoencoders
di: Chanin, David, et al.
Pubblicazione: (2025)

How Chain-of-Thought Works? Tracing Information Flow from Decoding, Projection, and Activation
di: Yang, Hao, et al.
Pubblicazione: (2025)

CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models
di: Zhao, Qingqing, et al.
Pubblicazione: (2025)

CoT-Drive: Efficient Motion Forecasting for Autonomous Driving with LLMs and Chain-of-Thought Prompting
di: Liao, Haicheng, et al.
Pubblicazione: (2025)

CoT-RAG: Integrating Chain of Thought and Retrieval-Augmented Generation to Enhance Reasoning in Large Language Models
di: Li, Feiyang, et al.
Pubblicazione: (2025)

Latent Chain-of-Thought? Decoding the Depth-Recurrent Transformer
di: Lu, Wenquan, et al.
Pubblicazione: (2025)

CoT-ICL Lab: A Synthetic Framework for Studying Chain-of-Thought Learning from In-Context Demonstrations
di: Kothapalli, Vignesh, et al.
Pubblicazione: (2025)

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

Video-Skill-CoT: Skill-based Chain-of-Thoughts for Domain-Adaptive Video Reasoning
di: Lee, Daeun, et al.
Pubblicazione: (2025)

Hindsight Hint Distillation: Scaffolded Reasoning for SWE Agents from CoT-free Answers
di: Wang, Shengjie, et al.
Pubblicazione: (2026)

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

MME-CoT: Benchmarking Chain-of-Thought in Large Multimodal Models for Reasoning Quality, Robustness, and Efficiency
di: Jiang, Dongzhi, et al.
Pubblicazione: (2025)

Biases in the Blind Spot: Detecting What LLMs Fail to Mention
di: Arcuschin, Iván, et al.
Pubblicazione: (2026)

Path Channels and Plan Extension Kernels: a Mechanistic Description of Planning in a Sokoban RNN
di: Taufeeque, Mohammad, et al.
Pubblicazione: (2025)