Copertina

The Hypocrisy Gap: Quantifying Divergence Between Internal Belief and Chain-of-Thought Explanation via Sparse Autoencoders

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Shiromani, Shikhar, Chaudhury, Archie, Kunda, Sri Pranav
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Alignment is Localized: A Causal Probe into Preference Layers
di: Chaudhury, Archie
Pubblicazione: (2025)

How does Chain of Thought Think? Mechanistic Interpretability of Chain-of-Thought Reasoning with Sparse Autoencoding
di: Chen, Xi, et al.
Pubblicazione: (2025)

Beyond Next Word Prediction: Developing Comprehensive Evaluation Frameworks for measuring LLM performance on real world applications
di: Agrawal, Vishakha, et al.
Pubblicazione: (2025)

Chain of Thought Explanation for Dialogue State Tracking
di: Xu, Lin, et al.
Pubblicazione: (2024)

COMPASS: Context-Modulated PID Attention Steering System for Hallucination Mitigation
di: Sahay, Kenji, et al.
Pubblicazione: (2025)

Unlocking the Capabilities of Thought: A Reasoning Boundary Framework to Quantify and Optimize Chain-of-Thought
di: Chen, Qiguang, et al.
Pubblicazione: (2024)

Interpreting and Steering LLMs with Mutual Information-based Explanations on Sparse Autoencoders
di: Wu, Xuansheng, et al.
Pubblicazione: (2025)

Mind the Gap: Bridging Thought Leap for Improved Chain-of-Thought Tuning
di: Xu, Haolei, et al.
Pubblicazione: (2025)

ProMoral-Bench: Evaluating Prompting Strategies for Moral Reasoning and Safety in LLMs
di: Thomas, Rohan Subramanian, et al.
Pubblicazione: (2026)

Critical or Compliant? The Double-Edged Sword of Reasoning in Chain-of-Thought Explanations
di: Park, Eunkyu, et al.
Pubblicazione: (2025)

Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought
di: Boppana, Siddharth, et al.
Pubblicazione: (2026)

Sparse-Autoencoder-Guided Internal Representation Unlearning for Large Language Models
di: Yamashita, Tomoya, et al.
Pubblicazione: (2025)

Internalizing ASR with Implicit Chain of Thought for Efficient Speech-to-Speech Conversational LLM
di: Yuen, Robin Shing-Hei, et al.
Pubblicazione: (2024)

Quantifying Feature Space Universality Across Large Language Models via Sparse Autoencoders
di: Lan, Michael, et al.
Pubblicazione: (2024)

Visual Chain of Thought: Bridging Logical Gaps with Multimodal Infillings
di: Rose, Daniel, et al.
Pubblicazione: (2023)

A Formal Comparison Between Chain of Thought and Latent Thought
di: Xu, Kevin, et al.
Pubblicazione: (2025)

Mind the Gap: The Divergence Between Human and LLM-Generated Tasks
di: Lu, Yi-Long, et al.
Pubblicazione: (2025)

Mitigating Spurious Correlations Between Question and Answer via Chain-of-Thought Correctness Perception Distillation
di: Xie, Hongyan, et al.
Pubblicazione: (2025)

Internal Chain-of-Thought: Empirical Evidence for Layer-wise Subtask Scheduling in LLMs
di: Yang, Zhipeng, et al.
Pubblicazione: (2025)

How a Bilingual LM Becomes Bilingual: Tracing Internal Representations with Sparse Autoencoders
di: Inaba, Tatsuro, et al.
Pubblicazione: (2025)

Efficient Reasoning via Chain of Unconscious Thought
di: Gong, Ruihan, et al.
Pubblicazione: (2025)

A Survey on Sparse Autoencoders: Interpreting the Internal Mechanisms of Large Language Models
di: Shu, Dong, et al.
Pubblicazione: (2025)

GRACE: Generative Recommendation via Journey-Aware Sparse Attention on Chain-of-Thought Tokenization
di: Ma, Luyi, et al.
Pubblicazione: (2025)

From Sparse Dependence to Sparse Attention: Unveiling How Chain-of-Thought Enhances Transformer Sample Efficiency
di: Wen, Kaiyue, et al.
Pubblicazione: (2024)

Detecting Hallucinations in Large Language Models via Internal Attention Divergence Signals
di: van Dijk, Gijs
Pubblicazione: (2026)

Faithful Logical Reasoning via Symbolic Chain-of-Thought
di: Xu, Jundong, et al.
Pubblicazione: (2024)

Rethinking Dense Sequential Chains: Reasoning Language Models Can Extract Answers from Sparse, Order-Shuffling Chain-of-Thoughts
di: Chen, Yi-Chang, et al.
Pubblicazione: (2026)

SteerRM: Debiasing Reward Models via Sparse Autoencoders
di: Sun, Mengyuan, et al.
Pubblicazione: (2026)

Guiding LLM Post-training Data Engineering with Model Internals from Sparse Autoencoders
di: Jing, Yi, et al.
Pubblicazione: (2026)

Clause-Internal or Clause-External? Testing Turkish Reflexive Binding in Adapted versus Chain of Thought Large Language Models
di: Karakaş, Sercan
Pubblicazione: (2026)

State-of-the-Art Arabic Language Modeling with Sparse MoE Fine-Tuning and Chain-of-Thought Distillation
di: Singh, Navan Preet, et al.
Pubblicazione: (2026)

Belief Attribution as Mental Explanation: The Role of Accuracy, Informativity, and Causality
di: Ying, Lance, et al.
Pubblicazione: (2025)

Probing the Lack of Stable Internal Beliefs in LLMs
di: Luo, Yifan, et al.
Pubblicazione: (2026)

Reliable Chain-of-Thought via Prefix Consistency
di: Iwase, Naoto, et al.
Pubblicazione: (2026)

Self-Harmonized Chain of Thought
di: Jin, Ziqi, et al.
Pubblicazione: (2024)

Sparse Autoencoder Insights on Voice Embeddings
di: Pluth, Daniel, et al.
Pubblicazione: (2025)

Enhancing Uncertainty Estimation in LLMs with Expectation of Aggregated Internal Belief
di: Xiao, Zeguan, et al.
Pubblicazione: (2025)

Improving Chain-of-Thought Reasoning via Quasi-Symbolic Abstractions
di: Ranaldi, Leonardo, et al.
Pubblicazione: (2025)

Revisiting the Capacity Gap in Chain-of-Thought Distillation from a Practical Perspective
di: Kajitsuka, Tokio, et al.
Pubblicazione: (2026)

Upfront Chain-of-Thought: A Cooperative Framework for Chain-of-Thought Compression
di: Li, Chengzhengxu, et al.
Pubblicazione: (2025)