Imagen de Portada

Are Reasoning LLMs Robust to Interventions on Their Chain-of-Thought?

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	von Recum, Alexander, Girrbach, Leander, Akata, Zeynep
Formato:	Preprint
Publicado:	2026
Materias:	Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Do LLMs Experience an Internal Polylogue? Investigating Reasoning through the Lens of Personas
por: Herrmann, Nils A., et al.
Publicado: (2026)

SUB: Benchmarking CBM Generalization via Synthetic Attribute Substitutions
por: Bader, Jessica, et al.
Publicado: (2025)

Sparse Autoencoders are Topic Models
por: Girrbach, Leander, et al.
Publicado: (2025)

APM: Evaluating Style Personalization in LLMs with Arbitrary Preference Mappings
por: Spohn, Philipp, et al.
Publicado: (2026)

Reference-Free Rating of LLM Responses via Latent Information
por: Girrbach, Leander, et al.
Publicado: (2025)

DeLoRA: Decoupling Angles and Strength in Low-rank Adaptation
por: Bini, Massimo, et al.
Publicado: (2025)

A Systematic Study of In-the-Wild Model Merging for Large Language Models
por: Hitit, Oğuz Kağan, et al.
Publicado: (2025)

A Large Scale Analysis of Gender Biases in Text-to-Image Generative Models
por: Girrbach, Leander, et al.
Publicado: (2025)

Align-then-Unlearn: Embedding Alignment for LLM Unlearning
por: Spohn, Philipp, et al.
Publicado: (2025)

Cannot or Should Not? Automatic Analysis of Refusal Composition in IFT/RLHF Datasets and Refusal Behavior of Black-Box LLMs
por: von Recum, Alexander, et al.
Publicado: (2024)

Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Models
por: Girrbach, Leander, et al.
Publicado: (2025)

Revealing and Reducing Gender Biases in Vision and Language Assistants (VLAs)
por: Girrbach, Leander, et al.
Publicado: (2024)

Improving Chain-of-Thought for Logical Reasoning via Attention-Aware Intervention
por: Phuong, Nguyen Minh, et al.
Publicado: (2026)

Improving Intervention Efficacy via Concept Realignment in Concept Bottleneck Models
por: Singhi, Nishad, et al.
Publicado: (2024)

Thoughts without Thinking: Reconsidering the Explanatory Value of Chain-of-Thought Reasoning in LLMs through Agentic Pipelines
por: Manuvinakurike, Ramesh, et al.
Publicado: (2025)

LLM Reasoning Is Latent, Not the Chain of Thought
por: Wang, Wenshuo
Publicado: (2026)

Draft-Thinking: Learning Efficient Reasoning in Long Chain-of-Thought LLMs
por: Cao, Jie, et al.
Publicado: (2026)

LLMs can Find Mathematical Reasoning Mistakes by Pedagogical Chain-of-Thought
por: Jiang, Zhuoxuan, et al.
Publicado: (2024)

SemioLLM: Evaluating Large Language Models for Diagnostic Reasoning from Unstructured Clinical Narratives in Epilepsy
por: Dani, Meghal, et al.
Publicado: (2024)

DICE: Structured Reasoning in LLMs through SLM-Guided Chain-of-Thought Correction
por: Li, Yiqi, et al.
Publicado: (2025)

Direct Value Optimization: Improving Chain-of-Thought Reasoning in LLMs with Refined Values
por: Zhang, Hongbo, et al.
Publicado: (2025)

AdvChain: Adversarial Chain-of-Thought Tuning for Robust Safety Alignment of Large Reasoning Models
por: Zhu, Zihao, et al.
Publicado: (2025)

Diagnosing Pathological Chain-of-Thought in Reasoning Models
por: Liu, Manqing, et al.
Publicado: (2026)

Reasoning Models Struggle to Control their Chains of Thought
por: Yueh-Han, Chen, et al.
Publicado: (2026)

Feasibility with Language Models for Open-World Compositional Zero-Shot Learning
por: Kim, Jae Myung, et al.
Publicado: (2025)

From Drop-off to Recovery: A Mechanistic Analysis of Segmentation in MLLMs
por: Wu, Boyong, et al.
Publicado: (2026)

Chain-of-Thought Degrades Visual Spatial Reasoning Capabilities of Multimodal LLMs
por: Kancheti, Sai Srinivas, et al.
Publicado: (2026)

Discovering Chunks in Neural Embeddings for Interpretability
por: Wu, Shuchen, et al.
Publicado: (2025)

Latent Chain-of-Thought for Visual Reasoning
por: Sun, Guohao, et al.
Publicado: (2025)

Mechanics of Bias and Reasoning: Interpreting the Impact of Chain-of-Thought Prompting on Gender Bias in LLMs
por: Pearman, Edie, et al.
Publicado: (2026)

Towards System 2 Reasoning in LLMs: Learning How to Think With Meta Chain-of-Thought
por: Xiang, Violet, et al.
Publicado: (2025)

Fractured Chain-of-Thought Reasoning
por: Liao, Baohao, et al.
Publicado: (2025)

Reasoning about Reasoning: BAPO Bounds on Chain-of-Thought Token Complexity in LLMs
por: Tomlinson, Kiran, et al.
Publicado: (2026)

Streaming Hallucination Detection in Long Chain-of-Thought Reasoning
por: Lu, Haolang, et al.
Publicado: (2026)

Reason from Future: Reverse Thought Chain Enhances LLM Reasoning
por: Xu, Yinlong, et al.
Publicado: (2025)

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

Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens
por: Zhao, Chengshuai, et al.
Publicado: (2025)

What Makes a Good Reasoning Chain? Uncovering Structural Patterns in Long Chain-of-Thought Reasoning
por: Jiang, Gangwei, et al.
Publicado: (2025)

The Molecular Structure of Thought: Mapping the Topology of Long Chain-of-Thought Reasoning
por: Chen, Qiguang, et al.
Publicado: (2026)

Markov Chain of Thought for Efficient Mathematical Reasoning
por: Yang, Wen, et al.
Publicado: (2024)