Image de couverture de livre

How Well Do Models Follow Their Constitutions?

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Jakkli, Arya, Rajamanoharan, Senthooran, Nanda, Neel
Format:	Preprint
Publié:	2026
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models
par: Ferrando, Javier, et autres
Publié: (2024)

Emergent Misalignment is Easy, Narrow Misalignment is Hard
par: Soligo, Anna, et autres
Publié: (2026)

Convergent Linear Representations of Emergent Misalignment
par: Soligo, Anna, et autres
Publié: (2025)

Model Organisms for Emergent Misalignment
par: Turner, Edward, et autres
Publié: (2025)

Subliminal Learning Is Steering Vector Distillation
par: Blank, Camila, et autres
Publié: (2026)

Thought Branches: Interpreting LLM Reasoning Requires Resampling
par: Macar, Uzay, et autres
Publié: (2025)

Are Sparse Autoencoders Useful? A Case Study in Sparse Probing
par: Kantamneni, Subhash, et autres
Publié: (2025)

Steering Out-of-Distribution Generalization with Concept Ablation Fine-Tuning
par: Casademunt, Helena, et autres
Publié: (2025)

Chain-of-Thought Reasoning In The Wild Is Not Always Faithful
par: Arcuschin, Iván, et autres
Publié: (2025)

Censored LLMs as a Natural Testbed for Secret Knowledge Elicitation
par: Casademunt, Helena, et autres
Publié: (2026)

Improving Dictionary Learning with Gated Sparse Autoencoders
par: Rajamanoharan, Senthooran, et autres
Publié: (2024)

Towards eliciting latent knowledge from LLMs with mechanistic interpretability
par: Cywiński, Bartosz, et autres
Publié: (2025)

Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2
par: Lieberum, Tom, et autres
Publié: (2024)

Simple Mechanistic Explanations for Out-Of-Context Reasoning
par: Wang, Atticus, et autres
Publié: (2025)

Explorations of Self-Repair in Language Models
par: Rushing, Cody, et autres
Publié: (2024)

Towards Best Practices of Activation Patching in Language Models: Metrics and Methods
par: Zhang, Fred, et autres
Publié: (2023)

Base Models Know How to Reason, Thinking Models Learn When
par: Venhoff, Constantin, et autres
Publié: (2025)

Dense SAE Latents Are Features, Not Bugs
par: Sun, Xiaoqing, et autres
Publié: (2025)

When Chain of Thought is Necessary, Language Models Struggle to Evade Monitors
par: Emmons, Scott, et autres
Publié: (2025)

One Instruction Does Not Fit All: How Well Do Embeddings Align Personas and Instructions in Low-Resource Indian Languages?
par: Shah, Arya, et autres
Publié: (2026)

Eliciting Secret Knowledge from Language Models
par: Cywiński, Bartosz, et autres
Publié: (2025)

BatchTopK Sparse Autoencoders
par: Bussmann, Bart, et autres
Publié: (2024)

Steering Evaluation-Aware Language Models to Act Like They Are Deployed
par: Hua, Tim Tian, et autres
Publié: (2025)

Jumping Ahead: Improving Reconstruction Fidelity with JumpReLU Sparse Autoencoders
par: Rajamanoharan, Senthooran, et autres
Publié: (2024)

How Well Do Multimodal Models Reason on ECG Signals?
par: Xu, Maxwell A., et autres
Publié: (2026)

How Well Do Large Language Models Truly Ground?
par: Lee, Hyunji, et autres
Publié: (2023)

Understanding Reasoning in Thinking Language Models via Steering Vectors
par: Venhoff, Constantin, et autres
Publié: (2025)

Learning Multi-Level Features with Matryoshka Sparse Autoencoders
par: Bussmann, Bart, et autres
Publié: (2025)

Sparse Autoencoders Do Not Find Canonical Units of Analysis
par: Leask, Patrick, et autres
Publié: (2025)

How Well Do LLMs Understand Tunisian Arabic?
par: Mahdi, Mohamed
Publié: (2025)

Because we have LLMs, we Can and Should Pursue Agentic Interpretability
par: Kim, Been, et autres
Publié: (2025)

SPIN-Bench: How Well Do LLMs Plan Strategically and Reason Socially?
par: Yao, Jianzhu, et autres
Publié: (2025)

Lost in the Pipeline: How Well Do Large Language Models Handle Data Preparation?
par: Spreafico, Matteo, et autres
Publié: (2025)

How Well Do Deep Learning Models Capture Human Concepts? The Case of the Typicality Effect
par: Vemuri, Siddhartha K., et autres
Publié: (2024)

Interpretable Embeddings with Sparse Autoencoders: A Data Analysis Toolkit
par: Jiang, Nick, et autres
Publié: (2025)

Thought Anchors: Which LLM Reasoning Steps Matter?
par: Bogdan, Paul C., et autres
Publié: (2025)

CountQA: How Well Do MLLMs Count in the Wild?
par: Tamarapalli, Jayant Sravan, et autres
Publié: (2025)

How Well Can AI Build SD Models?
par: Schoenberg, William, et autres
Publié: (2025)

Privacy Issues in Large Language Models: A Survey
par: Neel, Seth, et autres
Publié: (2023)

What's the plan? Metrics for implicit planning in LLMs and their application to rhyme generation and question answering
par: Maar, Jim, et autres
Publié: (2026)