The Answer-Basin Representation Hypothesis: We Are Not Probing or Steering Concepts

arXiv:2609.24821 · cs.CL · Submitted 2026-09-21 · Read on arXiv

cs.CL

Submitted: 2026-09-21

Updated: 2026-10-02

Comments: 19 pages, 7 figures

Code: https://github.com/V1centNevwake/answer-basin-representation

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Steering — 49× in this paper · explained in 2 episodes
• Alignment — 20× in this paper · explained in 1 episode
• Answer-Basin Representation Hypothesis — 14× in this paper
• Accuracy — 12× in this paper · explained in 1 episode
• Linear probe — 9× in this paper · explained in 2 episodes
• Activation Steering — 5× in this paper · explained in 4 episodes
• Aggregation — 5× in this paper · explained in 1 episode
• Consistency — 3× in this paper · explained in 3 episodes
• Hallucination — 3× in this paper · explained in 3 episodes
• Self-consistency — 3× in this paper · explained in 2 episodes
• Hallucination Detection — 2× in this paper · explained in 2 episodes
• Activation Addition — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Interpretability — 1× in this paper · explained in 15 episodes
• Linear Probing — 1× in this paper · explained in 2 episodes
• Regularization — 1× in this paper · explained in 3 episodes
• Representation Engineering — 1× in this paper · explained in 2 episodes
• Tokenization — 1× in this paper · explained in 2 episodes
• Transformer — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• Refusal in Language Models Is Mediated by a Single Direction
• No Answer Needed: Predicting LLM Answer Accuracy from Question-Only Linear Probes
• Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
• Training Verifiers to Solve Math Word Problems
• Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs
• Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation
• Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
• The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets
• The Geometry of Categorical and Hierarchical Concepts in Large Language Models
• The Information Geometry of Softmax: Probing and Steering
• Steering Language Models With Activation Engineering
• Representation Engineering: A Top-Down Approach to AI Transparency

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering