Sensitivity, Causality, and Repair Dissociate: A Layer-Wise Analysis of Perturbation Robustness and Its Scaling

arXiv:2608.03842 · cs.CL, cs.LG · Submitted 2026-08-04 · Read on arXiv

Nathan Labiosa, David Buff, Ena Nayak, Erica Donno

cs.CL, cs.LG

Submitted: 2026-08-04

Comments: 29 pages, 18 figures, 11 tables

Code: https://github.com/NathanLabiosa/Layer-Wise-Representation-Divergence

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

Terminology

• Harness — 61× in this paper · explained in 2 episodes
• Robustness — 10× in this paper · explained in 6 episodes
• Ablation — 9× in this paper · explained in 1 episode
• Transformer — 9× in this paper · explained in 2 episodes
• Accuracy — 8× in this paper · explained in 1 episode
• Activation Patching — 8× in this paper · explained in 2 episodes
• Anti-correlation — 6× in this paper · explained in 3 episodes
• Effective Rank — 4× in this paper · explained in 2 episodes
• Interpretability — 4× in this paper · explained in 15 episodes
• Papers — 3× in this paper · explained in 3 episodes
• Transformers — 3× in this paper · explained in 2 episodes
• Benchmark — 2× in this paper · explained in 2 episodes
• Core — 2× in this paper · explained in 2 episodes
• Frozen model — 2× in this paper · explained in 3 episodes
• LoRA Adapters — 2× in this paper · explained in 3 episodes
• Localization — 2× in this paper · explained in 2 episodes
• Attention Entropy — 1× in this paper · explained in 2 episodes
• Consistency — 1× in this paper · explained in 3 episodes
• Evaluation Protocol — 1× in this paper · explained in 2 episodes
• Generalization — 1× in this paper · explained in 14 episodes
• Hyperparameters — 1× in this paper · explained in 2 episodes
• LayerWise Representation Divergence — 1× in this paper
• Mechanistic Interpretability — 1× in this paper · explained in 4 episodes
• Monotonicity — 1× in this paper · explained in 2 episodes
• Representation Engineering — 1× in this paper · explained in 2 episodes
• Scaffolding — 1× in this paper · explained in 3 episodes
• Verifier — 1× in this paper · explained in 2 episodes
• Weight Decay — 1× in this paper · explained in 2 episodes

Sources

• Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
• Training Verifiers to Solve Math Word Problems
• The Llama 3 Herd of Models
• Mistral 7B
• Qwen2.5 Technical Report
• Gemma 2: Improving Open Language Models at a Practical Size

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