Mood Matters: How Syntactic Sensitivity Undermines Safety Alignment

arXiv:2608.05409 · cs.CL · Submitted 2026-08-05 · Read on arXiv

Alina Klerings, Jannik Brinkmann, Heiner Stuckenschmidt, Simone Paolo Ponzetto

cs.CL

Submitted: 2026-08-05

Code: https://github.com/decoderesearch/SAELens

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

Terminology

• Steering — 34× in this paper · explained in 2 episodes
• Accuracy — 7× in this paper · explained in 1 episode
• Generalization — 6× in this paper · explained in 14 episodes
• Robustness — 6× in this paper · explained in 6 episodes
• Benchmark — 5× in this paper · explained in 2 episodes
• Core — 4× in this paper · explained in 2 episodes
• Trace — 4× in this paper · explained in 2 episodes
• Harness — 3× in this paper · explained in 2 episodes
• Interpretability — 3× in this paper · explained in 15 episodes
• Memorization — 3× in this paper · explained in 4 episodes
• Sparsity — 3× in this paper · explained in 5 episodes
• Adam — 2× in this paper · explained in 3 episodes
• Augmentation — 2× in this paper · explained in 2 episodes
• Data Augmentation — 2× in this paper · explained in 10 episodes
• Feature Selection — 2× in this paper · explained in 4 episodes
• Fine-Tuning — 2× in this paper · explained in 5 episodes
• Grounding — 2× in this paper · explained in 3 episodes
• Mechanistic Interpretability — 2× in this paper · explained in 4 episodes
• RLVR — 2× in this paper · explained in 2 episodes
• Synthetic Data Generation — 2× in this paper · explained in 5 episodes
• Transformer — 2× in this paper · explained in 2 episodes
• Activation Steering — 1× in this paper · explained in 7 episodes
• Commentary — 1× in this paper · explained in 2 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 8 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 3 episodes
• Generator — 1× in this paper · explained in 2 episodes
• Judge Reliability — 1× in this paper · explained in 2 episodes
• LLM-as-a-Judge — 1× in this paper · explained in 4 episodes
• Language Models (LLMs) — 1× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 9 episodes
• LoRA Adapters — 1× in this paper · explained in 3 episodes
• Papers — 1× in this paper · explained in 3 episodes
• Reconstruction Error — 1× in this paper · explained in 2 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Representation Engineering — 1× in this paper · explained in 2 episodes
• Representation Learning — 1× in this paper · explained in 3 episodes
• Sparse Autoencoders (SAEs) — 1× in this paper · explained in 4 episodes

Sources

• Apertus: Democratizing Open and Compliant LLMs for Global Language Environments
• DeepSeek LLM: Scaling Open-Source Language Models with Longtermism
• The Llama 3 Herd of Models
• Model Spec Midtraining: Improving How Alignment Training Generalizes
• Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders
• Olmo 3
• Say It Differently: Linguistic Styles as Jailbreak Vectors
• Representation Engineering: A Top-Down Approach to AI Transparency
• Universal and Transferable Adversarial Attacks on Aligned Language Models

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