Toward Alignment Scaling Laws: A Framework and First Preregistered Measurements
cs.AI, cs.CL, cs.LG
Submitted: 2026-10-06
Updated: 2026-10-06
Code: https://github.com/AlignmentResearch/scaling-llm-robustness-paper
Terminology
Sources
- Concrete Problems in AI Safety
- Refusal in Language Models Is Mediated by a Single Direction
- A General Language Assistant as a Laboratory for Alignment
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs
- Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling
- Scaling Trends for Data Poisoning in LLMs
- Measuring Progress on Scalable Oversight for Large Language Models
- Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision
- Bias-Augmented Consistency Training Reduces Biased Reasoning in Chain-of-Thought
- DeepSeek-V3 Technical Report
- Maintaining Plasticity in Deep Continual Learning
- Scaling Laws For Scalable Oversight
- Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned
- The Capacity for Moral Self-Correction in Large Language Models
- Scaling Laws for Reward Model Overoptimization
- The Llama 3 Herd of Models
- Alignment faking in large language models
- OLMo: Accelerating the Science of Language Models
- Jailbreak Scaling Laws for Large Language Models: Polynomial-Exponential Crossover
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems