AlignMerge - Alignment-Preserving Large Language Model Merging via Fisher-Guided Geometric Constraints

arXiv:2512.16245 · cs.AI · Submitted 2025-12-18 · Read on arXiv

cs.AI

Submitted: 2025-12-18

Updated: 2026-10-10

Terminology

• Alignment — 12× in this paper · explained in 1 episode
• Alignment Quality Index — 1× in this paper
• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Model Merging — 1× in this paper · explained in 2 episodes

Sources

• Evolutionary Optimization of Model Merging Recipes
• The Internal State of an LLM Knows When It's Lying
• Scheming AIs: Will AIs fake alignment during training in order to get power?
• SALSA: Soup-based Alignment Learning for Stronger Adaptation in RLHF
• $C^2M^3$: Cycle-Consistent Multi-Model Merging
• SafeMERGE: Preserving Safety Alignment in Fine-Tuned Large Language Models via Selective Layer-Wise Model Merging
• Model Merging and Safety Alignment: One Bad Model Spoils the Bunch
• Risks from Learned Optimization in Advanced Machine Learning Systems
• Uncovering Deceptive Tendencies in Language Models: A Simulated Company AI Assistant
• TIES-Merging: Resolving Interference When Merging Models
• Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities

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