Towards Mechanistically Understanding Why Memorized Knowledge Fails to Generalize in Large Language Model Finetuning

arXiv:2607.08393 · cs.AI, cs.CL · Submitted 2026-07-09 · Read on arXiv

cs.AI, cs.CL

Submitted: 2026-07-09

Updated: 2026-09-27

Code: https://github.com/snap-stanford/stark

Terminology

• Generalization — 54× in this paper · explained in 13 episodes
• Accuracy — 32× in this paper · explained in 1 episode
• Fine-Tuning — 32× in this paper · explained in 5 episodes
• Transformer — 9× in this paper · explained in 2 episodes
• Ablation — 7× in this paper · explained in 1 episode
• Core — 7× in this paper · explained in 2 episodes
• Interpretability — 6× in this paper · explained in 15 episodes
• Alignment — 5× in this paper · explained in 1 episode
• Benchmark — 4× in this paper · explained in 2 episodes
• Multi-hop Reasoning — 4× in this paper · explained in 3 episodes
• Robustness — 4× in this paper · explained in 5 episodes
• Activation Patching — 3× in this paper · explained in 2 episodes
• Consistency — 3× in this paper · explained in 3 episodes
• In-Context Learning — 2× in this paper · explained in 4 episodes
• Institutional Review Board — 2× in this paper
• Knowledge Graph — 2× in this paper · explained in 7 episodes
• Trace — 2× in this paper · explained in 2 episodes
• Adam — 1× in this paper · explained in 2 episodes
• Adaptation — 1× in this paper · explained in 1 episode
• Alignment Methods — 1× in this paper · explained in 1 episode
• Alignment Problem — 1× in this paper · explained in 1 episode
• Anchor Tokens — 1× in this paper · explained in 1 episode
• Benchmarking — 1× in this paper · explained in 3 episodes
• Catastrophic Forgetting — 1× in this paper · explained in 5 episodes
• Decoupling — 1× in this paper · explained in 2 episodes
• Generalizability — 1× in this paper · explained in 6 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 2 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Linear probe — 1× in this paper · explained in 2 episodes
• RAG — 1× in this paper · explained in 3 episodes
• Representation Learning — 1× in this paper · explained in 2 episodes
• Retrieval-Augmented Generation (RAG) — 1× in this paper · explained in 16 episodes

Sources

• Key-value memory in the brain
• Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test
• How much do language models memorize?
• In-context Learning and Induction Heads
• Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets
• CaKE: Circuit-aware Editing Enables Generalizable Knowledge Learners

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
• 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
• VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection