Why Fine-Tuning Encourages Hallucinations and How to Fix It

arXiv:2604.15574 · cs.CL, cs.AI, cs.LG, cs.NE · Submitted 2026-04-16 · Read on arXiv

cs.CL, cs.AI, cs.LG, cs.NE

Submitted: 2026-04-16

Updated: 2026-09-25

Comments: Published in the CoLM 2026 conference

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

Terminology

• Distillation — 80× in this paper · explained in 2 episodes
• Self-distillation — 62× in this paper · explained in 6 episodes
• Hallucination — 57× in this paper · explained in 3 episodes
• Fine-Tuning — 32× in this paper · explained in 5 episodes
• Accuracy — 27× in this paper · explained in 1 episode
• Continual Learning — 24× in this paper · explained in 2 episodes
• Regularization — 8× in this paper · explained in 3 episodes
• Ablation — 7× in this paper · explained in 1 episode
• Adaptation — 5× in this paper · explained in 1 episode
• Transformer — 5× in this paper · explained in 2 episodes
• Alignment — 4× in this paper · explained in 1 episode
• Catastrophic Forgetting — 3× in this paper · explained in 5 episodes
• Supervised Fine-Tuning (SFT) — 3× in this paper · explained in 8 episodes
• Consistency — 2× in this paper · explained in 3 episodes
• Adam — 1× in this paper · explained in 2 episodes
• Benchmark — 1× in this paper · explained in 2 episodes
• Core — 1× in this paper · explained in 2 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Generative AI — 1× in this paper · explained in 2 episodes
• Hallucination Detection — 1× in this paper · explained in 2 episodes
• Interpretability — 1× in this paper · explained in 15 episodes
• Knowledge Distillation — 1× in this paper · explained in 11 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Scaling Laws — 1× in this paper · explained in 4 episodes
• Self-consistency — 1× in this paper · explained in 2 episodes
• Semantic Similarity — 1× in this paper · explained in 2 episodes
• Trace — 1× in this paper · explained in 2 episodes

Sources

• Physics of Language Models: Part 3.3, Knowledge Capacity Scaling Laws
• The Internal State of an LLM Knows When It's Lying
• From Single to Multi: How LLMs Hallucinate in Multi-Document Summarization
• Dark Experience for General Continual Learning: a Strong, Simple Baseline
• Analyzing Transformers in Embedding Space
• Inferring Functionality of Attention Heads from their Parameters
• Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?
• Inside-Out: Hidden Factual Knowledge in LLMs
• Thinking to Recall: How Reasoning Unlocks Parametric Knowledge in LLMs
• Transformer Feed-Forward Layers Are Key-Value Memories
• Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space
• The Llama 3 Herd of Models
• Distilling the Knowledge in a Neural Network
• Towards Continual Knowledge Learning of Language Models
• Why Language Models Hallucinate
• From Tokens to Words: On the Inner Lexicon of LLMs
• Follow the Flow: On Information Flow Across Textual Tokens in Text-to-Image Models
• Achieving a Better Stability-Plasticity Trade-off via Auxiliary Networks in Continual Learning
• A continual learning survey: Defying forgetting in classification tasks
• BiLD: Bi-directional Logits Difference Loss for Large Language Model Distillation

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