Farther the Shift, Sparser the Representation: Analyzing OOD Mechanisms in LLMs

arXiv:2603.03415 · cs.CL, cs.AI · Submitted 2026-03-03 · Read on arXiv

cs.CL, cs.AI

Submitted: 2026-03-03

Updated: 2026-09-26

Code: https://github.com/MingyuJ666/sparsityLLM

Terminology

• Sparsity — 169× in this paper · explained in 4 episodes
• Accuracy — 21× in this paper · explained in 1 episode
• Transformer — 20× in this paper · explained in 2 episodes
• Benchmark — 13× in this paper · explained in 2 episodes
• Interpretability — 11× in this paper · explained in 15 episodes
• In-Context Learning — 9× in this paper · explained in 4 episodes
• Knowledge Graph — 9× in this paper · explained in 7 episodes
• Pruning — 9× in this paper · explained in 4 episodes
• Activation Sparsity — 7× in this paper · explained in 1 episode
• Core — 7× in this paper · explained in 2 episodes
• Generalization — 7× in this paper · explained in 13 episodes
• Curriculum Learning — 6× in this paper
• Fine-Tuning — 6× in this paper · explained in 5 episodes
• International Conference on Machine Learning — 4× in this paper
• Large Language Models (LLMs) — 4× in this paper · explained in 6 episodes
• Robustness — 4× in this paper · explained in 5 episodes
• Adaptive Mechanism — 3× in this paper · explained in 1 episode
• Distribution Shift — 3× in this paper · explained in 5 episodes
• Embeddings — 3× in this paper · explained in 2 episodes
• Empirical Methods in Natural Language Processing — 3× in this paper
• Foundation Model — 3× in this paper · explained in 3 episodes
• Foundation Models — 3× in this paper · explained in 7 episodes
• Semantic Similarity — 3× in this paper · explained in 2 episodes
• Adam — 2× in this paper · explained in 2 episodes
• Adaptation — 2× in this paper · explained in 1 episode
• Adversarial Noise — 2× in this paper · explained in 1 episode
• Alignment — 2× in this paper · explained in 1 episode
• Chain-of-Thought Prompting — 2× in this paper · explained in 2 episodes
• Downstream Tasks — 2× in this paper · explained in 2 episodes
• Explainability — 2× in this paper · explained in 3 episodes
• Graph Construction — 2× in this paper · explained in 3 episodes
• Hallucination — 2× in this paper · explained in 3 episodes
• In International Conference on Learning Representations — 2× in this paper
• Meeting of the Association for Computational Linguistics — 2× in this paper
• Trace — 2× in this paper · explained in 2 episodes
• Activation Patching — 1× in this paper · explained in 2 episodes
• Benchmarking — 1× in this paper · explained in 3 episodes
• Cognitive Load — 1× in this paper · explained in 2 episodes
• Consistency — 1× in this paper · explained in 3 episodes
• Entrainment — 1× in this paper · explained in 2 episodes
• Feature Selection — 1× in this paper · explained in 4 episodes
• Hallucination Detection — 1× in this paper · explained in 2 episodes
• Instruction Tuning — 1× in this paper · explained in 2 episodes
• Knowledge graphs — 1× in this paper · explained in 2 episodes
• Mixture-of-Experts (MoE) — 1× in this paper · explained in 7 episodes
• Mixture-of-Experts (MoE) Models — 1× in this paper · explained in 3 episodes
• Modularity — 1× in this paper · explained in 9 episodes
• Multi-hop Reasoning — 1× in this paper · explained in 3 episodes
• Parsimony — 1× in this paper · explained in 2 episodes
• Regularization — 1× in this paper · explained in 3 episodes
• Representation Learning — 1× in this paper · explained in 2 episodes
• Sample Efficiency — 1× in this paper · explained in 4 episodes
• Sparse Autoencoders (SAEs) — 1× in this paper · explained in 3 episodes
• Steering — 1× in this paper · explained in 2 episodes
• Transfer Learning — 1× in this paper · explained in 10 episodes
• Transformer Models — 1× in this paper · explained in 2 episodes

Sources

• Eliciting Latent Predictions from Transformers with the Tuned Lens
• SAGE: An Agentic Explainer Framework for Interpreting SAE Features in Language Models
• SAE-SSV: Supervised Steering in Sparse Representation Spaces for Reliable Control of Language Models
• Measuring Faithfulness in Chain-of-Thought Reasoning
• Orca: Progressive Learning from Complex Explanation Traces of GPT-4
• LLaMA: Open and Efficient Foundation Language Models
• DUMP: Automated Distribution-Level Curriculum Learning for RL-based LLM Post-training
• Qwen2.5 Technical Report
• Towards Uncovering How Large Language Model Works: An Explainability Perspective

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