PPL
Jikun Wu
Papers (8)
Model Collapse as Cultural EvolutionDo Language Models Know What Not to Say? Causal Evidence for Statistical Preemption in LLMsSparse Autoencoders Map Brain-LLM Alignment onto Cortical Semantic TopographyBrain-LLM Alignment Tracks Training Data, Not TypologyParameter-Efficient Neuroevolution for Diverse LLM Generation: Quality-Diversity Optimization via Prompt Embedding EvolutionEvoPref: Multi-Objective Evolutionary Optimization Discovers Diverse LLM Alignments Beyond Gradient DescentBias by Necessity: Impossibility Theorems for Sequential Processing with Convergent AI and Human ValidationWhen Can Human-AI Teams Outperform Individuals? Tight Bounds with Impossibility Guarantees