Search papers, labs, and topics across Lattice.
Research organized by category. Examine each subsection of AI.
Sparklines show daily paper volume. Percentages show change in papers released between each week / 2 weeks / month - green means accelerating, red means decelerating.
Understanding the internal mechanisms of neural networks through circuit analysis, feature visualization, and mechanistic interpretability.
Training AI systems from human feedback using reinforcement learning, direct preference optimization, and reward modeling.
AI governance principles, value alignment through constitutions, fairness, bias mitigation, and ethical AI deployment.
Adversarial testing of AI systems, jailbreaking research, prompt injection defense, and robustness evaluation.
Evaluation methodology for AI systems, benchmark design, capability measurement, and safety evaluations.
Theoretical foundations of alignment, scalable oversight mechanisms, debate protocols, and iterated amplification.
Chain-of-thought prompting, mathematical reasoning, logical inference, and step-by-step problem solving in LLMs.
Model compression, quantization, pruning, distillation, and efficient inference for deployment.
AI for scientific research, protein structure prediction, drug discovery, materials science, and climate modeling.
