AI & ML Glossary

A

• Ablation study (9)
• Agent (3)
• Alignment
• Attention mechanism (25)

B

• Benchmark (11)

C

• Chain of thought (8)
• Context window
• Convolutional neural network (2)
• Cross-attention (6)
• Cross-modal
• Curriculum learning

D

• Diffusion model (5)
• Distillation (11)

E

• Embeddings (11)
• Embodied AI
• Emergent
• Ensemble (2)

F

• Fine-tuning (36)
• Flow matching (8)

G

• Generalization
• Ground truth
• Grounding (2)
• Group relative policy optimization (20)
• Guardrails

H

• Hallucination (9)
• Held-out set
• Hyperparameter (5)

I

• Image segmentation
• Imitation learning (3)
• In-context learning (3)
• Inference (15)

J

• Jailbreak

K

• KV cache (2)

L

• Large language model (18)
• Latent space (5)
• LoRA (13)

M

• Mixture of experts (9)
• Multi-agent system
• Multimodal (10)

N

• Neural network

O

• On-policy distillation (3)
• Orchestration
• Out-of-distribution (4)
• Overfitting (3)

P

• Parameters (10)
• Pass@k
• Policy (6)
• Policy gradient
• Policy network
• Power law
• PPO (2)
• Precision (3)
• Pretraining (2)
• Prompt injection (3)

Q

• Quantization (9)

R

• Recall (5)
• Red teaming
• Reinforcement learning (37)
• Retrieval-augmented generation (2)
• Reward hacking (6)
• Reward model (2)
• Reward signal
• RLHF
• Router (2)

S

• Self-consistency
• Self-distillation (5)
• Sim-to-real
• Supervised fine-tuning (15)
• Supervised learning (3)

T

• Test-time scaling
• Token (29)
• Tool use
• Trajectory (5)
• Transformer (38)
• Tree of thoughts

V

• Vision transformer
• Vision-language model (17)
• Vision-language-action model

W

• World model (5)

Z

• Zero-shot (6)