fig2

From: From large language models to AI agents in energy materials research: enabling discovery, design, and automation

Figure 2. Milestones in the development of LLMs, highlighting architectural advances, multimodal integration, and the emergence of agentic reasoning for scientific automation. LLMs: Large language models; RLHF: reinforcement learning from human feedback; AI: artificial intelligence; BERT: bidirectional encoder representations from transformers; GPT-2: generative pre-trained transformer 2; API: application programming interface.