[4] Bender, E. M., & Koller, A. (2020). Climbing towards NLU: On meaning, form, and understanding in the age of data. Proceedings of ACL , 5185-5198.
| Layer | Component | Responsibility | |-------|-----------|----------------| | 1 | | High-level states (e.g., GREETING , COLLECT_INFO , VERIFY , RESOLUTION ) and legal transitions. Defined in JSON or YAML. | | 2 | Entity Extractor | Small, fine-tuned model (or regex) that extracts key data (order number, problem type) from user input. | | 3 | LLM Utterance Generator | Takes current state + extracted entities + last user message + state-specific script prompt and generates the bot's reply. | chat scripts
"states": [ "name": "start", "prompt": "Greet user warmly.", "transitions": ["collect_info"] , "name": "collect_info", "prompt": "Ask for order number and problem type. Use extracted entities if present.", "transitions": ["verify", "escalate"] , "name": "verify", "prompt": "Confirm details. If correct, move to resolve.", "transitions": ["resolve", "collect_info"] , "name": "resolve", "prompt": "Provide solution from KB. Do not invent steps.", "transitions": ["end"] ] [4] Bender, E