Asking for What Was Never Requested: horizontal and vertical proactivity in LLM agents. Need-graph metrics (no LLM judge) and Q&D, which trains a questioner from the consequences of its questions.
nlp machine-learning lora customer-service preference-learning question-asking peft rag dpo huggingface information-seeking large-language-models retrieval-augmented-generation llm-agents agent-evaluation qwen3 proactive-agents multi-hop-qa tau-bench clarifying-questions
-
Updated
Oct 4, 2026 - Python