List your company
ResearchFormerly WizWand

Conversational response generation on TopiOCQA

30.6F1 ScoreChatR1-7b
5.74412.19718.6525.103Oct 15, 2025
Updated 6mo ago

Evaluation results

MethodLinks
ChatR1-7bRAG=RAG, LLM=Qwen-7b, Training Strategy=Supervised Fine-tuning or RL Training2025.1030.679.5
UniConvRAG=RAG, LLM=Mistral 7b, Training Strategy=Supervised Fine-tuning or RL Training2025.1029.6—
ChatR1-3bRAG=RAG, LLM=Qwen-3b, Training Strategy=Supervised Fine-tuning or RL Training2025.1029.480.9
ChatRetriever +Mis.RAG=RAG, LLM=Mistral 7b, Training Strategy=Supervised Fine-tuning or RL Training2025.1028.3—
Claude (DI)RAG=No, LLM=Claude, Training Strategy=Zero-Shot2025.1027.2—
conv-ANCE +Mis.RAG=RAG, LLM=Mistral 7b, Training Strategy=Supervised Fine-tuning or RL Training2025.1027.2—
ChatGPT (DI)RAG=No, LLM=GPT-3.5, Training Strategy=Zero-Shot2025.1025.577.5
ChatR1 (w/o Rint.)RAG=RAG, LLM=Qwen-3b, Training Strategy=Supervised Fine-tuning or RL Training2025.1024.473
QR Search R1RAG=RAG, LLM=Qwen-3b, Training Strategy=Supervised Fine-tuning or RL Training2025.1020.172.7
SFTRAG=No, LLM=Qwen-3b, Training Strategy=Supervised Fine-tuning or RL Training2025.101878.8
CoT R1RAG=No, LLM=Qwen-3b, Training Strategy=Supervised Fine-tuning or RL Training2025.1012.570.3
IRCoTRAG=RAG, LLM=Qwen-3b, Training Strategy=Zero-Shot2025.108.961
Qwen-Instr. (RAG)RAG=RAG, LLM=Qwen-3b, Training Strategy=Zero-Shot2025.108.854.7
Qwen-Instr. (DI)RAG=No, LLM=Qwen-3b, Training Strategy=Zero-Shot2025.106.757.3
Qwen-Instr. (CoT)RAG=No, LLM=Qwen-3b, Training Strategy=Zero-Shot2025.106.758.6