List your company

Multi-hop Question Answering on MuSiQue (test) (Cost and Repair Metrics)

4,987Token CostDR-RAG
4,336.928,724.9613,11317,501.04Apr 1, 2026

Evaluation results

MethodLinks
DR-RAGBackbone=Qwen3-8B, Baseline=Search-o1 [27]2026.044,98720.8101010
DR-RAGBackbone=Qwen3-8B, Baseline=ReAct [45]2026.045,26126.1109.19.1
DR-RAGBackbone=Qwen3-8B, Baseline=Search-R1 [21]2026.045,39520.61010.510.2
DR-RAGBackbone=Qwen3-4B, Baseline=ReAct [45]2026.045,43223.18.48.59.1
DR-RAGBackbone=Qwen3-4B, Baseline=Search-R1 [21]2026.045,72817.95.66.76.7
DR-RAGBackbone=Qwen3-4B, Baseline=Search-o1 [27]2026.045,73217.19.210.510.4
Step-wiseBackbone=Qwen3-4B, Baseline=Search-R1 [21]2026.046,1625.72.63.73.9
DR-RAGBackbone=LLaMA-3.1-8B-Instruct, Baseline=Search-o1 [27]2026.046,60913.95.64.84.6
RerunBackbone=Qwen3-8B, Baseline=Search-R1 [21]2026.046,63216.76.68.28.1
RerunBackbone=Qwen3-8B, Baseline=Search-o1 [27]2026.046,8097.34.63.63.6
Step-wiseBackbone=Qwen3-8B, Baseline=Search-R1 [21]2026.047,0577.31.422.2
DR-RAGBackbone=LLaMA-3.1-8B-Instruct, Baseline=ReAct [45]2026.047,19620.175.45.2
DR-RAGBackbone=LLaMA-3.1-8B-Instruct, Baseline=Search-R1 [21]2026.047,26812.85.25.95.9
Step-wiseBackbone=Qwen3-8B, Baseline=Search-o1 [27]2026.047,3207.733.43
RAG-Critic [8]Backbone=LLaMA-3.1-8B-Instruct, Baseline=Search-R1 [21]2026.047,5192.610.50.6
Step-wiseBackbone=Qwen3-4B, Baseline=Search-o1 [27]2026.047,6645.32.62.31.9
RerunBackbone=Qwen3-4B, Baseline=Search-R1 [21]2026.047,87413.54.25.96.3
RerunBackbone=Qwen3-4B, Baseline=Search-o1 [27]2026.047,8817.744.44.2
RerunBackbone=Qwen3-8B, Baseline=ReAct [45]2026.048,11315.767.47.3
RAG-Critic [8]Backbone=LLaMA-3.1-8B-Instruct, Baseline=ReAct [45]2026.048,12810.742.22.3
RerunBackbone=Qwen3-4B, Baseline=ReAct [45]2026.048,7731555.85.6
RAG-Critic [8]Backbone=LLaMA-3.1-8B-Instruct, Baseline=Search-o1 [27]2026.049,4864.61.81.61.7
RerunBackbone=LLaMA-3.1-8B-Instruct, Baseline=Search-R1 [21]2026.049,5026.73.44.64.9
Step-wiseBackbone=Qwen3-4B, Baseline=ReAct [45]2026.049,8719.84.23.33.3
Step-wiseBackbone=Qwen3-8B, Baseline=ReAct [45]2026.049,92711.53.63.83.7
RerunBackbone=LLaMA-3.1-8B-Instruct, Baseline=Search-o1 [27]2026.0412,0655.62.83.63.4
Step-wiseBackbone=LLaMA-3.1-8B-Instruct, Baseline=Search-R1 [21]2026.0413,1625.32.42.83.3
RAG-Critic [8]Backbone=Qwen3-4B, Baseline=ReAct [45]2026.0414,18612.14.84.34.4
RerunBackbone=LLaMA-3.1-8B-Instruct, Baseline=ReAct [45]2026.0414,787144.24.54.3
RAG-Critic [8]Backbone=Qwen3-8B, Baseline=ReAct [45]2026.0415,61017.98.68.18.5
RAG-Critic [8]Backbone=Qwen3-8B, Baseline=Search-R1 [21]2026.0416,09911.97.88.79.5
Step-wiseBackbone=LLaMA-3.1-8B-Instruct, Baseline=Search-o1 [27]2026.0416,4776.344.44
RAG-Critic [8]Backbone=Qwen3-4B, Baseline=Search-R1 [21]2026.0417,6856.11.21.11.7
RAG-Critic [8]Backbone=Qwen3-8B, Baseline=Search-o1 [27]2026.0419,7639.77.68.58.8
RAG-Critic [8]Backbone=Qwen3-4B, Baseline=Search-o1 [27]2026.0419,7705.833.73.9
Step-wiseBackbone=LLaMA-3.1-8B-Instruct, Baseline=ReAct [45]2026.0421,239134.64.74.8