Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Multi-hop Question Answering on HotPotQA (Task Success, CoT, Latency)

100CoT Match Rate

Teacher (OPT13B)

61.93671.81881.791.582May 20, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.05
10082.740.74.8
2025.05
1008139.94.7
2025.05
10084.339.14.6
2025.05
100---
2025.05
100---
2025.05
10078.510.86.2
2025.05
86.578.638.94.7
2025.05
84.775.239.84.8
2025.05
84.7---
2025.05
8473.940.24.9
2025.05
84---
2025.05
83.474.8425.1
2025.05
82.673.542.55.1
2025.05
81.572.443.15.2
2025.05
81.271.543.25.2
2025.05
81.2---
2025.05
80.473.111.76.6
2025.05
80.269.742.95.3
2025.05
80.2---
2025.05
8070.943.65.2
2025.05
79.86741.75
2025.05
79.8---
2025.05
7968.143.35.4
2025.05
7970.1445.3
2025.05
78.167.243.95.4
2025.05
78---
2025.05
77.1---
2025.05
76.5---
2025.05
76.469.1127.2
2025.05
75.562.945.25.6
2025.05
75.5---
2025.05
75.3---
2025.05
74.458.543.65.3
2025.05
74.4---
2025.05
74.367.412.37.3
2025.05
74.161.3465.6
2025.05
7465.512.27
2025.05
73.460.746.55.7
2025.05
73.166.212.67.4
2025.05
72.3---
2025.05
70.561.5137.9
2025.05
70.2---
2025.05
69.854.148.56
2025.05
69.8---
2025.05
69.360.213.48.1
2025.05
68.452.349.46.2
2025.05
68.358.813.88.2
2025.05
67.35150.16.3
2025.05
66.252.813.87.8
2025.05
65.748.314.88.9
2025.05
65.4---
2025.05
64.247.2159
2025.05
63.7---
2025.05
63.446.115.39.2