Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Question Answering on DetectiveQA
Loading...
67.25
Accuracy
QRRanker
25.754
36.527
47.3
58.073
Feb 12, 2026
Mar 5, 2026
Mar 26, 2026
Apr 16, 2026
May 7, 2026
May 28, 2026
Jun 18, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
QRRanker
Method Category=QRRank...
2026.02
67.25
QRHeads-4B
Method Category=QRRank...
2026.02
64.75
Qwen3-Reranker-4B (trained)
Method Category=Rerank...
2026.02
64.52
SFT-Embedding-8B
Method Category=Embedd...
2026.02
62.85
Qwen3-Reranker-4B (vanilla)
Method Category=Rerank...
2026.02
60.93
Qwen3-Embedding-8B
Method Category=Embedd...
2026.02
57.35
Top-down
Backbone=Qwen3-235B-A2...
2026.06
40.27
Bottom-up
Backbone=Qwen3-235B-A2...
2026.06
39.6
TOA
Backbone=Qwen3-235B-A2...
2026.06
38.76
Top-down
Backbone=Qwen3-30B-A3B...
2026.06
35.91
Bottom-up
Backbone=Qwen3-30B-A3B...
2026.06
34.06
Zero-shot
Backbone=Qwen3-235B-A2...
2026.06
33.72
TOA
Backbone=Qwen3-30B-A3B...
2026.06
32.72
QwenLong-L1.5
Backbone=Qwen3-30B-A3B...
2026.06
32.21
Zero-shot
Backbone=Qwen3-30B-A3B...
2026.06
30.87
MemAgent
Backbone=Qwen3-235B-A2...
2026.06
30.03
MemAgent
Backbone=Qwen3-30B-A3B...
2026.06
27.35
Feedback
Search any
task
Search any
task