Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Tracking Shuffled Objects on Tracking Shuffled Objects 5 objects (test)
Loading...
94.1
Accuracy (TSO 5-obj)
TextReg
33.26
49.055
64.85
80.645
May 20, 2026
Accuracy (TSO 5-obj)
Updated 13d ago
Evaluation Results
Method
Method
Links
Accuracy (TSO 5-obj)
TextReg
Test Engine=Phi-3.5-Mi...
2026.05
94.1
CoT
Test Engine=Phi-3.5-Mi...
2026.05
89.6
REVOLVE
Test Engine=Phi-3.5-Mi...
2026.05
86.8
TextGrad
Test Engine=Phi-3.5-Mi...
2026.05
83.6
TextReg
Test Engine=Llama-3.1-...
2026.05
79.7
REVOLVE
Test Engine=Llama-3.1-...
2026.05
76
TextGrad
Test Engine=Llama-3.1-...
2026.05
69.7
CoT
Test Engine=Llama-3.1-...
2026.05
58.8
TextReg
Test Engine=Llama-3-8B...
2026.05
54.3
REVOLVE
Test Engine=Llama-3-8B...
2026.05
46.1
TextGrad
Test Engine=Llama-3-8B...
2026.05
45.9
TextReg
Test Engine=Qwen2-7B-I...
2026.05
45.4
CoT
Test Engine=Qwen2-7B-I...
2026.05
42
REVOLVE
Test Engine=Qwen2-7B-I...
2026.05
40.3
TextGrad
Test Engine=Qwen2-7B-I...
2026.05
38.2
CoT
Test Engine=Llama-3-8B...
2026.05
35.6
Feedback
Search any
task
Search any
task