Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Length Prediction on ForeLen Reasoning 1.0 (test)
Loading...
138.01
MAE
TRAIL
110.8992
293.8971
476.895
659.8929
Feb 12, 2026
MAE
Updated 1mo ago
Evaluation Results
Method
Method
Links
MAE
TRAIL
Model=Qwen2.5 1.5B
2026.02
138.01
LTR-C
Model=Qwen2.5 1.5B
2026.02
139.02
EGTP
Model=Qwen2.5 0.5B
2026.02
145.85
TRAIL
Model=Qwen2.5 0.5B
2026.02
146.24
EGTP
Model=Qwen2.5 1.5B
2026.02
146.48
S3
Model=Qwen2.5 1.5B
2026.02
169.72
LTR-C
Model=Qwen2.5 0.5B
2026.02
178.28
S3
Model=Qwen2.5 0.5B
2026.02
200.91
SSJF-Reg
Model=Qwen2.5 0.5B
2026.02
221.15
PiA
Model=Qwen2.5 0.5B
2026.02
296.02
SSJF-Reg
Model=Qwen2.5 1.5B
2026.02
310.64
PiA
Model=Qwen2.5 1.5B
2026.02
432.05
SSJF-MC
Model=Qwen2.5 0.5B
2026.02
466
TPV
Model=Qwen2.5 1.5B
2026.02
584.08
TPV
Model=Qwen2.5 0.5B
2026.02
597
SSJF-MC
Model=Qwen2.5 1.5B
2026.02
815.78
Feedback
Search any
task
Search any
task