Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Algorithm Synthesis on Novel Hard Benchmark 15 problems version 2
Loading...
7
Correct Count (out of 15)
Reflexion
2.84
3.92
5
6.08
Jun 29, 2026
Correct Count (out of 15)
T-opt Score
Updated 25d ago
Evaluation Results
Method
Method
Links
Correct Count (out of 15)
T-opt Score
Reflexion
Source=version 2
2026.06
7
86
AlgoSkill-MCTS
Source=version 2
2026.06
7
100
MapCoder
Source=prior-only
2026.06
6
50
CoT
Source=version 2
2026.06
4
100
Self-Refine
Source=version 2
2026.06
4
-
AlgoSkill-Greedy
Source=version 2
2026.06
4
100
Direct LLM
Source=version 2
2026.06
3
100
Feedback
Search any
task
Search any
task