List your company

Skill Evolution on S-Bench

54.8AccuracyALIGNEVOSKILL
1.55215.37629.243.024Jun 29, 2025

Evaluation results

MethodLinks
ALIGNEVOSKILLLLM=GPT-5.4, Evolution Setting=Human-Labeled2025.0654.8—
CoEvoSkillsLLM=GPT-5.4, Evolution Setting=Human-Labeled2025.0652.4—
SkillXLLM=GPT-5.4, Evolution Setting=Human-Labeled2025.0651.2—
Trace2SkillLLM=GPT-5.4, Evolution Setting=Human-Labeled2025.0650—
ALIGNEVOSKILLLLM=GPT-5.4, Evolution Setting=From-Scratch2025.0648.8—
CoEvoSkillsLLM=GPT-5.4, Evolution Setting=From-Scratch2025.0646.4—
SkillXLLM=GPT-5.4, Evolution Setting=From-Scratch2025.0645.2—
BaseLLM=GPT-5.4, Evolution Setting=Human-Labeled2025.0645.2—
Trace2SkillLLM=GPT-5.4, Evolution Setting=From-Scratch2025.0642.9—
BaseLLM=GPT-5.4, Evolution Setting=From-Scratch2025.0639.3—
ALIGNEVOSKILLLLM=DEEPSEEK-V4-FLASH, Evolution Setting=Human-Labeled2025.0634.5—
CoEvoSkillsLLM=DEEPSEEK-V4-FLASH, Evolution Setting=Human-Labeled2025.0632.1—
SkillXLLM=DEEPSEEK-V4-FLASH, Evolution Setting=Human-Labeled2025.0631—
ALIGNEVOSKILLLLM=CLAUDE-HAIKU-4.5, Evolution Setting=Human-Labeled2025.0631—
Trace2SkillLLM=DEEPSEEK-V4-FLASH, Evolution Setting=Human-Labeled2025.0629.8—
ALIGNEVOSKILLLLM=CLAUDE-HAIKU-4.5, Evolution Setting=From-Scratch2025.0628.6—
CoEvoSkillsLLM=CLAUDE-HAIKU-4.5, Evolution Setting=Human-Labeled2025.0628.6—
SkillXLLM=CLAUDE-HAIKU-4.5, Evolution Setting=Human-Labeled2025.0627.4—
ALIGNEVOSKILLLLM=DEEPSEEK-V4-FLASH, Evolution Setting=From-Scratch2025.0626.2—
CoEvoSkillsLLM=CLAUDE-HAIKU-4.5, Evolution Setting=From-Scratch2025.0626.2—
ALIGNEVOSKILLLLM=GPT-5.4-NANO, Evolution Setting=Human-Labeled2025.0626.2—
SkillXLLM=CLAUDE-HAIKU-4.5, Evolution Setting=From-Scratch2025.0625—
CoEvoSkillsLLM=GPT-5.4-NANO, Evolution Setting=Human-Labeled2025.0625—
CoEvoSkillsLLM=DEEPSEEK-V4-FLASH, Evolution Setting=From-Scratch2025.0623.8—
Trace2SkillLLM=CLAUDE-HAIKU-4.5, Evolution Setting=Human-Labeled2025.0623.8—
SkillXLLM=GPT-5.4-NANO, Evolution Setting=Human-Labeled2025.0623.8—
SkillXLLM=DEEPSEEK-V4-FLASH, Evolution Setting=From-Scratch2025.0622.6—
BaseLLM=DEEPSEEK-V4-FLASH, Evolution Setting=Human-Labeled2025.0622.6—
Trace2SkillLLM=CLAUDE-HAIKU-4.5, Evolution Setting=From-Scratch2025.0622.6—
Trace2SkillLLM=GPT-5.4-NANO, Evolution Setting=Human-Labeled2025.0622.6—
Trace2SkillLLM=DEEPSEEK-V4-FLASH, Evolution Setting=From-Scratch2025.0621.4—
BaseLLM=CLAUDE-HAIKU-4.5, Evolution Setting=From-Scratch2025.0619—
ALIGNEVOSKILLLLM=GPT-5.4-NANO, Evolution Setting=From-Scratch2025.0619—
CoEvoSkillsLLM=GPT-5.4-NANO, Evolution Setting=From-Scratch2025.0616.7—
SkillXLLM=GPT-5.4-NANO, Evolution Setting=From-Scratch2025.0615.5—
Trace2SkillLLM=GPT-5.4-NANO, Evolution Setting=From-Scratch2025.0614.3—
BaseLLM=DEEPSEEK-V4-FLASH, Evolution Setting=From-Scratch2025.0613.1—
BaseLLM=CLAUDE-HAIKU-4.5, Evolution Setting=Human-Labeled2025.069.5—
BaseLLM=GPT-5.4-NANO, Evolution Setting=Human-Labeled2025.069.5—
BaseLLM=GPT-5.4-NANO, Evolution Setting=From-Scratch2025.063.6—