Unverified60% confidenceBenchmarkExact time
Reflexion improved the pass rate on the HumanEval programming benchmark from 80% to 91%.
2
Sources
60%
Confidence
Long-term
Relevance
8/20/2026
First Seen
Sources
Related Claims
UnverifiedReflexio distills experience into behavioral guidelines for the agent, clarifying what should be repeated and what should be avoided.61% similarUnverifiedReflexion (2023) proposed replacing numerical rewards with language feedback, allowing agents to summarize failure experiences in natural language and store them in memory.61% similarUnverifiedReflexio's core positioning is Behavioral Learning, enabling AI agents to improve with every interaction.57% similarUnverifiedThe concept of variable ratio reinforcement originates from behavioral psychologist B.F. Skinner's operant conditioning experiments.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/779382API
curl https://kongchang.com/api/v1/knowledge/claims/779382MCP
get_claim(id=779382)