agential-ai/agential

[Feature Request]: Reflexion, Standardizing Reflection Few-Shots

Closed this issue · 0 comments

alckasoc commented

Feature Description

Benchmark CoT ReAct
HotpotQA ✅ ✅
FEVER ✅ ✅
TriviaQA ✅ ✅
AmbigNQ ✅ ✅
GSM8K ✅ ✅
SVAMP ✅ ✅
TabMWP ✅ ✅
MBPP ✅ ✅
HumanEval ✅ ✅

Same as #185