Hi, thanks for organizing this collection of machine learning repositories.
I would like to propose adding WFGY to the section that covers LLM / NLP / tooling.
What it is:
- WFGY is a framework-agnostic TXT-based “debugging layer” that can sit on top of any LLM pipeline.
- The goal is to treat an LLM application as a system that can be diagnosed, instead of only tweaking prompts.
Important piece:
WFGY 2.0 ProblemMap
https://github.com/onestardao/WFGY/tree/main/ProblemMap
- 16 canonical failure modes that often show up in vector-store-based RAG, tools, and agents.
- Each with a diagnosis section and a set of practical remediation steps.
Additionally:
- WFGY 1.0 PDF explains the underlying math and reports quantitative results.
- WFGY 3.0 Singularity Demo provides 131 challenging evaluation prompts.
If you feel this resource fits your collection, I can send a PR with a concise entry.
Hi, thanks for organizing this collection of machine learning repositories.
I would like to propose adding WFGY to the section that covers LLM / NLP / tooling.
What it is:
Important piece:
WFGY 2.0 ProblemMap
https://github.com/onestardao/WFGY/tree/main/ProblemMap
Additionally:
If you feel this resource fits your collection, I can send a PR with a concise entry.