Spaces:

below-threshold
/

ai-response-validator

Sleeping

App Files Files Community

ai-response-validator / eval

70.1 kB

Ctrl+K

Ctrl+K

3 contributors

History: 8 commits

mbochniak01

Add parallel LangChain RAG engine; harden pipeline from architecture audit

4e2b267 16 days ago

bot-answers.json

13.3 kB
Faithfulness: mean sentence scoring, strip chunk title prefix, lower threshold to 0.35 3 months ago
calibrate.py

2.89 kB
Fix compat, bugs, and types; expand retail KB 3 months ago
compare_faithfulness.py

4.15 kB
Replace HHEM with sentence-level NLI, add claim decomposition and drift detection 3 months ago
compare_retrieval.py

3.38 kB
Add parallel LangChain RAG engine; harden pipeline from architecture audit 16 days ago
drift.py

6.34 kB
Replace HHEM with sentence-level NLI, add claim decomposition and drift detection 3 months ago
golden-dataset.yaml

17.9 kB
Address Gate 5 audit gaps 3 months ago
metrics.py

14.1 kB
Fix compat, bugs, and types; expand retail KB 3 months ago
simulate_traffic.py

8.01 kB
Replace HHEM with sentence-level NLI, add claim decomposition and drift detection 3 months ago