Spaces:

MukulRay
/

EmpathRAG

Running

App Files Files Community

EmpathRAG / eval

88.3 kB

Ctrl+K

4 contributors

History: 14 commits

Mukul Rayana

fix: MistralJudge uses from_json_schema grammar — correct structure for claim extraction and verification (Day 15)

5405e38 3 months ago

adversarial_probes.json

6.29 kB
Add eval suite: test prompts, adversarial probes, BERTScore references (Day 12) 3 months ago
adversarial_results.json

10.8 kB
fix: threshold sweep — calibrate DeBERTa guardrail threshold, patch eval imports (Day 14) 3 months ago
adversarial_results_t050.json

9.21 kB
eval: save adversarial results at both t=0.50 and t=0.85, reset production threshold to 0.50 (Day 14) 3 months ago
adversarial_results_t085.json

9.22 kB
eval: save adversarial results at both t=0.50 and t=0.85, reset production threshold to 0.50 (Day 14) 3 months ago
bertscore_references.json

13.5 kB
Add eval suite: test prompts, adversarial probes, BERTScore references (Day 12) 3 months ago
build_bertscore_references.py

4.39 kB
Add eval suite: test prompts, adversarial probes, BERTScore references (Day 12) 3 months ago
condition_a.py

1.61 kB
feat: eval scripts — BM25 baseline, adversarial probes, BERTScore, RAGAS, Wilcoxon (Day 14) 3 months ago
keyword_filter.py

1.38 kB
Day 1: data pipeline, session tracker, query router, adversarial probes, Colab training notebooks 3 months ago
run_adversarial.py

4.84 kB
eval: save adversarial results at both t=0.50 and t=0.85, reset production threshold to 0.50 (Day 14) 3 months ago
run_bertscore.py

3.18 kB
fix: guardrail dual-import path, bertscore key names, ragas reuse pipeline.llm (Day 14) 3 months ago
run_ragas.py

7.07 kB
fix: MistralJudge uses from_json_schema grammar — correct structure for claim extraction and verification (Day 15) 3 months ago
run_wilcoxon.py

5.6 kB
fix: wilcoxon sys.path for condition_a import, zero_method=pratt for binary scores (Day 15) 3 months ago
smoke_test_results.json

1.43 kB
feat: Gradio demo app, DeBERTa Colab notebook, updated smoke test results (Day 13) 3 months ago
test_prompts.json

8.92 kB
Add eval suite: test prompts, adversarial probes, BERTScore references (Day 12) 3 months ago
wilcoxon_results.json

920 Bytes
fix: MistralJudge JSON extraction for DeepEval faithfulness, commit Wilcoxon results (Day 15) 3 months ago