Performance Estimation on ToxicChat
0MAEBQ-RPF Rounded
Evaluation Results
| Method | Links | |
|---|---|---|
| BQ-RPF RoundedBudget=1%, Model=Gemma 3 27B, Selection Strategy=Active Selection2026.04 | 0 | |
| BQ-SF RandBudget=1%, Model=Gemma 3 27B, Selection Strategy=Random Selection2026.04 | 0.001 | |
| BQ-RPF RoundedBudget=1%, Target Model=GPT-4o, Selection Strategy=Active Selection + BQ2026.04 | 0.001 | |
| BQ-SF RoundedBudget=1%, Target Model=GPT-4o, Selection Strategy=Active Selection + BQ2026.04 | 0.001 | |
| BQ-SF RandTarget Model=GPT-5.2, Budget=1%, Selection Strategy=Random Selection2026.04 | 0.001 | |
| BQ-SFSelection Strategy=Random Selection, Feature Type=Score Features (SF), Labeling Budget=1%2026.04 | 0.001 | |
| BQ-SFSelection Strategy=Active Selection, Feature Type=Score Features (SF), Rounding=True, Labeling Budget=1%2026.04 | 0.001 | |
| BQ-SF RoundedModel=Claude 3.7 Sonnet, Budget=1%, Selection=Active2026.04 | 0.001 | |
| BQ-RPF RoundedBudget=1%, Target Model=Gemini 3 Flash, Selection=Active Selection + BQ2026.04 | 0.001 | |
| BQ-SF RandModel=Gemma 3 12B, Budget=1%, Selection Strategy=Random Selection2026.04 | 0.001 | |
| BQ-RPFBudget=1%, Target Model=GPT-4o, Selection Strategy=Active Selection + BQ2026.04 | 0.002 | |
| BQ-SFTarget Model=GPT-5.2, Budget=1%, Selection Strategy=Active Selection2026.04 | 0.002 | |
| BQ-SF RandBudget=1%, Target Model=Claude 4.5 Sonnet, Selection Strategy=Random Selection2026.04 | 0.002 | |
| BQ-RPFTarget Model=Claude 4.5 Opus, Budget=1%, Selection Strategy=Active Selection + BQ2026.04 | 0.002 | |
| BQ-SF RandBudget=1%, Target Model=GPT-3.5 Turbo, Selection Type=Random Selection2026.04 | 0.002 | |
| BQ-SFBudget=1%, Target Model=Gemini 3 Pro, Selection Strategy=Active Selection2026.04 | 0.002 | |
| BQ-SFBudget=1%, Target Model=Gemini 3 Flash, Selection=Active Selection + BQ2026.04 | 0.002 | |
| BQ-SFBudget=1%, Target Model=Claude 4.5 Sonnet, Selection Strategy=Active Selection2026.04 | 0.003 | |
| BQ-SF RoundedBudget=1%, Target Model=Claude 4.5 Sonnet, Selection Strategy=Active Selection2026.04 | 0.003 | |
| BQ-SF RandTarget Model=Claude 4.5 Opus, Budget=1%, Selection Strategy=Random Selection + BQ2026.04 | 0.003 | |
| BQ-SF RoundedBudget=1%, Target Model=Gemini 3 Pro, Selection Strategy=Active Selection, Integer Rounding=true2026.04 | 0.003 | |
| BQ-SF RandBudget=1%, Model=Gemini 2.5 Pro2026.04 | 0.003 | |
| BQ-SFBudget=1%, Model=Gemini 2.5 Pro2026.04 | 0.003 | |
| BQ-SFModel=Gemma 3 12B, Budget=1%, Selection Strategy=Active Selection2026.04 | 0.003 | |
| BQ-SFBudget=1%, Target Model=Claude 3.5 Haiku, Selection Strategy=Active Selection + BQ2026.04 | 0.004 | |
| BQ-RPF RoundedBudget=1%, Target Model=Gemini 3 Pro, Selection Strategy=Active Selection, Integer Rounding=true2026.04 | 0.004 | |
| BQ-RPF RoundedBudget=1%, Model=Gemini 2.5 Pro2026.04 | 0.004 | |
| BQ-SF RoundedTarget Model=GPT-5.2, Budget=1%, Selection Strategy=Active Selection2026.04 | 0.005 | |
| BQ-RPF RoundedBudget=1%, Target Model=Claude 4.5 Sonnet, Selection Strategy=Active Selection2026.04 | 0.005 | |
| BQ-SFBudget=1%, Target Model=GPT-3.5 Turbo, Selection Type=Active Selection2026.04 | 0.005 | |
| BQ-RPFSelection Mode=Active Selection, Budget=1%, Target Model=GPT-52026.04 | 0.005 | |
| BQ-SFSelection Mode=Active Selection, Budget=1%, Target Model=GPT-52026.04 | 0.005 | |
| BQ-SF RandBudget=1%, Target Model=Gemini 3 Flash, Selection=Random Selection + BQ2026.04 | 0.005 | |
| BQ-SFBudget=1%, Evaluated Model=GPT-5.1, Selection Strategy=Active Selection + BQ, Rounding=false2026.04 | 0.006 | |
| BQ-SFTarget Model=Claude 4.5 Opus, Budget=1%, Selection Strategy=Active Selection + BQ2026.04 | 0.006 | |
| BQ-SF RoundedBudget=1%, Target Model=Gemini 3 Flash, Selection=Active Selection + BQ2026.04 | 0.006 | |
| BQ-SFBudget=1%, Model=Gemma 3 27B, Selection Strategy=Active Selection2026.04 | 0.007 | |
| BQ-RPF RoundedBudget=1%, Evaluated Model=GPT-5.1, Selection Strategy=Active Selection + BQ, Rounding=true2026.04 | 0.007 | |
| BQ-SF RoundedTarget Model=Claude 4.5 Opus, Budget=1%, Selection Strategy=Active Selection + BQ2026.04 | 0.007 | |
| BQ-SF RandBudget=1%, Target Model=Gemini 3 Pro, Selection Strategy=Random2026.04 | 0.007 | |
| BQ-SF RoundedBudget=1%, Model=Gemini 2.5 Pro2026.04 | 0.007 | |
| BQ-SF RoundedSelection Mode=Active Selection, Budget=1%, Target Model=GPT-52026.04 | 0.007 | |
| BQ-SF RandBudget=1%, Target Model=GPT-4o, Selection Strategy=Random Selection + BQ2026.04 | 0.008 | |
| BQ-SFBudget=1%, Target Model=GPT-4o, Selection Strategy=Active Selection + BQ2026.04 | 0.008 | |
| BQ-SF RandBudget=1%, Target Model=Claude 3.5 Haiku, Selection Strategy=Random Selection + BQ2026.04 | 0.008 | |
| BQ-RPF RoundedModel=Gemma 3 12B, Budget=1%, Selection Strategy=Active Selection2026.04 | 0.008 | |
| BQ-SF RoundedBudget=1%, Evaluated Model=GPT-5.1, Selection Strategy=Active Selection + BQ, Rounding=true2026.04 | 0.009 | |
| BQ-SF RoundedBudget=1%, Target Model=Claude 3.5 Haiku, Selection Strategy=Active Selection + BQ2026.04 | 0.009 | |
| BQ-RPF RoundedTarget Model=Claude 4.5 Opus, Budget=1%, Selection Strategy=Active Selection + BQ2026.04 | 0.009 | |
| BQ-SF RandBudget=1%, Evaluated Model=GPT-5.1, Selection Strategy=Random Selection + BQ2026.04 | 0.01 | |
| BQ-SF RandSelection Mode=Random Selection, Budget=1%, Target Model=GPT-52026.04 | 0.01 | |
| BQ-RPFBudget=1%, Model=Gemini 2.5 Pro2026.04 | 0.011 | |
| BQ-RPF RoundedSelection Mode=Active Selection, Budget=1%, Target Model=GPT-52026.04 | 0.011 | |
| BQ-SF RoundedModel=Gemma 3 12B, Budget=1%, Selection Strategy=Active Selection2026.04 | 0.011 | |
| BQ-SF RoundedBudget=1%, Target Model=GPT-3.5 Turbo, Selection Type=Active Selection2026.04 | 0.012 | |
| BQ-RPFBudget=1%, Target Model=Gemini 3 Pro, Selection Strategy=Active Selection2026.04 | 0.012 | |
| BQ-SFModel=Claude 3.7 Sonnet, Budget=1%, Selection=Active2026.04 | 0.012 | |
| BQ-RPF RoundedBudget=1%, Target Model=GPT-3.5 Turbo, Selection Type=Active Selection2026.04 | 0.013 | |
| BQ-RPFBudget=1%, Model=Gemma 3 27B, Selection Strategy=Active Selection2026.04 | 0.014 | |
| BQ-RPFBudget=1%, Target Model=Claude 4.5 Sonnet, Selection Strategy=Active Selection2026.04 | 0.014 | |
| BQ-RPF RoundedBudget=1%, Target Model=Claude 3.5 Haiku, Selection Strategy=Active Selection + BQ2026.04 | 0.015 | |
| BQ-RPF RandBudget=1%, Target Model=Claude 4.5 Sonnet, Selection Strategy=Random Selection2026.04 | 0.015 | |
| BQ-SF RandModel=Claude 3.7 Sonnet, Budget=1%, Selection=Random2026.04 | 0.016 | |
| BQ-RPF RandTarget Model=Claude 4.5 Opus, Budget=1%, Selection Strategy=Random Selection + BQ2026.04 | 0.017 | |
| BQ-RPF RoundedModel=Claude 3.7 Sonnet, Budget=1%, Selection=Active2026.04 | 0.017 | |
| BQ-RPF RandBudget=1%, Target Model=Gemini 3 Flash, Selection=Random Selection + BQ2026.04 | 0.017 | |
| LR+LUREModel=Gemma 3 12B, Budget=1%2026.04 | 0.017 | |
| BQ-SF RoundedBudget=1%, Model=Gemma 3 27B, Selection Strategy=Active Selection2026.04 | 0.018 | |
| LR+ISBudget=1%, Target Model=GPT-52026.04 | 0.018 | |
| BQ-RPF RandModel=Claude 3.7 Sonnet, Budget=1%, Selection=Random2026.04 | 0.019 | |
| LR+ISBudget=1%, Target Model=Gemini 3 Pro, Selection Strategy=Random2026.04 | 0.021 | |
| BQ-TPFModel=Claude 3.7 Sonnet, Budget=1%, Selection=Active2026.04 | 0.021 | |
| BQ-TPF RandModel=Claude 3.7 Sonnet, Budget=1%, Selection=Random2026.04 | 0.022 | |
| BQ-RPF RoundedTarget Model=GPT-5.2, Budget=1%, Selection Strategy=Active Selection2026.04 | 0.023 | |
| LR+ISBudget=1%, Target Model=Claude 4.5 Sonnet2026.04 | 0.023 | |
| LR+ISBudget=1%, Target Model=GPT-3.5 Turbo2026.04 | 0.023 | |
| LR+ISBudget=1%, Target Model=Gemini 3 Flash2026.04 | 0.023 | |
| BQ-RPF RandBudget=1%, Model=Gemma 3 27B, Selection Strategy=Random Selection2026.04 | 0.026 | |
| LR+ISBudget=1%, Target Model=GPT-4o, Selection Strategy=Random Selection2026.04 | 0.026 | |
| LR+LUREBudget=1%, Target Model=GPT-4o, Selection Strategy=Random Selection2026.04 | 0.026 | |
| LR+ISTarget Model=GPT-5.2, Budget=1%, Selection Strategy=Random Selection2026.04 | 0.026 | |
| BQ-RPF RandBudget=1%, Target Model=GPT-4o, Selection Strategy=Random Selection + BQ2026.04 | 0.027 | |
| BQ-TPFBudget=1%, Target Model=Claude 3.5 Haiku, Selection Strategy=Active Selection + BQ2026.04 | 0.027 | |
| BQ-RPF RandBudget=1%, Target Model=Gemini 3 Pro, Selection Strategy=Random2026.04 | 0.027 | |
| BQ-RPFBudget=1%, Target Model=Gemini 3 Flash, Selection=Active Selection + BQ2026.04 | 0.027 | |
| BQ-RPFModel=Gemma 3 12B, Budget=1%, Selection Strategy=Active Selection2026.04 | 0.027 | |
| BQ-TPF RandBudget=1%, Target Model=GPT-4o, Selection Strategy=Random Selection + BQ2026.04 | 0.028 | |
| BQ-TPFBudget=1%, Target Model=GPT-4o, Selection Strategy=Active Selection + BQ2026.04 | 0.029 | |
| LR+ISTarget Model=Claude 4.5 Opus, Budget=1%2026.04 | 0.029 | |
| LR+ISBudget=1%, Evaluated Model=GPT-5.1, Selection Strategy=Random Selection2026.04 | 0.03 | |
| BQ-RPF RandBudget=1%, Evaluated Model=GPT-5.1, Selection Strategy=Random Selection + BQ2026.04 | 0.03 | |
| BQ-TPF RandBudget=1%, Target Model=Claude 3.5 Haiku, Selection Strategy=Random Selection + BQ2026.04 | 0.03 | |
| BQ-RPF RandBudget=1%, Target Model=GPT-3.5 Turbo, Selection Type=Random Selection2026.04 | 0.031 | |
| BQ-RPF RandTarget Model=GPT-5.2, Budget=1%, Selection Strategy=Random Selection2026.04 | 0.032 | |
| BQ-RPF RandBudget=1%, Target Model=Claude 3.5 Haiku, Selection Strategy=Random Selection + BQ2026.04 | 0.032 | |
| LR+ISModel=Claude 3.7 Sonnet, Budget=1%2026.04 | 0.032 | |
| BQ-RPF RandBudget=1%, Model=Gemini 2.5 Pro2026.04 | 0.032 | |
| BQ-RPF RandModel=Gemma 3 12B, Budget=1%, Selection Strategy=Random Selection2026.04 | 0.032 | |
| BQ-TPFBudget=1%, Target Model=GPT-3.5 Turbo, Selection Type=Active Selection2026.04 | 0.033 | |
| BQ-TPFBudget=1%, Model=Gemma 3 27B, Selection Strategy=Active Selection2026.04 | 0.034 |