Failure Discovery on StrategyQA
99.5Topic EntropyRand-Anchor-Gen
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Rand-Anchor-GenTarget Model=Gemini 2.5 Flash, Generator=Gemini 3 Pro, Iterations=1002026.04 | 99.5 | 98 | 99 | 22.3 | 4.3 | |
| Rand-GenTarget Model=Gemini 2.5 Flash, Generator=Gemini 3 Pro, Iterations=1002026.04 | 99.3 | 57 | 64 | 30.3 | 3.7 | |
| Rand-T-GenTarget Model=Gemini 2.5 Flash, Generator=Gemini 3 Pro, Iterations=1002026.04 | 99.2 | 87 | 71 | 20 | 4.3 | |
| TSS-RPFTarget Model=Gemini 2.5 Flash, Generator=Gemini 3 Pro, Iterations=1002026.04 | 99 | 100 | 74 | 26 | 4.3 | |
| SS-Gen-TPFTarget Model=Gemini 2.5 Flash, Generator=Gemini 3 Pro, Iterations=1002026.04 | 96.8 | 97 | 73 | 33 | 2 | |
| TSS-TPFTarget Model=Gemini 2.5 Flash, Generator=Gemini 3 Pro, Iterations=1002026.04 | 96.5 | 98 | 73 | 33 | 4.7 | |
| SS-Gen-RPFTarget Model=Gemini 2.5 Flash, Generator=Gemini 3 Pro, Iterations=1002026.04 | 95.2 | 95 | 71 | 40.3 | 3 | |
| SS-RPFTarget Model=Gemini 2.5 Flash, Generator=Gemini 3 Pro, Iterations=1002026.04 | 71.9 | 71 | 54 | 48 | 3 | |
| RandTarget Model=Gemini 2.5 Flash, Generator=Gemini 3 Pro, Iterations=1002026.04 | 68.4 | 100 | 59 | 17.3 | 9.3 | |
| SS-TPFTarget Model=Gemini 2.5 Flash, Generator=Gemini 3 Pro, Iterations=1002026.04 | 65 | 75 | 51 | 46 | 4 |