Knowledge QA on TriviaQA
72.93ScoreLLaDA2.1-flash
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| LLaDA2.1-flashInference Mode=Q Mode2026.02 | 72.93 | 2.92 | — | |
| LLaDA2.1-flashInference Mode=S Mode2026.02 | 72.55 | 4.3 | — | |
| Ling-flash-2.02026.02 | 69.76 | 1 | — | |
| LLaDA2.0-flash2026.02 | 66.88 | 1.94 | — | |
| Qwen3-30B-A3B-Inst-25072026.02 | 65.61 | 1 | — | |
| Ling-mini-2.02026.02 | 55.63 | — | — | |
| LLaDA2.1-minimode=Q Mode2026.02 | 54.24 | 2.02 | — | |
| LLaDA2.1-minimode=S Mode2026.02 | 53.33 | 3.21 | — | |
| Qwen3-8Bno think=true2026.02 | 52.51 | — | — | |
| LLaDA2.0-mini2026.02 | 51.33 | 1.54 | — | |
| Original (T2T)Backbone=LLaDA2.1-mini, Inference Strategy=Text-to-Text2026.04 | — | — | 30.54 | |
| T2MBackbone=LLaDA2.1-mini, Inference Strategy=Token-to-Mask, Remasking Strategy=LOWPROB, τ=0.3, Cmax=1, ρmax=0.252026.04 | — | — | 30.96 |