List your company

Speech-to-Text Question Answering on OBQA

83.08AccuracyAPin
9.572828.656447.7466.8236Jan 30, 2026Feb 10, 2026Feb 21, 2026Mar 5, 2026Mar 16, 2026Mar 27, 2026Apr 8, 2026

Evaluation results

MethodLinks
APinSetting=Aggressive (τin=0.80, τdeep=0.70), lin=✓, ldeep=-, FRR=86.28, Pre. GFLOPs=351.18, FLOPs Ratio=86.242026.0483.08
APinSetting=Conservative (τin=0.90, τdeep=0.80), lin=✓, ldeep=-, FRR=96.44, Pre. GFLOPs=392.69, FLOPs Ratio=96.432026.0482.64
APdeepSetting=Conservative (τin=0.90, τdeep=0.80), lin=-, ldeep=✓, FRR=71.18, Pre. GFLOPs=390.42, FLOPs Ratio=95.872026.0482.64
DAPSetting=Aggressive (τin=0.80, τdeep=0.70), lin=✓, ldeep=✓, FRR=40.71, Pre. GFLOPs=324.64, FLOPs Ratio=79.722026.0482.42
APdeepSetting=Aggressive (τin=0.80, τdeep=0.70), lin=-, ldeep=✓, FRR=48.81, Pre. GFLOPs=377.39, FLOPs Ratio=92.682026.0482.2
Vanillalin=-, ldeep=-, FRR=100.0, Pre. GFLOPs=407.22, FLOPs Ratio=100.02026.0481.98
DAPSetting=Conservative (τin=0.90, τdeep=0.80), lin=✓, ldeep=✓, FRR=68.19, Pre. GFLOPs=376.14, FLOPs Ratio=92.362026.0481.98
Phi-4-MultimodalType=AR, S→S capability=false2026.0165.9
DIFFUSPEECHType=Diff., S→S capability=true2026.0151.3
Qwen2-AudioType=AR, S→S capability=false2026.0149.5
MinMoType=AR, S→S capability=true2026.0144.5
DiFFAType=Diff., S→S capability=false2026.0135.6
Llama-Omni2Type=AR, S→S capability=true2026.0128.1
MoshiType=AR, S→S capability=true2026.0126
SpiritLMType=AR, S→S capability=true2026.0121.7
SpeechGPTType=AR, S→S capability=true2026.0112.4