List your company

Accuracy on MuSiQue (Question Answering)

71.5Accuracy (MuSiQue QA)BRANE
2.96420.75738.5556.343Mar 25, 2026Apr 9, 2026Apr 24, 2026May 10, 2026May 25, 2026Jun 9, 2026Jun 25, 2026

Evaluation results

MethodLinks
BRANEBudget Goal=50%2026.0571.5
CARROT-KNN (Extended)Budget Goal=50%2026.0568.9
Best Static Config. (Murakkab)Budget Goal=50%2026.0568.17
CARROT-RoBERTa (Extended)Budget Goal=50%2026.0567.5
CARROT-RoBERTa (LLM-only)Budget Goal=50%2026.0565.6
CARROT-KNN (LLM-only)Budget Goal=50%2026.0558.3
Adaptive-RAGBudget Goal=50%2026.0556.5
METISBudget Goal=50%2026.0548.5
IRCoTBackbone=GPT-4o2026.0344.2
Iter-RetGenBackbone=GPT-4o2026.0342
MARCHBackbone=Llama3.1-8B, Prompting=10-Shots2026.0340.8
RAG-StarBackbone=GPT-4o2026.0340
MARCHBackbone=Llama3.1-8B, Prompting=CoT2026.0336.2
FE-SFT & FC-RLBackbone=Youtu-LLM-2B-Base, Framework=WM-AMT, Optimization Stage=RL stage2026.0634.2
IWM & RLBackbone=Youtu-LLM-2B-Base, Framework=Standard, Optimization Stage=RL stage2026.0633.2
FE-SFT & RLBackbone=Youtu-LLM-2B-Base, Framework=WM-AMT, Optimization Stage=RL stage2026.0632.6
RQ-RAGBackbone=GPT-4o2026.0331.6
State-SFT & RL†Backbone=Youtu-LLM-2B-Base, Framework=Standard, Optimization Stage=RL stage2026.0631.4
State-SFT & RL‡Backbone=Youtu-LLM-2B-Base, Framework=WM-AMT, Optimization Stage=RL stage2026.0630.4
SFT & RL*Backbone=Youtu-LLM-2B-Base, Framework=Standard, Optimization Stage=RL stage2026.0630.2
RAG-StandardBackbone=GPT-4o2026.0329.8
FE-SFT & RL*Backbone=Youtu-LLM-2B-Base, Framework=Standard, Optimization Stage=RL stage2026.0629.8
IRCoTBackbone=Llama3.1-8B2026.0329.2
CoTBackbone=GPT-4o2026.0326.8
FE-SFTBackbone=Youtu-LLM-2B-Base, Framework=WM-AMT, Optimization Stage=SFT stage2026.0625.6
IWMBackbone=Youtu-LLM-2B-Base, Framework=Standard, Optimization Stage=SFT stage2026.0625.4
SFT & RLBackbone=Youtu-LLM-2B-Base, Framework=WM-AMT, Optimization Stage=RL stage2026.0625.4
FE-SFT*Backbone=Youtu-LLM-2B-Base, Framework=Standard, Optimization Stage=SFT stage2026.0623.2
RAFTBackbone=Llama3.1-8B2026.0322
SFT*Backbone=Youtu-LLM-2B-Base, Framework=Standard, Optimization Stage=SFT stage2026.0621
SFTBackbone=Youtu-LLM-2B-Base, Framework=WM-AMT, Optimization Stage=SFT stage2026.0620.6
GenGroundBackbone=7B-class2026.0320.2
BaselineBackbone=Llama3.1-8B2026.035.6