List your company

Question Answering on Average of 5 Datasets

78.9Average ScoreGUARANTRAG
58.72463.96269.274.438Apr 9, 2026

Evaluation results

MethodLinks
GUARANTRAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=RetroMAE2026.0478.9————
GUARANTRAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=SPLADE2026.0477.1————
GUARANTRAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=HyDE2026.0476.6————
P-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=RetroMAE2026.0475.8————
P-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=SPLADE2026.0474.4————
GUARANTRAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=BM252026.0474.2————
SOLARBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=RetroMAE2026.0474.1————
P-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=HyDE2026.0473.7————
DRAGINBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=RetroMAE2026.0473.6————
FLAREBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=RetroMAE2026.0473.1————
SOLARBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=SPLADE2026.0472.5————
RQ-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=RetroMAE2026.0472.4————
DRAGINBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=SPLADE2026.0472————
SOLARBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=HyDE2026.0471.6————
FLAREBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=SPLADE2026.0471.6————
GUARANTRAG-7BRetriever=BM252026.0471.6————
P-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=BM252026.0471.3————
DRAGINBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=HyDE2026.0471.2————
FLAREBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=HyDE2026.0470.9————
RQ-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=SPLADE2026.0470.8————
SelfRAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=RetroMAE2026.0470.6————
GUARANTRAGModel Variant=w/o DPO Training, Retriever=BM252026.0470.5————
RQ-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=HyDE2026.0470.1————
SOLARBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=BM252026.0469.4————
DA-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=RetroMAE2026.0469.4————
SelfRAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=SPLADE2026.0469.1————
DRAGINBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=BM252026.0469.1————
GUARANTRAGModel Variant=w/o Relevance Control, Retriever=BM252026.0469.1————
GUARANTRAGModel Variant=w/o Segment-Level Fusion, Retriever=BM252026.0468.9————
FLAREBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=BM252026.0468.7————
SelfRAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=HyDE2026.0468.5————
DA-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=SPLADE2026.0468.2————
RQ-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=BM252026.0467.9————
DA-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=HyDE2026.0467.3————
GUARANTRAG-4BRetriever=BM252026.0467.3————
GUARANTRAGModel Variant=w/o Length Control, Retriever=BM252026.0466.8————
SelfRAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=BM252026.0466.2————
RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=RetroMAE2026.0465.7————
DA-RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=BM252026.0465.1————
RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=SPLADE2026.0464.3————
RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=HyDE2026.0464————
RAGBase Model=Qwen3-14B-T, Thinking Mode=true, Retriever=BM252026.0462.1————
Qwen3-14B-TThinking Mode=true, Retriever=BM252026.0459.5————
Qwen3-14B-TThinking Mode=true, Retriever=SPLADE2026.0459.5————
Qwen3-14B-TThinking Mode=true, Retriever=RetroMAE2026.0459.5————
Qwen3-14B-TThinking Mode=true, Retriever=HyDE2026.0459.5————
CompActReader=Llama-3.1-8B Instruct, top-k chunks=202026.03—30.339.40.296.3
EXITReader=Llama-3.1-8B Instruct, top-k chunks=202026.03—30.138.40.756.3
LongLLMLinguaReader=Llama-3.1-8B Instruct, top-k chunks=202026.03—31.3400.760.8
LooCompReader=Llama-3.1-8B Instruct, top-k chunks=202026.03—32.441.46.390.3
ProvenceReader=Llama-3.1-8B Instruct, top-k chunks=202026.03—3139.85.284.5
RECOMP-absReader=Llama-3.1-8B Instruct, top-k chunks=202026.03—29380.898.2
RECOMP-extReader=Llama-3.1-8B Instruct, top-k chunks=202026.03—25.833.627.788.1
RefinerReader=Llama-3.1-8B Instruct, top-k chunks=202026.03—29.337.80.294.3