Share your thoughts, 1 month free Claude Pro on usSee more

Question Answering on LongBench Qasper

0.4459F1

InnerQBase

Updated 11d ago

Evaluation Results

Method	Links
InnerQBase 2026.02		0.4459
Baseline 2026.02		0.4458
KIVI 2026.02		0.4408
KIVISink 2026.02		0.4361
InnerQHybrid 2026.02		0.424
InnerQSmall 2026.02		0.4236
Baseline 2026.02		0.4063
InnerQBase 2026.02		0.4032
InnerQHybrid 2026.02		0.3959
InnerQSmall 2026.02		0.3884
KIVISink 2026.02		0.3849
KIVI 2026.02		0.3709
PIC 2026.01		0.2838
M-RAG†-P 2026.01		0.2693
M-RAG†-S 2026.01		0.269
Semantic 2026.01		0.2683
M-RAG-S 2026.01		0.2655
DOS 2026.01		0.2635
PIC 2026.01		0.253
Semantic 2026.01		0.2482
M-RAG-P 2026.01		0.2456
Fixed-Size 2026.01		0.2425
M-RAG†-P 2026.01		0.237
M-RAG-P 2026.01		0.2317
M-RAG-S 2026.01		0.2308
DOS 2026.01		0.2269
M-RAG†-S 2026.01		0.2224
InnerQHybrid 2026.02		0.2218
InnerQBase 2026.02		0.2184
KIVI 2026.02		0.2111
Baseline 2026.02		0.2099
InnerQHybrid 2026.02		0.2043
KIVISink 2026.02		0.203
Baseline 2026.02		0.2029
InnerQSmall 2026.02		0.1984
Fixed-Size 2026.01		0.1908
InnerQBase 2026.02		0.1837
M-RAG†-S 2026.01		0.182
M-RAG†-P 2026.01		0.1806
KIVI 2026.02		0.18
InnerQBase 2026.02		0.1785
InnerQSmall 2026.02		0.175
KIVISink 2026.02		0.1723
Baseline 2026.02		0.1659
InnerQHybrid 2026.02		0.1656
Semantic 2026.01		0.1654
DOS 2026.01		0.1643
PIC 2026.01		0.1581
KIVISink 2026.02		0.1548
KIVI 2026.02		0.1509
M-RAG-P 2026.01		0.1436
M-RAG-S 2026.01		0.139
Fixed-Size 2026.01		0.1316
InnerQSmall 2026.02		0.1312
CurvPrune 2026.02		0.102
BM25 2026.02		0.095
BM25+Tex 2026.02		0.093
Head+Tail 2026.02		0.082
Recency 2026.02		0.077
Sel. Ctx 2026.02		0.074
Random 2026.02		0.063
LLMLingua 2026.02		0.059