List your company

ResearchFormerly WizWand

NarrativeQA

Benchmarks

TaskDatasetSOTA resultTrendResults

Question Answering

NarrativeQA

38.12F1 Score

124

Question Answering

NarrativeQA (test)

76.2ROUGE-L

88

Question Answering

NarrativeQA

21.96Score

40

Question Answering

NarrativeQA

15EM

38

Long-context Question Answering

NarrativeQA

53.56F1 Score

38

Long-context Question Answering

NarrativeQA

22SubEM

36

Question Answering

NarrativeQA

33.61F1

36

Question Answering

NarrativeQA LongBench

15.92F1 Score

29

Question Answering

NarrativeQA

21.5BLEU-1

28

Single-hop Question Answering

NarrativeQA

23.89Score

22

Long-context Question Answering

NarrativeQA Passage Split

32.64Score

18

Long-context Question Answering

NarrativeQA Fixed Chunk 2048

32.64Score

18

Question Answering

NarrativeQA

34.79EM

18

Question Answering

NarrativeQA No Trun. latest (test)

23.96Average Score

18

Question Answering

NarrativeQA

28.94F1 Score

16

Question Answering

NarrativeQA

45Rouge-L

15

Discourse Question Answering

NarrativeQA

26.6F1 Score

14

Long narrative understanding QA

NarrativeQA

55Accuracy

14

Traceback (Prompt Injection Attacks)

NarrativeQA

98Precision

13

Question Answering

NarrativeQA

355.12TTFT (ms)

12

Question Answering

NarrativeQA

0.58Peak GPU Memory (GB)

12

Question Answering

NarrativeQA LongBench 32K context

17.2F1 Score

12

Multi-session Retrieval-Augmented Generation

NarrativeQA (test)

38.4F1 Score

12

Document Retrieval

NarrativeQA (test)

61.7nDCG@10

12

Long-context retrieval

NarrativeQA LongEmbed

71.5nDCG@10

11