List your company

Question Answering on NQ (EM, F1)

59.46EMPrompt tuning
1.126416.270731.41546.5593Feb 3, 2026

Evaluation results

MethodLinks
Prompt tuningBackbone=Llama-3.2-3B-Base, Evaluation Protocol=Prompt tuning, Context Setting=w/ context2026.0259.4673.32
Prompt tuningBackbone=Llama-3.2-1B-Base, Evaluation Protocol=Prompt tuning, Context Setting=w/ context2026.0253.0366.59
ComprExITBackbone=Llama-3.2-3B-Base, Evaluation Protocol=Context Compression2026.0252.2368.95
500xBackbone=Llama-3.2-3B-Base, Evaluation Protocol=Context Compression2026.0251.3266.6
ICAEBackbone=Llama-3.2-3B-Base, Evaluation Protocol=Context Compression2026.0250.8766.75
ComprExITBackbone=Llama-3.2-1B-Base, Evaluation Protocol=Context Compression2026.0245.9363.88
BeaconBackbone=Llama-3.2-3B-Base, Evaluation Protocol=Context Compression2026.0245.0559.56
ICAEBackbone=Llama-3.2-1B-Base, Evaluation Protocol=Context Compression2026.0242.2658
Zero-shotBackbone=Llama-3.2-3B-Base, Evaluation Protocol=Zero-shot, Context Setting=w/ context2026.0232.7452.09
BeaconBackbone=Llama-3.2-1B-Base, Evaluation Protocol=Context Compression2026.0221.0437.05
Zero-shotBackbone=Llama-3.2-1B-Base, Evaluation Protocol=Zero-shot, Context Setting=w/ context2026.0218.1438.9
Zero-shotBackbone=Llama-3.2-3B-Base, Evaluation Protocol=Zero-shot, Context Setting=w/o context2026.0210.3118.47
500xBackbone=Llama-3.2-1B-Base, Evaluation Protocol=Context Compression2026.025.5813.89
Zero-shotBackbone=Llama-3.2-1B-Base, Evaluation Protocol=Zero-shot, Context Setting=w/o context2026.023.378.43