List your company

Long-context Understanding on LongBench HotpotQA

52.15F1 ScoreBase+ctx
9.395620.495331.59542.6947Jun 10, 2026

Evaluation results

MethodLinks
Base+ctxLLM Backbone=Qwen3-4B-Instruct, Evaluation Protocol=Zero-shot2026.0652.1551.78
Base+ctxLLM Backbone=Gemma-2-2B-It, Evaluation Protocol=Zero-shot2026.064140.78
Doc2AtomLLM Backbone=Gemma-2-2B-It, Evaluation Protocol=Zero-shot2026.0627.9627.82
D2LckptLLM Backbone=Gemma-2-2B-It, Evaluation Protocol=Zero-shot2026.0626.7926.37
D2LrawLLM Backbone=Qwen3-4B-Instruct, Evaluation Protocol=Zero-shot2026.0625.8825.88
D2LatomLLM Backbone=Qwen3-4B-Instruct, Evaluation Protocol=Zero-shot2026.0625.4325.24
Doc2AtomLLM Backbone=Qwen3-4B-Instruct, Evaluation Protocol=Zero-shot2026.0624.7724.67
D2LrawLLM Backbone=Gemma-2-2B-It, Evaluation Protocol=Zero-shot2026.0622.322.21
D2LatomLLM Backbone=Gemma-2-2B-It, Evaluation Protocol=Zero-shot2026.0619.4919.49
D2LckptLLM Backbone=Qwen3-4B-Instruct, Evaluation Protocol=Zero-shot2026.0611.0410.5