List your company

Long-document and Multi-hop Question Answering on 2WikiMultiHopQA

50.39F1 ScoreD2Latom
9.684420.252230.8241.3878Jun 10, 2026

Evaluation results

MethodLinks
D2LatomLLM Backbone=Gemma-2-2B-It, Evaluation Protocol=In-domain2026.0650.3950.25
D2LrawLLM Backbone=Gemma-2-2B-It, Evaluation Protocol=In-domain2026.0649.8449.79
D2LatomLLM Backbone=Qwen3-4B-Instruct, Evaluation Protocol=In-domain2026.0647.4347.43
D2LrawLLM Backbone=Qwen3-4B-Instruct, Evaluation Protocol=In-domain2026.0646.0646.03
Base+ctxLLM Backbone=Qwen3-4B-Instruct, Evaluation Protocol=In-domain2026.0645.3145.27
Doc2AtomLLM Backbone=Gemma-2-2B-It, Evaluation Protocol=In-domain2026.0639.6239.62
Base+ctxLLM Backbone=Gemma-2-2B-It, Evaluation Protocol=In-domain2026.0637.6937.69
D2LckptLLM Backbone=Gemma-2-2B-It, Evaluation Protocol=In-domain2026.0636.0836.08
Doc2AtomLLM Backbone=Qwen3-4B-Instruct, Evaluation Protocol=In-domain2026.0631.7431.74
D2LckptLLM Backbone=Qwen3-4B-Instruct, Evaluation Protocol=In-domain2026.0611.2511.17