List your company

Question Answering on 2WikiHop (test)

67.1Exact Match (EM)Oracle
24.56435.60746.6557.693May 18, 2026

Evaluation results

MethodLinks
OracleBackbone=Llama-3.1-8B, Retrieval Infrastructure=Wikipedia corpus, FAISS IVF index, Contriever embeddings2026.0567.173.4453.14824.64.4
Ours (Predictive)Backbone=Llama-3.1-8B, Retrieval Infrastructure=Wikipedia corpus, FAISS IVF index, Contriever embeddings2026.0565.972.11085.25914.47.6
DRAGINBackbone=Llama-3.1-8B, Retrieval Infrastructure=Wikipedia corpus, FAISS IVF index, Contriever embeddings2026.0565.771.91786.46411.49.5
Sync-RAGBackbone=Llama-3.1-8B, Retrieval Infrastructure=Wikipedia corpus, FAISS IVF index, Contriever embeddings2026.0565.471.82879.2868.213.3
Adaptive-RAGBackbone=Llama-3.1-8B, Retrieval Infrastructure=Wikipedia corpus, FAISS IVF index, Contriever embeddings2026.0564.871.22157.16810.510.4
Self-RAGBackbone=Llama-3.1-8B, Retrieval Infrastructure=Wikipedia corpus, FAISS IVF index, Contriever embeddings2026.0564.270.52347.8729.411.5
FLAREBackbone=Llama-3.1-8B, Retrieval Infrastructure=Wikipedia corpus, FAISS IVF index, Contriever embeddings2026.0563.870.12066.87110.610.3
PipeRAGBackbone=Llama-3.1-8B, Retrieval Infrastructure=Wikipedia corpus, FAISS IVF index, Contriever embeddings, Retrieval Interval=322026.0563.469.71185.666.8138.4
Entropy-ThresholdBackbone=Llama-3.1-8B, Retrieval Infrastructure=Wikipedia corpus, FAISS IVF index, Contriever embeddings2026.0561.368.22688.7828.113.6
No-RAGBackbone=Llama-3.1-8B2026.0526.233.1422.3017.17