Claim Verification on HoVer (test)
73.1AccuracyTOME-2
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TOME-2#Params=220M, #Encoded=12021.10 | 73.1 | — | |
| TOME-2#Params=220M, #Encoded=12021.10 | 73.1 | — | |
| TOME-1#Params=220M, #Encoded=12021.10 | 72.8 | — | |
| TOME-1#Params=220M, #Encoded=12021.10 | 72.8 | — | |
| ClaimPKGBackbone=Llama 3, Strategy=Support-Predicted2025.05 | 70.7 | 14.3 | |
| ClaimPKGBackbone=Qwen 2.5, Strategy=Support-Predicted2025.05 | 69.4 | 15.7 | |
| HERA (Qwen-3-14B)category=Ours2026.04 | 67.35 | — | |
| ExSearchcategory=Advanced RAG, Backbone=Qwen2.5-7B2026.04 | 66.62 | — | |
| EaE#Params=360M, #Encoded=12021.10 | 66.6 | — | |
| Entities as Experts#Params=360M, #Encoded=12021.10 | 66.6 | — | |
| ClaimPKGBackbone=Llama 3, Strategy=Zero-shot CoT2025.05 | 66.6 | — | |
| REALM#Params=330M, #Encoded=52021.10 | 66.1 | — | |
| REALM#Params=330M, #Encoded=52021.10 | 66.1 | — | |
| ClaimPKGBackbone=Qwen 2.5, Strategy=Zero-shot CoT2025.05 | 65.3 | — | |
| HERA (Llama-3.1-8)category=Ours2026.04 | 60.31 | — | |
| AceSearcher-8Bcategory=Advanced RAG2026.04 | 59.37 | — | |
| DeepResearcher-7Bcategory=Advanced RAG2026.04 | 58.19 | — | |
| MMAO-RAG-8Bcategory=Advanced RAG2026.04 | 58.19 | — | |
| InstructRAG-8Bcategory=Advanced RAG2026.04 | 56.76 | — | |
| IRCoTcategory=Advanced RAG, Backbone=GPT-4o-mini2026.04 | 55.37 | — | |
| R1-Searcher 7Bcategory=Advanced RAG2026.04 | 51.56 | — | |
| Iter-RetGencategory=Advanced RAG, Backbone=Qwen2.5-7B2026.04 | 51.29 | — | |
| Plan-RAGcategory=Advanced RAG, Backbone=8b2026.04 | 50.33 | — | |
| Qwen3-8Bcategory=Single-turn Retrieval2026.04 | 50.15 | — | |
| Llama3-3.1-8B-instructcategory=Single-turn Retrieval2026.04 | 49.9 | — | |
| Qwen3-14Bcategory=Single-turn Retrieval2026.04 | 49.05 | — | |
| GPT-4o-minicategory=Single-turn Retrieval2026.04 | 47.95 | — | |
| Search-r1-7Bcategory=Advanced RAG2026.04 | 47.76 | — | |
| Qwen2.5-14B-Instructcategory=Single-turn Retrieval2026.04 | 45.5 | — | |
| CORAG-8B (Greedy)category=Advanced RAG2026.04 | 40.82 | — | |
| SELF-RAGcategory=Advanced RAG, Backbone=llama2-7B2026.04 | 40.52 | — |