Open-domain Question Answering on TriviaQA (TQA)
0.4177AccuracyUR-RAG
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| UR-RAGLLM=Qwen-14B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.4177 | 0.4203 | |
| RePlugLLM=Qwen-14B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.401 | 0.4077 | |
| LURE-RAG (w/o topics)LLM=Qwen-14B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.401 | 0.4017 | |
| LURE-RAGLLM=Qwen-14B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.3994 | 0.4039 | |
| Gemma3 1BParameter Count=1B, Evaluation Protocol=5-shot2026.03 | 0.398 | — | |
| k-shotLLM=Qwen-14B, Rank-By=Relevance, Retrieval Model=Contriever2026.01 | 0.3966 | 0.3954 | |
| UR-RAGLLM=phi-3.8B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.3699 | 0.3497 | |
| LURE-RAGLLM=phi-3.8B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.3635 | 0.3304 | |
| LURE-RAG (w/o topics)LLM=phi-3.8B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.3628 | 0.322 | |
| MobileLLM-Flash 1.4BParameter Count=1.4B, Evaluation Protocol=5-shot2026.03 | 0.3606 | — | |
| RePlugLLM=phi-3.8B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.3582 | 0.34 | |
| 0-shotLLM=Qwen-14B, Rank-By=Relevance, Retrieval Model=Contriever2026.01 | 0.3566 | 0.3587 | |
| k-shotLLM=phi-3.8B, Rank-By=Relevance, Retrieval Model=Contriever2026.01 | 0.3509 | 0.3137 | |
| LFM2 1.2BParameter Count=1.2B, Evaluation Protocol=5-shot2026.03 | 0.335 | — | |
| UR-RAGLLM=llama-1B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.3184 | 0.3098 | |
| RePlugLLM=llama-1B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.3111 | 0.3042 | |
| LURE-RAGLLM=llama-1B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.3059 | 0.2951 | |
| LURE-RAG (w/o topics)LLM=llama-1B, Rank-By=Utility, Retrieval Model=Contriever2026.01 | 0.2994 | 0.2897 | |
| k-shotLLM=llama-1B, Rank-By=Relevance, Retrieval Model=Contriever2026.01 | 0.2834 | 0.262 | |
| 0-shotLLM=phi-3.8B, Rank-By=Relevance, Retrieval Model=Contriever2026.01 | 0.2779 | 0.2214 | |
| MobileLLM-Flash 650MParameter Count=650M, Evaluation Protocol=5-shot2026.03 | 0.2449 | — | |
| Llama3.2 1BParameter Count=1B, Evaluation Protocol=5-shot2026.03 | 0.2381 | — | |
| 0-shotLLM=llama-1B, Rank-By=Relevance, Retrieval Model=Contriever2026.01 | 0.2258 | 0.2137 | |
| LFM2 700MParameter Count=700M, Evaluation Protocol=5-shot2026.03 | 0.225 | — | |
| MobileLLM-Flash 350MParameter Count=350M, Evaluation Protocol=5-shot2026.03 | 0.1984 | — | |
| Gemma3 270MParameter Count=270M, Evaluation Protocol=5-shot2026.03 | 0.154 | — | |
| LFM2 350MParameter Count=350M, Evaluation Protocol=5-shot2026.03 | 0.1497 | — | |
| Qwen3 0.6BParameter Count=0.6B, Evaluation Protocol=5-shot2026.03 | 0.0241 | — |