TriviaQA

Overview edit

Variants: TriviaQA, KILT: TriviaQA

Associated Benchmarks

This dataset is used in 4 benchmarks:

• Question Answering - Metrics: EM, F1
• Text Generation - Metrics: acc
• Question Generation - Metrics: QAE, R-QAE
• Open-Domain Question Answering - Metrics: Exact Match

Recent Benchmark Submissions

Research Papers

Recent papers with results on this dataset:

• Search-o1: Agentic Search-Enhanced Large Reasoning Models (2025) -
• SHAKTI: A 2.5 Billion Parameter Small Language Model Optimized for Edge AI and Low-Resource Environments (2024) -
• RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMs (2024) -
• Understand What LLM Needs: Dual Preference Alignment for Retrieval-Augmented Generation (2024) -
• Breaking the Ceiling of the LLM Community by Treating Token Generation as a Classification for Ensembling (2024) -
• Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM (2024) -
• ChatQA: Surpassing GPT-4 on Conversational QA and RAG (2024) -
• Mistral 7B (2023) -
• RA-DIT: Retrieval-Augmented Dual Instruction Tuning (2023) -
• Llama 2: Open Foundation and Fine-Tuned Chat Models (2023) -
• PaLM 2 Technical Report (2023) -
• GPT-4 Technical Report (2023) -
• LLaMA: Open and Efficient Foundation Language Models (2023) -
• REPLUG: Retrieval-Augmented Black-Box Language Models (2023) -
• FiE: Building a Global Probability Space by Leveraging Early Fusion in Encoder for Open-Domain Question Answering (2022) -
• DyREx: Dynamic Query Representation for Extractive Question Answering (2022) -
• PaLM: Scaling Language Modeling with Pathways (2022) -
• LinkBERT: Pretraining Language Models with Document Links (2022) -
• GLaM: Efficient Scaling of Language Models with Mixture-of-Experts (2021) -
• Mention Memory: incorporating textual knowledge into Transformers through entity mention attention (2021) -