List your company

Research

Benchmarks for “question answering”

Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.

9,723 benchmarksPage 319 of 389

Reset

Search results

TaskDatasetDomainTrendResultsLast update
Retrieval-Augmented Question AnsweringMMLU-Proretrieval search ragnlp4Jun 11, 2026
Question AnsweringBoolQnlp4Jun 11, 2026
Visual Question AnsweringVQA v2multimodalcomputer vision+14Jun 11, 2026
Multiple Choice Question AnsweringGPQA, ARC, MMLU (OOD)nlp4Jun 11, 2026
Question AnsweringARC Easynlp4Jun 11, 2026
Knowledge-intensive Question AnsweringAverage (CRAG, NQ, HotpotQA, MuSiQue)nlpretrieval search rag4Jun 11, 2026
Knowledge-intensive Question AnsweringMuSiQuenlpretrieval search rag4Jun 11, 2026
Multiple-choice Question AnsweringMMLUnlp4Jun 8, 2026
Question AnsweringSciQ (held-out)nlp4Jun 8, 2026
Graph-based Question Answering23 Queries (Hold-out)graphnlp4Jun 5, 2026
Out-of-distribution Question AnsweringCPIQA Adversarial Attacknlpllm chat instruction4Jun 4, 2026
Out-of-distribution Question AnsweringAEROnlpllm chat instruction4Jun 4, 2026
Out-of-distribution Question AnsweringCPIQA Cleannlpllm chat instruction4Jun 4, 2026
Out-of-distribution Question AnsweringAERO Cleannlpllm chat instruction4Jun 4, 2026
Video Question AnsweringVidCapBenchmultimodalcomputer vision4Jun 4, 2026
Video Question AnsweringPLM-SGQAmultimodalcomputer vision4Jun 4, 2026
Video Question AnsweringActivityNet-QAmultimodalcomputer vision4Jun 4, 2026
Video Question AnsweringNExT-QA openmultimodalcomputer vision4Jun 4, 2026
Truthful Question AnsweringTruthfulQAllm chat instructionnlp4Jun 4, 2026
Long-term Memory Question AnsweringLoCoMo (overall)llm chat instructionnlp4Jun 4, 2026
Science Question AnsweringGPQAnlpbio medical4Jun 4, 2026
Mathematics & Science Question AnsweringGPQAnlpbio medical4Jun 4, 2026
Graduate-Level Question AnsweringGPQAnlp4Jun 4, 2026
Sparse retrieval-style question answeringRULER synthetic QA (standard)retrieval search ragnlp4Jun 4, 2026
Question AnsweringMMLUnlp4Jun 2, 2026