# GQA

Source: [GQA](<https://www.sota2.com/research/dataset/gqa>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Benchmarks

Page 1 of 5. This is an index page, not the full collection.

- [Next index page](<https://www.sota2.com/research/dataset/gqa.md?page=2>): Follow this link to discover more pages.

- [Visual Question Answering on GQA](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa.md>): Visual Question Answering; GQA.

- [Visual Question Answering on GQA (Accuracy)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-accuracy.md>): Visual Question Answering; GQA.

- [Visual Question Answering on GQA (test-dev)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-test-dev.md>): Visual Question Answering; GQA (test-dev).

- [Visual Question Answering Accuracy on GQA](<https://www.sota2.com/research/sota/visual-question-answering-accuracy-on-gqa.md>): Visual Question Answering; GQA.

- [Visual Question Answering on GQA (test)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-test.md>): Visual Question Answering; GQA (test).

- [Visual Question Answering on GQA (Mean accuracy)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-mean-accuracy.md>): Visual Question Answering; GQA.

- [Visual Question Answering on GQA (Score)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-score.md>): Visual Question Answering; GQA.

- [Performance Estimation on GQA (MAE)](<https://www.sota2.com/research/sota/performance-estimation-on-gqa-mae.md>): Performance Estimation; GQA.

- [Visual Question Answering on GQA (GQA, Avg.)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-gqa-avg.md>): Visual Question Answering; GQA.

- [Visual Reasoning on GQA (Accuracy)](<https://www.sota2.com/research/sota/visual-reasoning-on-gqa-accuracy.md>): Visual Reasoning; GQA.

- [Error Prediction on AmbigQA (val)](<https://www.sota2.com/research/sota/error-prediction-on-ambigqa-val.md>): Error Prediction; AmbigQA (val).

- [Object Hallucination Evaluation on POPE GQA](<https://www.sota2.com/research/sota/object-hallucination-evaluation-on-pope-gqa.md>): Object Hallucination Evaluation; POPE GQA.

- [Visual Question Answering on GQA (GQA Score and Average)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-gqa-score-and-average.md>): Visual Question Answering; GQA.

- [Visual Question Answering on GQA (test-std)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-test-std.md>): Visual Question Answering; GQA (test-std).

- [Object Hallucination Evaluation on POPE GQA Popular](<https://www.sota2.com/research/sota/object-hallucination-evaluation-on-pope-gqa-popular.md>): Object Hallucination Evaluation; POPE GQA Popular.

- [Grounded Video Question Answering on NExT-GQA](<https://www.sota2.com/research/sota/grounded-video-question-answering-on-next-gqa.md>): Grounded Video Question Answering; NExT-GQA.

- [Compositional Zero-Shot Learning on C-GQA (open world)](<https://www.sota2.com/research/sota/compositional-zero-shot-learning-on-c-gqa-open-world.md>): Compositional Zero-Shot Learning; C-GQA open world.

- [Visual Question Answering on GQA (Accuracy and Average performance retention)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-accuracy-and-average-performance-retention.md>): Visual Question Answering; GQA.

- [Temporal Question Grounding on NExT-GQA](<https://www.sota2.com/research/sota/temporal-question-grounding-on-next-gqa.md>): Temporal Question Grounding; NExT-GQA.

- [Multimodal Understanding on MLLM Evaluation Suite (GQA, ScienceQA-IMG, TextVQA, POPE, MME, MMBench, MM-Vet, MMStar, AI2D, HallusionBench) (test val)](<https://www.sota2.com/research/sota/multimodal-understanding-on-mllm-evaluation-suite-gqa-scienceqa-img-textvqa-pope-mme-mmbench-mm-vet-mmstar-ai2d-hallusionbench-test-val.md>): Multimodal Understanding; MLLM Evaluation Suite GQA, ScienceQA-IMG, TextVQA, POPE, MME, MMBench, MM-Vet, MMStar, AI2D, HallusionBench (test val).

- [Compositional Zero-Shot Learning on C-GQA Closed World](<https://www.sota2.com/research/sota/compositional-zero-shot-learning-on-c-gqa-closed-world.md>): Compositional Zero-Shot Learning; C-GQA Closed World.

- [Object Hallucination Probing on GQA Adversarial](<https://www.sota2.com/research/sota/object-hallucination-probing-on-gqa-adversarial.md>): Object Hallucination Probing; GQA Adversarial.

- [Multi-modal Understanding on LLaVA Multi-modal Evaluation Suite v1.6 (test)](<https://www.sota2.com/research/sota/multi-modal-understanding-on-llava-multi-modal-evaluation-suite-v1-6-test.md>): Multi-modal Understanding; LLaVA Multi-modal Evaluation Suite (GQA, MMB, MME, POPE, SQA, VQAv2, TextVQA, MMMU, SEED-I) v1.6 (test).

- [Multi-modal Vision-Language Understanding on GQA](<https://www.sota2.com/research/sota/multi-modal-vision-language-understanding-on-gqa.md>): Multi-modal Vision-Language Understanding; GQA.

- [Object Hallucination Probing on GQA POPE Popular](<https://www.sota2.com/research/sota/object-hallucination-probing-on-gqa-pope-popular.md>): Object Hallucination Probing; GQA POPE Popular.

- [Visual Question Answering on LLaVA Evaluation Suite (GQA, MMB, MMMU, MME, VQAv2, VQAtext, POPE, SQA, SEED, MMVet)](<https://www.sota2.com/research/sota/visual-question-answering-on-llava-evaluation-suite-gqa-mmb-mmmu-mme-vqav2-vqatext-pope-sqa-seed-mmvet.md>): Visual Question Answering; LLaVA Evaluation Suite GQA, MMB, MMMU, MME, VQAv2, VQAtext, POPE, SQA, SEED, MMVet 1.5 7B (various).

- [Generalized Compositional Zero-shot Learning on C-GQA (test)](<https://www.sota2.com/research/sota/generalized-compositional-zero-shot-learning-on-c-gqa-test.md>): Generalized Compositional Zero-shot Learning; C-GQA (test).

- [Grounded Video Question Answering on NExT-GQA (test)](<https://www.sota2.com/research/sota/grounded-video-question-answering-on-next-gqa-test.md>): Grounded Video Question Answering; NExT-GQA (test).

- [Object Hallucination Evaluation on GQA (Random)](<https://www.sota2.com/research/sota/object-hallucination-evaluation-on-gqa-random.md>): Object Hallucination Evaluation; GQA (Random).

- [Object Hallucination Probing on GQA POPE Random](<https://www.sota2.com/research/sota/object-hallucination-probing-on-gqa-pope-random.md>): Object Hallucination Probing; GQA POPE Random.

- [Multimodal Understanding on LLaVA Evaluation Suite (GQA, MMB, MMB-CN, MME, POPE, SQA, VQAV2, VQAText, VizWiz)](<https://www.sota2.com/research/sota/multimodal-understanding-on-llava-evaluation-suite-gqa-mmb-mmb-cn-mme-pope-sqa-vqav2-vqatext-vizwiz.md>): Multimodal Understanding; LLaVA Evaluation Suite GQA, MMB, MMB-CN, MME, POPE, SQA, VQAV2, VQAText, VizWiz.

- [Vision-Language Understanding and Reasoning on LLaVA Multimodal Evaluation Suite 1.5 (test/val)](<https://www.sota2.com/research/sota/vision-language-understanding-and-reasoning-on-llava-multimodal-evaluation-suite-1-5-test-val.md>): Vision-Language Understanding and Reasoning; LLaVA Multimodal Evaluation Suite (GQA, MMBench, MME, POPE, ScienceQA, VQAv2, TextVQA, SEED-Bench, MM-Vet, VizWiz) 1.5 (test/val).

- [Question Answering on AmbigQA (One Intent Accuracy, Any Intent Accuracy)](<https://www.sota2.com/research/sota/question-answering-on-ambigqa-one-intent-accuracy-any-intent-accuracy.md>): Question Answering; AmbigQA.

- [Visual Question Answering on GQA (Accuracy, Cost)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-accuracy-cost.md>): Visual Question Answering; GQA.

- [General Visual Question Answering on GQA](<https://www.sota2.com/research/sota/general-visual-question-answering-on-gqa.md>): General Visual Question Answering; GQA.

- [Object Hallucination Evaluation on POPE GQA (Adversarial)](<https://www.sota2.com/research/sota/object-hallucination-evaluation-on-pope-gqa-adversarial.md>): Object Hallucination Evaluation; POPE GQA (Adversarial).

- [Multi-turn Visual Question Answering on MT-GQA (Accuracy Metrics)](<https://www.sota2.com/research/sota/multi-turn-visual-question-answering-on-mt-gqa-accuracy-metrics.md>): Multi-turn Visual Question Answering; MT-GQA.

- [Multimodal Understanding on Standard Multimodal Evaluation Suite 1.5 (test val)](<https://www.sota2.com/research/sota/multimodal-understanding-on-standard-multimodal-evaluation-suite-1-5-test-val.md>): Multimodal Understanding; Standard Multimodal Evaluation Suite (GQA, MMBench, MME, TextVQA, ScienceQA, VQA v2) 1.5 (test val).

- [Visual Question Answering on GQA (Exact Match)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-exact-match.md>): Visual Question Answering; GQA.

- [Visual Question Answering on GQA balanced (test-dev)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-balanced-test-dev.md>): Visual Question Answering; GQA balanced (test-dev).

- [Visual Question Answering on GQA (val)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-val.md>): Visual Question Answering; GQA (val).

- [Accuracy on GQA (Visual Question Answering)](<https://www.sota2.com/research/sota/accuracy-on-gqa-visual-question-answering.md>): Visual Question Answering; GQA.

- [Object Hallucination Evaluation on GQA Popular split](<https://www.sota2.com/research/sota/object-hallucination-evaluation-on-gqa-popular-split.md>): Object Hallucination Evaluation; GQA Popular split.

- [Visual Question Answering on GQA v1.0 (test)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-v1-0-test.md>): Visual Question Answering; GQA v1.0 (test).

- [Scene Graph Generation on GQA-200 (test)](<https://www.sota2.com/research/sota/scene-graph-generation-on-gqa-200-test.md>): Scene Graph Generation; GQA-200 (test).

- [Refusal Rate Evaluation on GQA](<https://www.sota2.com/research/sota/refusal-rate-evaluation-on-gqa.md>): Refusal Rate Evaluation; GQA.

- [Visual Question Answering on GQA (Accuracy, Relative Performance)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-accuracy-relative-performance.md>): Visual Question Answering; GQA.

- [Video Question Answering on AGQA v2 (test)](<https://www.sota2.com/research/sota/video-question-answering-on-agqa-v2-test.md>): Video Question Answering; AGQA v2 (test).

- [Ambiguity Detection on AmbigQA](<https://www.sota2.com/research/sota/ambiguity-detection-on-ambigqa.md>): Ambiguity Detection; AmbigQA.

- [Object Hallucination Evaluation on POPE GQA (test)](<https://www.sota2.com/research/sota/object-hallucination-evaluation-on-pope-gqa-test.md>): Object Hallucination Evaluation; POPE GQA (test).

- [Retrieval-Augmented Question Answering on RAGQA Leaderboard (test)](<https://www.sota2.com/research/sota/retrieval-augmented-question-answering-on-ragqa-leaderboard-test.md>): Retrieval-Augmented Question Answering; RAGQA Leaderboard (test).

- [Visual Question Answering on GQA (Accuracy and Relative Performance)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-accuracy-and-relative-performance.md>): Visual Question Answering; GQA.

- [Video Question Answering on NExT-GQA (test)](<https://www.sota2.com/research/sota/video-question-answering-on-next-gqa-test.md>): Video Question Answering; NExT-GQA (test).

- [Multimodal Visual Question Answering and Reasoning on LLaVA Evaluation Suite 1.5](<https://www.sota2.com/research/sota/multimodal-visual-question-answering-and-reasoning-on-llava-evaluation-suite-1-5.md>): Multimodal Visual Question Answering and Reasoning; LLaVA Evaluation Suite (GQA, SQA-IMG, VQA-Text, POPE, MME, MMBench-EN, MMBench-CN, MM-Vet) 1.5.

- [Multimodal Understanding and Question Answering on LLaVA-NeXT Evaluation Suite 13B (test)](<https://www.sota2.com/research/sota/multimodal-understanding-and-question-answering-on-llava-next-evaluation-suite-13b-test.md>): Multimodal Understanding and Question Answering; LLaVA-NeXT Evaluation Suite (GQA, SQA-IMG, VQA-Text, POPE, MME, MMB-EN, MMB-CN, MMVet) 13B (test).

- [Multimodal Question Answering and Reasoning on LMM Evaluation Suite (GQA, SQA-IMG, VQA-Text, POPE, MME, MMBench, MM-Vet) (test val)](<https://www.sota2.com/research/sota/multimodal-question-answering-and-reasoning-on-lmm-evaluation-suite-gqa-sqa-img-vqa-text-pope-mme-mmbench-mm-vet-test-val.md>): Multimodal Question Answering and Reasoning; LMM Evaluation Suite (GQA, SQA-IMG, VQA-Text, POPE, MME, MMBench, MM-Vet) (test val).

- [Multimodal Evaluation on Combined Benchmark Suite (GQA, MMB, MME, VQA-T, SQA-I, VQA-v2)](<https://www.sota2.com/research/sota/multimodal-evaluation-on-combined-benchmark-suite-gqa-mmb-mme-vqa-t-sqa-i-vqa-v2.md>): Multimodal Evaluation; Combined Benchmark Suite (GQA, MMB, MME, VQA-T, SQA-I, VQA-v2).

- [Visual Question Answering on GQA v1.2 (test)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-v1-2-test.md>): Visual Question Answering; GQA v1.2 (test).

- [Multi-turn Visual Question Answering on MT-GQA balanced (test-dev)](<https://www.sota2.com/research/sota/multi-turn-visual-question-answering-on-mt-gqa-balanced-test-dev.md>): Multi-turn Visual Question Answering; MT-GQA balanced (test-dev).

- [Visual Question Answering on GQA (ECE)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-ece.md>): Visual Question Answering; GQA.

- [Visual Question Answering on GQA (Clean & Robust Accuracy)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-clean-and-robust-accuracy.md>): Visual Question Answering; GQA.

- [Visual Question Answering on GQA (GQA and RelAcc.)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-gqa-and-relacc.md>): Visual Question Answering; GQA.

- [Visual Understanding on ScienceQA, TextVQA, and GQA](<https://www.sota2.com/research/sota/visual-understanding-on-scienceqa-textvqa-and-gqa.md>): Visual Understanding; ScienceQA, TextVQA, and GQA.

- [Multimodal Question Answering and Understanding on Standard Multimodal Evaluation Suite (GQA, MMB, MME, VQA-T, SQA-I, VQA-v2, POPE, MMMU, MM-Vet)](<https://www.sota2.com/research/sota/multimodal-question-answering-and-understanding-on-standard-multimodal-evaluation-suite-gqa-mmb-mme-vqa-t-sqa-i-vqa-v2-pope-mmmu-mm-vet.md>): Multimodal Question Answering and Understanding; Standard Multimodal Evaluation Suite GQA, MMB, MME, VQA-T, SQA-I, VQA-v2, POPE, MMMU, MM-Vet.

- [Object Hallucination Evaluation on POPE averaged across MS-COCO, A-OKVQA, and GQA (test)](<https://www.sota2.com/research/sota/object-hallucination-evaluation-on-pope-averaged-across-ms-coco-a-okvqa-and-gqa-test.md>): Object Hallucination Evaluation; POPE averaged across MS-COCO, A-OKVQA, and GQA (test).

- [Visual Question Answering on GQA (Accuracy, Avg.)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-accuracy-avg.md>): Visual Question Answering; GQA.

- [Multimodal Image Understanding on LLaVA-1.5-7B Evaluation Suite](<https://www.sota2.com/research/sota/multimodal-image-understanding-on-llava-1-5-7b-evaluation-suite.md>): Multimodal Image Understanding; LLaVA-1.5-7B Evaluation Suite (GQA, MMBench, MMBench-CN, MME, POPE, ScienceQA, TextVQA, VizWiz, MM-Vet).

- [Visual Question Answering on GQA (Accuracy and Average accuracy proportion w.r.t. vanilla)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-accuracy-and-average-accuracy-proportion-w-r-t-vanilla.md>): Visual Question Answering; GQA.

- [Visual Question Answering on GQA 40 (test)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-40-test.md>): Visual Question Answering; GQA 40 (test).

- [Visual Question Answering on GQA (test val)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-test-val.md>): Visual Question Answering; GQA (test val).

- [Visual Question Answering on POPE GQA](<https://www.sota2.com/research/sota/visual-question-answering-on-pope-gqa.md>): Visual Question Answering; POPE GQA.

- [Object Hallucination Evaluation on POPE GQA (random)](<https://www.sota2.com/research/sota/object-hallucination-evaluation-on-pope-gqa-random.md>): Object Hallucination Evaluation; POPE GQA (random).

- [Visual Reasoning on GQA (test)](<https://www.sota2.com/research/sota/visual-reasoning-on-gqa-test.md>): Visual Reasoning; GQA (test).

- [Visual Question Answering on GQA (accuracy, relative accuracy)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-accuracy-relative-accuracy.md>): Visual Question Answering; GQA.

- [Visual Question Answering on VQAv2, VizWiz, GQA, TextVQA, and MME](<https://www.sota2.com/research/sota/visual-question-answering-on-vqav2-vizwiz-gqa-textvqa-and-mme.md>): Visual Question Answering; General VQA VQAv2, VizWiz, GQA, TextVQA, MME.

- [Grounded Question Answering on NExT-GQA (test)](<https://www.sota2.com/research/sota/grounded-question-answering-on-next-gqa-test.md>): Grounded Question Answering; NExT-GQA (test).

- [Multimodal Understanding and Question Answering on LLaVA Evaluation Suite 1.5 (GQA, MMBench, MMBench-CN, MME, POPE, ScienceQA, VQAv2, TextVQA, SEED-Bench, VizWiz)](<https://www.sota2.com/research/sota/multimodal-understanding-and-question-answering-on-llava-evaluation-suite-1-5-gqa-mmbench-mmbench-cn-mme-pope-scienceqa-vqav2-textvqa-seed-bench-vizwiz.md>): Multimodal Understanding and Question Answering; LLaVA 7B Evaluation Suite (GQA, MMBench, MMBench-CN, MME, POPE, ScienceQA, VQAv2, TextVQA, SEED-Bench, VizWiz) 1.5.

- [Visual Question Answering on GQA (GQA, Rel.)](<https://www.sota2.com/research/sota/visual-question-answering-on-gqa-gqa-rel.md>): Visual Question Answering; GQA.

- [Vision-Language Question Answering on Pooled MLLM Suite (GQA, LLaVA-Wild, MMMU Pro, MME-Finance) (test)](<https://www.sota2.com/research/sota/vision-language-question-answering-on-pooled-mllm-suite-gqa-llava-wild-mmmu-pro-mme-finance-test.md>): Vision-Language Question Answering; Pooled Shared (GQA, POPE, LLaVA-Wild, MMMU Pro, GMAI-MMBench, MME-Finance) (test).

- [Visual Grounding on GQA](<https://www.sota2.com/research/sota/visual-grounding-on-gqa.md>): Visual Grounding; GQA.
