Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/SlowFast-LLaVA-34B

SlowFast-LLaVA-34B

Reported on 29 benchmarks across 6 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Reasoning23 results

Video Question AnsweringonNExT-QA
Accuracy· 2024-07-22
64.2
best: 85.5 (LinVT-Qwen2-VL (7B))
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video Question AnsweringonMSVD-QA
Accuracy· 2024-07-22
79.9
best: 80.3 (Tarsier (34B))
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video Question AnsweringonMSVD-QA
Confidence Score· 2024-07-22
4.1
best: 2.5 (Video LLaMA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video Question AnsweringonTGIF-QA
Accuracy· 2024-07-22
80.6
best: 82.5 (Tarsier (34B))
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video Question AnsweringonTGIF-QA
Confidence Score· 2024-07-22
4.3
best: 2.3 (Video Chat-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video Question AnsweringonMSRVTT-QA
Accuracy· 2024-07-22
67.4
best: 72.4 (Flash-VStream)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video Question AnsweringonMSRVTT-QA
Confidence Score· 2024-07-22
3.7
best: 1.8 (Video LLaMA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video Question AnsweringonIntentQA
Accuracy· 2024-07-22
60.1
best: 71.5 (ENTER)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video Question AnsweringonEgoSchema (subset)
Accuracy· 2024-07-22
47.2
best: 68.6 (Tarsier (34B))
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video Question AnsweringonActivityNet-QA
Accuracy· 2024-07-22
59.2
best: 61.6 (Tarsier (34B))
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video Question AnsweringonActivityNet-QA
Confidence Score· 2024-07-22
3.5
best: 1.1 (Video LLaMA)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Generative Visual Question AnsweringonVideoInstruct
mean· 2024-07-22
3.32
best: 3.73 (PPLLaVA-7B-dpo)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Generative Visual Question AnsweringonVideoInstruct
gpt-score· 2024-07-22
3.84
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Generative Visual Question AnsweringonVideoInstruct
gpt-score· 2024-07-22
3.48
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Generative Visual Question AnsweringonVideoInstruct
gpt-score· 2024-07-22
2.96
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Generative Visual Question AnsweringonVideoInstruct
gpt-score· 2024-07-22
2.77
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Generative Visual Question AnsweringonVideoInstruct
gpt-score· 2024-07-22
3.57
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video-based Generative Performance BenchmarkingonVideoInstruct
mean· 2024-07-22
3.32
best: 3.73 (PPLLaVA-7B-dpo)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video-based Generative Performance BenchmarkingonVideoInstruct
gpt-score· 2024-07-22
3.84
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video-based Generative Performance BenchmarkingonVideoInstruct
gpt-score· 2024-07-22
3.48
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video-based Generative Performance BenchmarkingonVideoInstruct
gpt-score· 2024-07-22
2.96
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video-based Generative Performance BenchmarkingonVideoInstruct
gpt-score· 2024-07-22
2.77
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Video-based Generative Performance BenchmarkingonVideoInstruct
gpt-score· 2024-07-22
3.57
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841

Natural Language Processing17 results

Question AnsweringonNExT-QA
Accuracy· 2024-07-22
64.2
best: 79.6 (VideoMultiAgent (GPT-4o))
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Question AnsweringonMSVD-QA
Accuracy· 2024-07-22
79.9
best: 80.3 (Tarsier (34B))
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Question AnsweringonMSVD-QA
Confidence Score· 2024-07-22
4.1
best: 2.5 (Video LLaMA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Question AnsweringonTGIF-QA
Accuracy· 2024-07-22
80.6
best: 82.5 (Tarsier (34B))
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Question AnsweringonTGIF-QA
Confidence Score· 2024-07-22
4.3
best: 2.3 (Video Chat-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Question AnsweringonMSRVTT-QA
Accuracy· 2024-07-22
67.4
best: 72.4 (Flash-VStream)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Question AnsweringonMSRVTT-QA
Confidence Score· 2024-07-22
3.7
best: 1.8 (Video LLaMA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Question AnsweringonIntentQA
Accuracy· 2024-07-22
60.1
best: 71.5 (ENTER)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Question AnsweringonEgoSchema (subset)
Accuracy· 2024-07-22
47.2
best: 68.6 (Tarsier (34B))
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Question AnsweringonActivityNet-QA
Accuracy· 2024-07-22
59.2
best: 61.6 (Tarsier (34B))
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Question AnsweringonActivityNet-QA
Confidence Score· 2024-07-22
3.5
best: 1.1 (Video LLaMA)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Visual Question Answering (VQA)onVideoInstruct
mean· 2024-07-22
3.32
best: 3.73 (PPLLaVA-7B-dpo)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Visual Question Answering (VQA)onVideoInstruct
gpt-score· 2024-07-22
3.84
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Visual Question Answering (VQA)onVideoInstruct
gpt-score· 2024-07-22
3.48
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Visual Question Answering (VQA)onVideoInstruct
gpt-score· 2024-07-22
2.96
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Visual Question Answering (VQA)onVideoInstruct
gpt-score· 2024-07-22
2.77
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841
Visual Question Answering (VQA)onVideoInstruct
gpt-score· 2024-07-22
3.57
best: 4.21 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841

Other1 result

Video-based Generative Performance Benchmarking (Correctness of Information)onVideoInstruct
gpt-score· 2024-07-22
3.48
best: 3.85 (PPLLaVA-7B)
SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models arXiv:2407.15841