TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

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

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

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

Models/Video-LLaVA-7B

Video-LLaVA-7B

Reported on 18 benchmarks across 4 tasks · 1 paper · 10 SOTA

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

Natural Language Processing12 results

  • Relation ExtractiononVinoground
    Group Score· 2023-11-16
    6.6
    best: 35 (GPT-4o (CoT))
    SOTA
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Relation ExtractiononVinoground
    Text Score· 2023-11-16
    24.8
    best: 59.2 (GPT-4o (CoT))
    SOTA
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Relation ExtractiononVinoground
    Video Score· 2023-11-16
    25.8
    best: 51 (GPT-4o (CoT))
    SOTA
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Question AnsweringonTGIF-QA
    Accuracy· 2023-11-16
    70
    best: 82.5 (Tarsier (34B))
    SOTA
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Question AnsweringonMSRVTT-QA
    Accuracy· uses extra data· 2023-11-16
    59.2
    best: 72.4 (Flash-VStream)
    SOTA
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Temporal Relation ExtractiononVinoground
    Group Score· 2023-11-16
    6.6
    best: 35 (GPT-4o (CoT))
    SOTA
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Temporal Relation ExtractiononVinoground
    Text Score· 2023-11-16
    24.8
    best: 59.2 (GPT-4o (CoT))
    SOTA
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Temporal Relation ExtractiononVinoground
    Video Score· 2023-11-16
    25.8
    best: 51 (GPT-4o (CoT))
    SOTA
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Question AnsweringonMSVD-QA
    Accuracy· 2023-11-16
    70.7
    best: 80.3 (Tarsier (34B))
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Question AnsweringonMSVD-QA
    Confidence Score· 2023-11-16
    3.9
    best: 2.5 (Video LLaMA-7B)
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Question AnsweringonTGIF-QA
    Confidence Score· 2023-11-16
    4
    best: 2.3 (Video Chat-7B)
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Question AnsweringonMSRVTT-QA
    Confidence Score· uses extra data· 2023-11-16
    3.5
    best: 1.8 (Video LLaMA-7B)
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122

Reasoning6 results

  • Video Question AnsweringonTGIF-QA
    Accuracy· 2023-11-16
    70
    best: 82.5 (Tarsier (34B))
    SOTA
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Video Question AnsweringonMSRVTT-QA
    Accuracy· uses extra data· 2023-11-16
    59.2
    best: 72.4 (Flash-VStream)
    SOTA
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Video Question AnsweringonMSVD-QA
    Accuracy· 2023-11-16
    70.7
    best: 80.3 (Tarsier (34B))
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Video Question AnsweringonMSVD-QA
    Confidence Score· 2023-11-16
    3.9
    best: 2.5 (Video LLaMA-7B)
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Video Question AnsweringonTGIF-QA
    Confidence Score· 2023-11-16
    4
    best: 2.3 (Video Chat-7B)
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122
  • Video Question AnsweringonMSRVTT-QA
    Confidence Score· uses extra data· 2023-11-16
    3.5
    best: 1.8 (Video LLaMA-7B)
    Video-LLaVA: Learning United Visual Representation by Alignment Before ProjectionarXiv:2311.10122