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/Mirasol3B

Mirasol3B

Reported on 8 benchmarks across 4 tasks · 1 paper · 5 SOTA

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

Reasoning3 results

Video Question AnsweringonActivityNet-QA
Accuracy· 2023-11-09
51.13
best: 61.6 (Tarsier (34B))
Mirasol3B: A Multimodal Autoregressive model for time-aligned and contextual modalities arXiv:2311.05698
Video Question AnsweringonMSRVTT-QA
Accuracy· 2023-11-09
50.42
best: 72.4 (Flash-VStream)
Mirasol3B: A Multimodal Autoregressive model for time-aligned and contextual modalities arXiv:2311.05698
Video Question AnsweringonNExT-QA
Accuracy· 2023-11-09
72
best: 85.5 (LinVT-Qwen2-VL (7B))
Mirasol3B: A Multimodal Autoregressive model for time-aligned and contextual modalities arXiv:2311.05698

Audio2 results

Audio ClassificationonEPIC-SOUNDS
Accuracy· 2023-11-09
78.2
SOTA
Mirasol3B: A Multimodal Autoregressive model for time-aligned and contextual modalities arXiv:2311.05698
Audio ClassificationonVGGSound
Top 1 Accuracy· 2023-11-09
69.8
SOTA
Mirasol3B: A Multimodal Autoregressive model for time-aligned and contextual modalities arXiv:2311.05698

Methodology2 results

ClassificationonEPIC-SOUNDS
Accuracy· 2023-11-09
78.2
SOTA
Mirasol3B: A Multimodal Autoregressive model for time-aligned and contextual modalities arXiv:2311.05698
ClassificationonVGGSound
Top 1 Accuracy· 2023-11-09
69.8
SOTA
Mirasol3B: A Multimodal Autoregressive model for time-aligned and contextual modalities arXiv:2311.05698

Computer Vision1 result

VideoonKinetics-Sounds
Top 1 Accuracy· 2023-11-09
90.1
best: 93.3 (CA2ST(B/16))
SOTA
Mirasol3B: A Multimodal Autoregressive model for time-aligned and contextual modalities arXiv:2311.05698