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/LLaMA Adapter

LLaMA Adapter

Reported on 30 benchmarks across 7 tasks · 2 papers · 29 SOTA

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

Reasoning24 results

Video Question AnsweringonActivityNet-QA
Confidence Score· 2023-04-28
2.7
best: 1.1 (Video LLaMA)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
Consistency· 2023-04-28
2.15
best: 3.81 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
Contextual Understanding· 2023-04-28
2.3
best: 4.21 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
Correctness of Information· 2023-04-28
2.03
best: 3.85 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
Detail Orientation· 2023-04-28
2.32
best: 3.56 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
Temporal Understanding· 2023-04-28
1.98
best: 3.23 (VLM-RLAIF)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
mean· 2023-04-28
2.16
best: 3.73 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
gpt-score· 2023-04-28
2.32
best: 4.21 (PPLLaVA-7B)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
Consistency· 2023-04-28
2.15
best: 3.81 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
Contextual Understanding· 2023-04-28
2.3
best: 4.21 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
Correctness of Information· 2023-04-28
2.03
best: 3.85 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
Detail Orientation· 2023-04-28
2.32
best: 3.56 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
Temporal Understanding· 2023-04-28
1.98
best: 3.23 (VLM-RLAIF)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
mean· 2023-04-28
2.16
best: 3.73 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
gpt-score· 2023-04-28
2.32
best: 4.21 (PPLLaVA-7B)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video Question AnsweringonActivityNet-QA
Accuracy· 2023-04-28
34.2
best: 61.6 (Tarsier (34B))
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
gpt-score· 2023-04-28
2.3
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
gpt-score· 2023-04-28
2.03
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
gpt-score· 2023-04-28
1.98
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Generative Visual Question AnsweringonVideoInstruct
gpt-score· 2023-04-28
2.15
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
gpt-score· 2023-04-28
2.3
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
gpt-score· 2023-04-28
2.03
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
gpt-score· 2023-04-28
1.98
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Video-based Generative Performance BenchmarkingonVideoInstruct
gpt-score· 2023-04-28
2.15
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010

Natural Language Processing13 results

Question AnsweringonActivityNet-QA
Accuracy· 2023-04-28
34.2
best: 61.6 (Tarsier (34B))
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Question AnsweringonActivityNet-QA
Confidence Score· 2023-04-28
2.7
best: 1.1 (Video LLaMA)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
Consistency· 2023-04-28
2.15
best: 3.81 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
Contextual Understanding· 2023-04-28
2.3
best: 4.21 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
Correctness of Information· 2023-04-28
2.03
best: 3.85 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
Detail Orientation· 2023-04-28
2.32
best: 3.56 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
Temporal Understanding· 2023-04-28
1.98
best: 3.23 (VLM-RLAIF)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
mean· 2023-04-28
2.16
best: 3.73 (PPLLaVA-7B-dpo)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
gpt-score· 2023-04-28
2.32
best: 4.21 (PPLLaVA-7B)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
gpt-score· 2023-04-28
2.3
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
gpt-score· 2023-04-28
2.03
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
gpt-score· 2023-04-28
1.98
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010
Visual Question Answering (VQA)onVideoInstruct
gpt-score· 2023-04-28
2.15
best: 4.21 (PPLLaVA-7B)
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010

Music4 results

Music Question AnsweringonMusicQA
BERT Score· 2023-03-28
0.895
best: 0.901 (MU-LLaMA)
SOTA
LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention arXiv:2303.16199
Music Question AnsweringonMusicQA
BLEU· 2023-03-28
0.273
best: 0.306 (MU-LLaMA)
SOTA
LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention arXiv:2303.16199
Music Question AnsweringonMusicQA
METEOR· 2023-03-28
0.334
best: 0.385 (MU-LLaMA)
SOTA
LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention arXiv:2303.16199
Music Question AnsweringonMusicQA
ROUGE· 2023-03-28
0.413
best: 0.466 (MU-LLaMA)
SOTA
LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention arXiv:2303.16199

Other1 result

Video-based Generative Performance Benchmarking (Correctness of Information)onVideoInstruct
gpt-score· 2023-04-28
2.03
best: 3.85 (PPLLaVA-7B)
SOTA
LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model arXiv:2304.15010