Research map / Vision-language and multimodal / Robot action and navigation
Hulk: A Universal Knowledge Translator for Human-Centric Tasks
Yizhou Wang, Yixuan Wu, Weizhen He, Xun Guo, Feng Zhu, Lei Bai, Rui Zhao, Jian Wu, Tong He, Wanli Ouyang, Shixiang Tang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| share modality tokenizers instead of dataset-specific ones (ablation) | Hulk · 8-task ablation average | average score: 41.1 -> 41.8 | Improved |
| share decoder parameters in addition to encoder (ablation) | Hulk · 8-task ablation average | average score: 29.0 -> 41.1 | Improved |
| share decoder parameters in addition to encoder (ablation) | Hulk · NTU60-XSub | Acc: 1.7 -> 93.8 | Improved |
| add 2D pose task to joint training (ablation) | Hulk · CrowdHuman | AP: 52.6 -> 71.6 | Improved |
| add 3D pose task to joint training (ablation) | Hulk · CrowdHuman | AP: 52.6 -> 57.8 | Improved |
| add 2D pose task to joint training (ablation) | Hulk · RAPv2 | mA: 75.4 -> 78.7 | Improved |
| add 2D pose task to joint training (ablation) | Hulk · NTU60-XSub | Acc: 93.1 -> 93.5 | No clear effect |
| add 2D pose task to training (ablation) | Hulk · CrowdHuman | AP: 65.0 -> 81.8 | Improved |
| add 2D pose task to training (ablation) | Hulk · 3DPW mesh | 100-MPVPE: -18.9 -> -0.3 | Improved |
| add parsing task to training (ablation) | Hulk · LIP parsing | mIoU: 56.8 -> 57.4 | Hurt |
| add detection task to training (ablation) | Hulk · AIC 2D pose | AP: 26.4 -> 25.5 | Hurt |
| add mesh and 3D pose tasks to training (ablation) | Hulk · 8-task average | average score: -0.1 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- BLIP — BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation (ICML 2022) · better on 0 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Human-MME: A Holistic Evaluation Benchmark for Human-Centric Multimodal Large Language Models (ICLR 2026 · similarity 0.88)
- L-Man: A Large Multi-modal Model Unifying Human-centric Tasks (AAAI 2025 · similarity 0.87)
- HumanPCR: Probing MLLM Capabilities in Diverse Human-Centric Scenes (ICLR 2026 · similarity 0.86)
- A Large-Scale Human-Centric Benchmark for Referring Expression Comprehension in the LMM Era (NeurIPS 2024 · similarity 0.86)
- An Embodied Generalist Agent in 3D World (ICML 2024 · similarity 0.86)
- ChatHuman: Chatting about 3D Humans with Tools (CVPR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
