Research map / Vision-language and multimodal / Human and embodied modeling
Human perception modeling
Models human visual motion, saliency, and eye movement with trainable networks and human signals.
71 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Human and embodied modeling, keyword cluster “computer vision · humans · abstract”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
29 papers in 2023–24 and 16 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 5.5% to 2.4%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Some Like It Hot - Visual Guidance for Preference Prediction (CVPR 2016 · 74 citations)
- 2017 · Detecting Visual Relationships With Deep Relational Networks (CVPR 2017 · 507 citations)
- 2018 · Guide Me: Interacting With Deep Networks (CVPR 2018 · 26 citations)
- 2020 · Hide-and-Tell: Learning to Bridge Photo Streams for Visual Storytelling (AAAI 2020 · 28 citations)
- 2021 · The Affective Growth of Computer Vision (CVPR 2021 · 26 citations)
- 2022 · V2C: Visual Voice Cloning (CVPR 2022 · 30 citations)
- 2023 · MetaCLUE: Towards Comprehensive Visual Metaphors Research (CVPR 2023 · 25 citations)
- 2024 · Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions (CVPR 2024 · 49 citations)
- 2025 · Image Quality Assessment: From Human to Machine Preference (CVPR 2025 · 11 citations)
- 2026 · Image Quality Assessment for Embodied AI (ICLR 2026 · 0 citations)
Earliest papers here
- Some Like It Hot - Visual Guidance for Preference Prediction (CVPR 2016 · 74 citations)
- We Are Humor Beings: Understanding and Predicting Visual Humor (CVPR 2016 · 49 citations)
- Predicting Motivations of Actions by Leveraging Text (CVPR 2016 · 40 citations)
What has been tried
63 papers here report 567 measured changes (a component added, removed or swapped, compared with a baseline): helped 66%, no clear effect 10%, hurt 19%, mixed 4%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Remove the editing transformer | COCAS+ Real2 CTCC-ReID · mAP | 32.2 -> 32.6 | No clear effect | Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions (CVPR 2024 · 49 citations) |
| add FTS | Gait3D · rank-1 accuracy | 73.2 -> 75.4 | Helped | Learning Visual Prompt for Gait Recognition (CVPR 2024 · 29 citations) |
| add observer encoder | OSIE-ASD · SM | 0.341 -> 0.377 | Helped | Beyond Average: Individualized Visual Scanpath Prediction (CVPR 2024 · 17 citations) |
| Provide human-generated oracle captions rather than model-generated captions | Visual Riddles open-ended VQA · human rating accuracy | 23 -> 50 | Helped | Visual Riddles: a Commonsense and World Knowledge Challenge for Large Vision and Language Models (NeurIPS 2024 · 12 citations) |
| Fine-tune DINOv2 on DepthCues | NYUv2 · accuracy | 87.78 -> 87.06 | Hurt | DepthCues: Evaluating Monocular Depth Perception in Large Vision Models (CVPR 2025 · 8 citations) |
| Use expanded textual description oti rather than only semantic character osi for conditioning | Oracle visual-guide test set · TransOV accuracy | 54.05% -> 72.97% | Helped | Making Visual Sense of Oracle Bones for You and Me (CVPR 2024 · 6 citations) |
| Add the unit representing mashed avocado dip or spread | ImageNet guacamole category · test accuracy | 84% | Helped | Automated Natural Language Explanation of Deep Visual Neurons with Large Models (Student Abstract) (AAAI 2024 · 3 citations) |
| Use deeper MoLFormer layers for odorant representations | Snitz and Ravia · representational alignment | Helped | Can Transformers Smell Like Humans? (NeurIPS 2024 · 2 citations) | |
| Tune on NIGHTS instead of BAPPS, THINGS, or ImageNet triplets | Counting and DeepFashion2 instance retrieval · counting RMSE and retrieval recall | Helped | When does perceptual alignment benefit vision representations? (NeurIPS 2024 · 2 citations) | |
| Add vision salience attraction point | EyEar-20k test set · PDS | 0.602 -> 0.614 | Helped | EyEar: Learning Audio Synchronized Human Gaze Trajectory Based on Physics-Informed Dynamics (AAAI 2025 · 1 citation) |
| Add CoRs to the extracted features | Semantic Complexity Scoring · PCC | 0.818 (0.003) -> 0.823 (0.002) | Helped | Is Your Image a Good Storyteller? (AAAI 2025 · 1 citation) |
| Train Stage 3 with images and textual annotations rather than images only | BlendFace · AUC (%) | 98.45 -> 99.48 | Helped | Guard Me If You Know Me: Protecting Specific Face-Identity from Deepfakes (NeurIPS 2025 · 1 citation) |
What reviewers will expect
Datasets and baselines named in the 12 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- COCO · 2 of 12 papers
Baselines
- CLIP · 2 of 12 papers
- GPT-4o · 2 of 12 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 45 papers; from OpenAlex and the papers' affiliations).
- Google · 4 papers
- Zhejiang University · 3 papers
- Massachusetts Institute of Technology · 3 papers
- University of Georgia · 2 papers
- Warsaw University of Technology · 2 papers
- Shanghai Jiao Tong University · 2 papers
- Beijing Institute of Technology · 2 papers
- Beijing Normal University · 2 papers
- Peking University · 2 papers
Use it
Download BibTeX of 37 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Multimodal and affective interaction modeling (87 papers)
- Human-like visual cognition in VLMs (97 papers)
- Abstract visual reasoning benchmarks (63 papers)
- Scientific domain vision datasets (47 papers)
Elsewhere in the topic
- Vision-language pretraining (182 papers)
- Medical vision-language models (130 papers)
- Scene graph generation (126 papers)
- Language-grounded scene text and structure (124 papers)
- fMRI visual decoding (106 papers)
- Multimodal retrieval-augmented modeling (85 papers)
- Language-grounded visual concept learning (83 papers)
- Composed and text-based image retrieval (78 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).