Research map / Large language models / AI and society
AI for health and public decisions
Apply AI frameworks to health monitoring, mental health, and equitable public decisions.
55 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in AI and society, keyword cluster “human · systems · argue”
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.
17 papers in 2023–24 and 27 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 7.5% to 5.3%.
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 · Topical Analysis of Interactions Between News and Social Media (AAAI 2016 · 20 citations)
- 2017 · TweetFit: Fusing Multiple Social Media and Sensor Data for Wellness Profile Learning (AAAI 2017 · 32 citations)
- 2019 · Learning to Address Health Inequality in the United States with a Bayesian Decision Network (AAAI 2019 · 8 citations)
- 2020 · Crowd-Assisted Disaster Scene Assessment with Human-AI Interactive Attention (AAAI 2020 · 35 citations)
- 2021 · Automated Lay Language Summarization of Biomedical Scientific Reviews (AAAI 2021 · 50 citations)
- 2022 · METS-CoV: A Dataset of Medical Entity and Targeted Sentiment on COVID-19 Related Tweets (NeurIPS 2022 · 10 citations)
- 2023 · Demo Alleviate: Demonstrating Artificial Intelligence Enabled Virtual Assistance for Telehealth: The Mental Health Case (AAAI 2023 · 11 citations)
- 2024 · Your Career Path Matters in Person-Job Fit (AAAI 2024 · 9 citations)
- 2025 · SNAP: Semantic Stories for Next Activity Prediction (AAAI 2025 · 6 citations)
- 2026 · SmellNet: A Large-scale Dataset for Real-world Smell Recognition (ICLR 2026 · 1 citation)
Earliest papers here
- Topical Analysis of Interactions Between News and Social Media (AAAI 2016 · 20 citations)
- An Algorithm to Coordinate Measurements Using Stochastic Human Mobility Patterns in Large-Scale Participatory Sensing Settings (AAAI 2016 · 10 citations)
- Predicting Online Protest Participation of Social Media Users (AAAI 2016 · 9 citations)
What has been tried
32 papers here report 228 measured changes (a component added, removed or swapped, compared with a baseline): helped 70%, no clear effect 8%, hurt 18%, 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 |
|---|---|---|---|---|
| Add career path identification (CPI) | Person-job matching test set · accuracy | 0.813 -> 0.827 | Helped | Your Career Path Matters in Person-Job Fit (AAAI 2024 · 9 citations) |
| use semantic stories instead of a list of feature values | BPI13cp, BPI13in, Env Permit, Sepsis, and Nasa · accuracy and weighted F1-score | BPI13cp: 0.616 -> 0.679 accuracy, 0.614 -> 0.670 F1; BPI13in: 0.733 -> 0.758 accuracy, 0.731 -> 0.754 F1; Env Permit: 0. | Helped | SNAP: Semantic Stories for Next Activity Prediction (AAAI 2025 · 6 citations) |
| polynomial instead of linear SVM kernel | Activity classification 70-30 split · weighted test F1 | 0.817 -> 0.879 | Helped | Indoor Air Quality Dataset with Activities of Daily Living in Low to Middle-income Communities (NeurIPS 2024 · 5 citations) |
| use the proposed quality-control indicator calculation method instead of ICL inference | QCIC test set · accuracy | 88.53 -> 93.31; improvement of 4.78% | Helped | IMQC: A Large Language Model Platform for Medical Quality Control (AAAI 2025 · 1 citation) |
| use standard feature names rather than raw laboratory test names | Critical Outcome · F1 | 0.397 -> 0.413 | Helped | ED-Copilot: Reduce Emergency Department Wait Time with Language Model Diagnostic Assistance (ICML 2024 · 1 citation) |
| Use no reward instead of both task-relevance and smoothness rewards | PDCARE test · UPC | 60.7% -> 57.5% | Hurt | Breaking Barriers: A Paradigm Shift in Technology Accessibility for Individuals with Physical Disabilities (AAAI 2025 · 0 citations) |
| Add global context optimization | Collected dataset · Acc | 83.2 -> 86.1 | Helped | XCotton: Advancing AI-Enabled Hardware/Software Integrated System for Foreign Fiber Cleaning (AAAI 2025 · 0 citations) |
| Add MIDR-specific fine-tuning to the extractive QA pipeline | MFQuAD All · Exact Match | 0.416 -> 0.671 | Helped | Symbol Description Reading (AAAI 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 21 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Baselines
- Llama2-7B · 2 of 21 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 44 papers; from OpenAlex and the papers' affiliations).
- Carnegie Mellon University · 6 papers
- IBM · 4 papers
- Howard University · 2 papers
- Harvard University · 2 papers
- Purdue University · 2 papers
- Ben-Gurion University of the Negev · 2 papers
- University of Washington · 2 papers
- Monash University · 2 papers
- Massachusetts Institute of Technology · 2 papers
- Johns Hopkins University · 2 papers
Use it
Download BibTeX of 34 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
- Human-centered trust and agency (161 papers)
- Social-science view of AI evaluation (159 papers)
- Data commons and human baselines (159 papers)
- Critiques of LLM claims and judging (104 papers)
- Benchmarking and reproducibility practice (104 papers)
Elsewhere in the topic
- Teacher needs and AI curricula (45 papers)
- Classroom tools and student perceptions (40 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).