New AACL paper: using multimodal models to analyze social constructs in short-form videos at scale
krisgligoric · x · 2026-10-02
A new paper appearing at AACL main targets researchers who want to use multimodal models to analyze social constructs like emotion, topics, or framing in short-form videos, addressing three questions: how well it works, whether conclusions will be valid, and how costs scale.
It offers a methodological framework and cost analysis for using model annotations in place of human labeling—a practical reference for computational social science meets LLMs.
More from Research
- Marin's 535B MoE hero run hits ~27% MFU on 11 NVL72 racks: expert parallelism deep dive — dlwh · 2026-10-03
- SWE-sweep benchmark tests whether AI agents can find bugs before users hit them — klieret · 2026-10-02
- Harvard-led team unveils brain imaging 60x faster than fMRI, tracking activity in ~100ms — melnykowycz · 2026-10-02
- RExBench: best coding agent implements research extensions only 33% of the time — najoungkim · 2026-10-02
- Retrieval-augmented episodic memory narrows LLMs' lexical frequency gap in syntax tests — najoungkim · 2026-10-02
- EMPIRIC teaches robots missing physics as code, solving all 25 tasks where baselines get 14-16 — tomssilver · 2026-10-02