AIWalker
Mar 5, 2026 · Artificial Intelligence
How ViDA-UGC Leverages Large Multimodal Models for Fine-Grained Visual Quality Assessment
The article introduces ViDA-UGC, a large‑scale UGC visual‑quality dataset and its companion benchmark ViDA‑Bench, explains the MILP‑driven sampling, expert annotation pipeline, and CoT‑based evaluation framework, and shows how fine‑tuning popular multimodal LLMs on this data markedly improves low‑level quality perception, grounding, and description capabilities.
benchmarkchain of thoughtdataset
0 likes · 12 min read