AWhat changed
Natural-language prompts could specify many visual categories without task-specific training in the tested benchmarks.
Scientific scope
2021 twelve-author CLIP paper; data biases, later systems and all visual understanding claims excluded.
Evidence
The linked sources identify this contribution and its scope. WRS has not independently reconstructed the scientific claims. The record is the twelve-author 2021 CLIP paper. Dataset construction, benchmark limitations, biases and later multimodal models remain distinct; zero-shot results do not imply universal visual understanding.
Contributors
- The named authors listed in the linked research record · SCIENCE: Complete source byline retained in the research record; this collective display does not merge person identities or measure labor shares.
- Radford et al., Learning Transferable Visual Models From Natural Language Supervision
- OpenAI: CLIP, connecting text and images
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction