Research map / Vision-language and multimodal / Human and embodied modeling
VidLanKD: Improving Language Understanding via Video-Distilled Knowledge Transfer
Zineng Tang, Jaemin Cho, Hao Tan, Mohit Bansal
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Distill with video-grounded NST+CRD instead of text-only pretraining | BERT-base · 7 downstream NLU tasks (Avg.) | average score: 78.9 -> 82.6 | Improved |
| Use NST+CRD instead of image-based vokenization | BERT-base · 7 downstream NLU tasks (Avg.) | average score: 81.4 -> 82.6 | Improved |
| Use a CLIP frame encoder instead of ResNet (ablation) | VidLanKD · 7 downstream NLU tasks (Avg.) | average score: 82.4 -> 82.6 | Improved |
| Use an MLM-only teacher instead of the text-only baseline (ablation) | BERT-small with NST distillation · SST-2 | accuracy: 88.8 -> 88.1 | Hurt |
| Use a contrastive-only teacher instead of the text-only baseline (ablation) | BERT-small with NST distillation · QNLI | accuracy: 84.9 -> 85.2 | Improved |
| Train the teacher with both MLM and contrastive objectives (ablation) | BERT-small with NST distillation · SST-2 | accuracy: 88.8 -> 91.1 | Improved |
| Use video-then-text two-stage pretraining instead of VidLanKD (ablation) | BERT-small · SST-2 | accuracy: 91.1 -> 90.3 | Hurt |
| Use soft-label distillation instead of NST+CRD (ablation) | BERT-small · SST-2 | accuracy: 91.5 -> 87.2 | Hurt |
| Combine NST and CRD rather than using NST alone (ablation) | BERT-small · SST-2 | accuracy: 91.1 -> 91.5 | Improved |
| Use video-level NST rather than video vokenization (ablation) | BERT-small · SST-2 | accuracy: 89.7 -> 91.1 | Improved |
| Use video-level rather than image-level NST supervision (ablation) | BERT-small · SST-2 | accuracy: 88.9 -> 91.1 | Improved |
| Distill video-grounded knowledge with NST | BERT-small · GLUE diagnostics Knowledge | score: +4.6 | Improved |
Closest papers by meaning
- Distilling Vision-Language Models on Millions of Videos (CVPR 2024 · similarity 0.88)
- PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding (NeurIPS 2025 · similarity 0.88)
- Language-aware Visual Semantic Distillation for Video Question Answering (CVPR 2024 · similarity 0.87)
- Language-Informed Visual Concept Learning (ICLR 2024 · similarity 0.87)
- Cross-Lingual Text-Rich Visual Comprehension: An Information Theory Perspective (AAAI 2025 · similarity 0.87)
- MoVE-KD: Knowledge Distillation for VLMs with Mixture of Visual Encoders (CVPR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
