AI Paper Atlas

Research map / Video / Video-language understanding

Token Mixing: Parameter-Efficient Transfer Learning from Image-Language to Video-Language

Yuqi Liu, Luhui Xu, Pengfei Xiong, Qin Jin

AAAI 2023 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 7 citations (OpenAlex)

Read the paper

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 changedWhereMeasuredEffect
add token mixing to a transformer layer (ablation)partially fine-tuned BLIP ViT-B/16 · MSVDCIDEr: 146.3 -> 146.9Improved
add token mixing to a transformer layer (ablation)partially fine-tuned BLIP ViT-B/16 · MSRVTTR@1: 46.1 -> 47.1Improved
freeze visual projection instead of fine-tuning it (ablation)Token Mixing on BLIP ViT-B/16 · MSVDCIDEr: 146.92 -> 138.87Hurt
freeze language head instead of fine-tuning it (ablation)Token Mixing on BLIP ViT-B/16 · MSVDCIDEr: 146.92 -> 141.15Hurt
freeze both attention and FFN in the token-mixing layer (ablation)Token Mixing on BLIP ViT-B/16 · MSVDCIDEr: 146.92 -> 134.39Hurt
freeze FFN in the token-mixing layer (ablation)Token Mixing on BLIP ViT-B/16 · MSVDCIDEr: 146.92 -> 144.45Hurt
freeze attention in the token-mixing layer (ablation)Token Mixing on BLIP ViT-B/16 · MSVDCIDEr: 146.92 -> 145.46Hurt
apply token mixing rather than spatial-only attention (ablation)fully fine-tuned BLIP ViT-B/16 · MSVDCIDEr: 150.1 -> 151.2Improved
apply token mixing rather than spatial-only attention (ablation)fully fine-tuned BLIP ViT-B/16 · MSRVTTR@1: 47.6 -> 48.5Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy