AI Paper Atlas

Research map / Video / Video-language understanding

TempR1: Improving Temporal Understanding of MLLMs via Temporal-Aware Multi-Task Reinforcement Learning

Tao Wu, Li Yang, Gen Zhan, Yabin Zhang, Yiting Liao, Junlin Li, Deliang Fu, Li Zhang, Limin Wang

CVPR 2026 · Video › Video-language understanding · approach: Long video understanding · 5 citations (OpenAlex)

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 changedWhereMeasuredEffect
add instance number reward to TAL reward (ablation)TempR1 · ActivityNet-v1.3 TALmF1: 69.8 -> 70.6Improved
use DP-based matching instead of sequential matching (ablation)TempR1 · ActivityNet-v1.3 TALmF1: 45.4 -> 70.6Improved
GRPO instead of DAPO (ablation)TempR1 · ActivityNet-v1.3 TALmF1: 47.6 -> 71.0Improved
GSPO instead of GRPO (ablation)TempR1 · ActivityNet-v1.3 TALmF1: 71.0 -> 71.7No clear effect
add DTG and VHD tasks to TG+GVQA joint training (ablation)TempR1 · QVHighlights VHDmIoU: 7.1 -> 70.3Improved
add TAL task to joint training (ablation)TempR1 · ActivityNet-v1.3 TALmF1: 68.2 -> 71.0Improved
add GVQA to TG-only training (ablation)TempR1 · NExTGQAAcc: 62.5 -> 67.5Improved
add DTG to TG+GVQA training (ablation)TempR1 · ActivityNet DTGmIoU: 17.7 -> 59.4Improved
add TAL to four-task training (ablation)TempR1 · NExTGQAAcc: 66.7 -> 70.1Improved
dataset-specific fine-tuning after multi-task RLTempR1 · ActivityNet TGmIoU: 32.2 -> 51.9Improved
RL fine-tuning instead of SFT on same dataTempR1 · general video benchmarksaccuracyImproved
TempR1 multi-task vs VideoChat-R1 on Charades-STATempR1 · Charades-STAmIoU: 60.8 -> 61.4Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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