AI Paper Atlas

Research map / Video / Video-language understanding

Towards Temporal Compositional Reasoning in Long-Form Sports Videos

Siyu Cao, Lu Zhang, Ruizhe Zeng, Zhi-yong Liu

ECCV 2026 · Video › Video-language understanding · approach: Event and anomaly reasoning in video · 1 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
fine-tune with direct SFT (ablation)Qwen3-VL-4B · SportsTime (Avg)open-ended QA accuracy (%): 25.15 -> 17.61Hurt
fine-tune with native GRPO (ablation)Qwen3-VL-4B · SportsTime (Avg)open-ended QA accuracy (%): 25.15 -> 26.72Improved
fine-tune with native GRPO (ablation)Qwen3-VL-4B · SportsTime (Temporal)open-ended QA accuracy (%): 13.14 -> 11.76Hurt
replace native GRPO with temporal-reward GRPO (tr-GRPO w/ ts) (ablation)Qwen3-VL-4B · SportsTime (Avg)open-ended QA accuracy (%): 26.72 -> 27.31Improved
replace native GRPO with temporal-reward GRPO (tr-GRPO w/ ts) (ablation)Qwen3-VL-4B · SportsTime (Tactical)open-ended QA accuracy (%): 32.65 -> 35.73Improved
add timestamp overlay to frames in tr-GRPO (ablation)Qwen3-VL-4B · SportsTime (Avg)open-ended QA accuracy (%): 26.12 -> 27.31Improved
add anchor-triggered interactive observation (AT-IO) at inference (ablation)Qwen3-VL-4B · SportsTime (Avg)open-ended QA accuracy (%): 27.31 -> 29.24Improved
add anchor-triggered interactive observation (AT-IO) at inference (ablation)Qwen3-VL-4B · SportsTime (Temporal)open-ended QA accuracy (%): 13.73 -> 15.43Improved
add anchor-triggered interactive observation (AT-IO) at inference (ablation)Qwen3-VL-4B · SportsTime (Tactical)open-ended QA accuracy (%): 35.73 -> 34.55Hurt
add tr-GRPO reward over native GRPO for temporal groundingQwen3-VL-4B · SportsTime (SGA objective)mIoU / Hit@0.5: 0.12 -> 0.58Improved
prompt zero-shot CoT for time spans instead of no time promptQwen3-VL-4B · SportsTime (SGA objective)anchor rate (%): 19.49% anchor, mIoU 0.12Mixed
apply CoTR on general long-video benchmarksQwen3-VL-4B · LVBenchaccuracy (%): 56.2 -> 59.9Improved

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