Experiments on video temporal grounding and highlight detection benchmarks show that TEMPURA substantially improves strong base VLMs across model families and scales, confirming that combining event-level reasoning with fine-grained temporal segmentation is an effective recipe for video temporal understanding.
This work introduces a closed-loop, test-time prompt refinement framework that requires no additional training of the underlying T2I model, termed TIR, and demonstrates that this closed-loop strategy improves alignment and visual coherence across multiple benchmark datasets, all while maintaining plug-and-play integrat...
Mohammad Abdul Hafeez Khan, Y. Jain, Siddhartha Bhattacharyya et al.· 2025 IEEE/CVF International...· 17 citations· ⚡2
This work presents Echoverse, which compiles specifications into stateful applications whose tasks are graded against the application's own database, and a co-evolution loop that reads every graded rollout twice: as repairs to the environment, its tasks and its verifier, and as training signal for the model.
