VideoNet: A Large-Scale Dataset for Domain-Specific Action Recognition
VideoNet is a large-scale domain-specific action recognition benchmark and training dataset with 1,000 distinct actions across 37 domains, designed to revitalize action recognition evaluation for modern vision-language models.
Jun 2, 2026
SAGE: Training Smart Any-Horizon Agents for Long Video Reasoning with Reinforcement Learning
SAGE is an agent system that enables flexible video reasoning across different time scales by using multi-turn processing and reinforcement learning, achieving improved performance on long-form video analysis tasks.
Jun 1, 2026
HinTel-AlignBench: A Framework and Benchmark for Hindi-Telugu with English-Aligned Samples
We present HinTel-AlignBench, a scalable framework and comprehensive benchmark for evaluating Vision-Language Models in Hindi and Telugu with English-aligned samples.
May 31, 2026
Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models
We present Molmo, a new family of state-of-the-art VLMs. Starting from pre-trained vision encoders and language-only LLMs, the entire remainder of our VLM pipeline – weights, code, data, and evaluations – is open and free from VLM distillation.
Jun 1, 2025
SIDGAN: High-Resolution Dubbed Video Generation via Shift-Invariant Learning
This work introduces shift-invariant learning, which generates photo-realistic high-resolution videos with accurate Lip-Sync with outperforms state-of-the-art methods on multiple benchmark datasets, in terms of photo-realism, identity preservation, and Lip-Sync accuracy.
Oct 1, 2023
MODEFORMER: Modality-Preserving Embedding For Audio-Video Synchronization Using Transformers
Jun 4, 2023
Automatic Generation and Evaluation of Usable and Secure Audio reCAPTCHA
Oct 24, 2019
Semantic Segmentation with Scarce Data
Jul 16, 2018
Enterprise Scale Privacy Aware Occupancy Sensing
Jul 2, 2018