PixMo Collection A set of vision-language datasets built by Ai2 and used to train the Molmo family of models. Read more at https://molmo.allenai.org/blog β’ 9 items β’ Updated 27 days ago β’ 50
Inf-CL Collection The corresponding demos/checkpoints/papers/datasets of Inf-CL. β’ 2 items β’ Updated Oct 25 β’ 3
TOMATO: Assessing Visual Temporal Reasoning Capabilities in Multimodal Foundation Models Paper β’ 2410.23266 β’ Published Oct 30 β’ 20
Breaking the Memory Barrier: Near Infinite Batch Size Scaling for Contrastive Loss Paper β’ 2410.17243 β’ Published Oct 22 β’ 89
Stabilize the Latent Space for Image Autoregressive Modeling: A Unified Perspective Paper β’ 2410.12490 β’ Published Oct 16 β’ 8
The Curse of Multi-Modalities: Evaluating Hallucinations of Large Multimodal Models across Language, Visual, and Audio Paper β’ 2410.12787 β’ Published Oct 16 β’ 30
A Controlled Study on Long Context Extension and Generalization in LLMs Paper β’ 2409.12181 β’ Published Sep 18 β’ 43
SeaLLMs 3: Open Foundation and Chat Multilingual Large Language Models for Southeast Asian Languages Paper β’ 2407.19672 β’ Published Jul 29 β’ 55
3D-GRAND: A Million-Scale Dataset for 3D-LLMs with Better Grounding and Less Hallucination Paper β’ 2406.05132 β’ Published Jun 7 β’ 27
What If We Recaption Billions of Web Images with LLaMA-3? Paper β’ 2406.08478 β’ Published Jun 12 β’ 39
VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs Paper β’ 2406.07476 β’ Published Jun 11 β’ 32
Video-LaVIT: Unified Video-Language Pre-training with Decoupled Visual-Motional Tokenization Paper β’ 2402.03161 β’ Published Feb 5 β’ 14