4 citations · 12 across the 17 of their papers we have counts for
11 papers · 1 filter
Cosmos 3: Omnimodal World Models for Physical AI
NVIDIA, :, Aditi +293
We introduce Cosmos 3, a family of omnimodal world models designed to jointly process and generate language, image, video, audio, and action sequences within a unified mixture-of-t…
Guidance Contrastive Token Credit Assignment for Discrete Policy Optimization
Shufan Li, Konstantinos Kallidromitis, Akash Gokul +2
Group-advantage-based reinforcement learning methods, such as GRPO and DAPO, have demonstrated strong performance across diverse domains, including mathematical reasoning and text-…
LaViDa: A Large Diffusion Language Model for Multimodal Understanding
Shufan Li, Konstantinos Kallidromitis, Hritik Bansal +7
Modern Vision-Language Models (VLMs) can solve a wide range of tasks requiring visual reasoning. In real-world scenarios, desirable properties for VLMs include fast inference and c…
Reflect-DiT: Inference-Time Scaling for Text-to-Image Diffusion Transformers via In-Context Reflection
Shufan Li, Konstantinos Kallidromitis, Akash Gokul +4
The predominant approach to advancing text-to-image generation has been training-time scaling, where larger models are trained on more data using greater computational resources. W…
Aligning Diffusion Models by Optimizing Human Utility
Shufan Li, Konstantinos Kallidromitis, Akash Gokul +2
We present Diffusion-KTO, a novel approach for aligning text-to-image diffusion models by formulating the alignment objective as the maximization of expected human utility. Since t…
BootPIG: Bootstrapping Zero-shot Personalized Image Generation Capabilities in Pretrained Diffusion Models
Senthil Purushwalkam, Akash Gokul, Shafiq Joty +1
Recent text-to-image generation models have demonstrated incredible success in generating images that faithfully follow input prompts. However, the requirement of using words to de…
End-to-End Diffusion Latent Optimization Improves Classifier Guidance
Bram Wallace, Akash Gokul, Stefano Ermon +1
Classifier guidance -- using the gradients of an image classifier to steer the generations of a diffusion model -- has the potential to dramatically expand the creative control ove…
EDICT: Exact Diffusion Inversion via Coupled Transformations
Bram Wallace, Akash Gokul, Nikhil Naik
Finding an initial noise vector that produces an input image when fed into the diffusion process (known as inversion) is an important problem in denoising diffusion models (DDMs),…
Prior Knowledge-Guided Attention in Self-Supervised Vision Transformers
Kevin Miao, Akash Gokul, Raghav Singh +5
Recent trends in self-supervised representation learning have focused on removing inductive biases from training pipelines. However, inductive biases can be useful in settings when…
Refine and Represent: Region-to-Object Representation Learning
Akash Gokul, Konstantinos Kallidromitis, Shufan Li +4
Recent works in self-supervised learning have demonstrated strong performance on scene-level dense prediction tasks by pretraining with object-centric or region-based correspondenc…
Remembering for the Right Reasons: Explanations Reduce Catastrophic Forgetting
Sayna Ebrahimi, Suzanne Petryk, Akash Gokul +4
The goal of continual learning (CL) is to learn a sequence of tasks without suffering from the phenomenon of catastrophic forgetting. Previous work has shown that leveraging memory…