CoDiRe blends VLM and target model predictions via MSP-based weighting and Optimal Transport rectification to enable stable continual test-time adaptation, outperforming CoTTA by 10.55% on ImageNet-C at 48% of the compute cost.
Scaling up visual and vision-language representation learning with noisy text supervision
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
representative citing papers
VFMTok builds a generalist image tokenizer on frozen VFMs using adaptive quantization and semantic alignment, delivering gFID 1.36 for autoregressive and 1.25 for continuous generation on ImageNet with 3x faster convergence.
GenLIP pretrains ViTs to generate language tokens from images via LM objective without contrastive batches or extra decoders, matching baselines on less data and improving on OCR after multi-resolution continued pretraining.
CLAMP pretrains 3D multi-view encoders with contrastive learning on point clouds and actions, then initializes diffusion policies for more sample-efficient fine-tuning on robotic tasks.
GAF creates 4D dynamic scene models by adding motion to 3D Gaussians, enabling better reconstruction and 7.3% higher success in robotic tasks.
Generative purification with consensus aggregation reduces adversarial illusion attack success rates to near zero on ImageBind while improving alignment on both clean and attacked inputs.
JAM aligns frozen vision and language models via joint autoencoders and multimodal Spread Loss, reliably inducing cross-modal alignment across layer depths, objectives, and model scales.
citing papers explorer
-
Test-Time Distillation for Continual Model Adaptation
CoDiRe blends VLM and target model predictions via MSP-based weighting and Optimal Transport rectification to enable stable continual test-time adaptation, outperforming CoTTA by 10.55% on ImageNet-C at 48% of the compute cost.
-
Vision Foundation Models as Generalist Tokenizers for Image Generation
VFMTok builds a generalist image tokenizer on frozen VFMs using adaptive quantization and semantic alignment, delivering gFID 1.36 for autoregressive and 1.25 for continuous generation on ImageNet with 3x faster convergence.
-
Let ViT Speak: Generative Language-Image Pre-training
GenLIP pretrains ViTs to generate language tokens from images via LM objective without contrastive batches or extra decoders, matching baselines on less data and improving on OCR after multi-resolution continued pretraining.
-
CLAMP: Contrastive Learning for 3D Multi-View Action-Conditioned Robotic Manipulation Pretraining
CLAMP pretrains 3D multi-view encoders with contrastive learning on point clouds and actions, then initializes diffusion policies for more sample-efficient fine-tuning on robotic tasks.
-
GAF: Gaussian Action Field as a 4D Representation for Dynamic World Modeling in Robotic Manipulation
GAF creates 4D dynamic scene models by adding motion to 3D Gaussians, enabling better reconstruction and 7.3% higher success in robotic tasks.
-
Breaking the Illusion: Consensus-Based Generative Mitigation of Adversarial Illusions in Multi-Modal Embeddings
Generative purification with consensus aggregation reduces adversarial illusion attack success rates to near zero on ImageBind while improving alignment on both clean and attacked inputs.
-
Escaping Plato's Cave: JAM for Aligning Independently Trained Vision and Language Models
JAM aligns frozen vision and language models via joint autoencoders and multimodal Spread Loss, reliably inducing cross-modal alignment across layer depths, objectives, and model scales.
- Adjustable Text-Guided Backdoor Attacks with Natural-Word Triggers on Multimodal Pretrained Models