2408.03326] LLaVA-OneVision: Easy Visual Task Transfer Skip to main content We gratefully acknowledge support from the

2408.03326] LLaVA-OneVision: Easy Visual Task Transfer Skip to main content We gratefully acknowledge support from the

2407.00369] How to Train Your Fact Verifier: Knowledge Transfer with Multimodal Open Models Skip to main

2407.00369] How to Train Your Fact Verifier: Knowledge Transfer with Multimodal Open Models Skip to main

2502.10377] ReStyle3D: Scene-Level Appearance Transfer with Semantic Correspondences arXiv Is Hiring a DevOps Engineer Work on one

2502.10377] ReStyle3D: Scene-Level Appearance Transfer with Semantic Correspondences Happy Open Access Week from arXiv! YOU make open access possible

to Stay in the Pre-train Basin: Insights on Ensembling in Transfer Learning Skip to

2304.02744] StyleGAN Salon: Multi-View Latent Optimization for Pose-Invariant Hairstyle Transfer Happy Open Access Week from arXiv

2108.12847] Non-Parametric Neural Style Transfer Skip to main content We gratefully acknowledge support from the Simons

2205.02841] Understanding Transfer Learning for Chest Radiograph Clinical Report Generation with Modified Transformer Architectures Skip to

2303.09665] LOCATE: Localize and Transfer Object Parts for Weakly Supervised Affordance Grounding Skip to main

2303.09665] LOCATE: Localize and Transfer Object Parts for Weakly Supervised Affordance Grounding Skip to main

2210.00912] Federated Domain Generalization for Image Recognition via Cross-Client Style Transfer Skip to main content

2503.09838] BioSpark: Beyond Analogical Inspiration to LLM-augmented Transfer Skip to main content We gratefully acknowledge

2503.09838] BioSpark: Beyond Analogical Inspiration to LLM-augmented Transfer Skip to main content We gratefully acknowledge

cond-mat/0702502] Spatial Transportation Networks with Transfer Costs: Asymptotic Optimality of Hub and Spoke Models Skip

2210.13702] DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Reality Skip

2206.06522] LST: Ladder Side-Tuning for Parameter and Memory Efficient Transfer Learning Skip to main

2112.06825] VL-Adapter: Parameter-Efficient Transfer Learning for Vision-and-Language Tasks Skip to main

1710.00756] Progressive Color Transfer with Dense Semantic Correspondences Skip to main content We gratefully acknowledge support from the

1910.10683] Exploring the Limits of Transfer Learning with a Unified Text-to-Text

Abstract: We present TextureDreamer, a novel image-guided texture synthesis method to transfer relightable textures from a

modules for every new task misses an opportunity for cross-task knowledge transfer. We propose Improvise to

Abstract: We present TextureDreamer, a novel image-guided texture synthesis method to transfer relightable textures from a

and action recognition. Moreover, Florence demonstrates outstanding performance in many types of transfer learning: fully sampled fine

and action recognition. Moreover, Florence demonstrates outstanding performance in many types of transfer learning: fully sampled fine

data efficiency is multitask learning with shared neural network parameters, where efficiency may be improved through transfer across related tasks. In

concepts (or describe new ones) to enable zero-shot and few-shot transfer of the

data efficiency is multitask learning with shared neural network parameters, where efficiency may be improved through transfer across related tasks. In

concepts (or describe new ones) to enable zero-shot and few-shot transfer of the

datasets, and performs competitively to state-of-the-art transfer-learning methods requiring greater

demonstrate the quality of the learned dense correspondences through segmentation mask transfer on multiple datasets. We

methods can help mitigate forgetting during multimodal task learning, but do not enable cross-task knowledge transfer. We envision that CLiMB

opening the door to a notion of generative texture transfer. Comments: CVPR 2024. Code

dual pathway architecture that allows locale-specific features to be encapsulated, while still enabling transfer to multiple

has achieved impressive results on image classification in both zero-shot and transfer learning settings. However, we

dual pathway architecture that allows locale-specific features to be encapsulated, while still enabling transfer to multiple

has achieved impressive results on image classification in both zero-shot and transfer learning settings. However, we

multi-teacher distillation, where each teacher is specialized in one domain, to transfer detailed domain-specific knowledge

control module with our model to capture identity-disentangled facial expressions, facilitating accurate expression transfer for enhanced

We further demonstrate the suitability of the dataset for transfer learning by showing that

multi-teacher distillation, where each teacher is specialized in one domain, to transfer detailed domain-specific knowledge

guiding the hand pose estimation. Such structural and contact priors can easily transfer to real

of images, and training and evaluating simulation-to-real transfer for a

control module with our model to capture identity-disentangled facial expressions, facilitating accurate expression transfer for enhanced

We further demonstrate the suitability of the dataset for transfer learning by showing that

The model is designed and trained to be promptable, so it can transfer zero-shot to

materials in real photos, and employ 3D-2D alignment techniques to transfer materials to different

PDF Abstract: Universal Domain Adaptation (UniDA) deals with the problem of knowledge transfer between two datasets with

can fully exploit unlabeled target domain data in object detection training and transfer knowledge between domains via

margin, while using up to $95\%$ fewer weights. Furthermore, we devise a transfer learning setting for

that while slight noise in pre-training can benefit in-domain (ID) transfer performance, where the

The model is designed and trained to be promptable, so it can transfer zero-shot to

Pangea. In extensive experiments, our new system shows significant superiority, especially in transfer learning. Our code and

to-real adaptation, the multitask policies pretrained on GPT4-generated simulation tasks exhibit stronger transfer to unseen

particular, we consider models that invoke processes of gene birth (duplication and transfer) and death

the generated demonstrations and exhibits strong sim-to-real zero-shot transfer. Combining the proposed

the generated demonstrations and exhibits strong sim-to-real zero-shot transfer. Combining the proposed

text-to-image generation, and even edit prompts for style transfer! Finally, being a

label as supervision. To this end, we devise a cross-view knowledge transfer framework that extracts affordance

motion synthesis is already extensively studied and benchmarked, it remains challenging to transfer this success to

different views. To this end, we devise a cross-view affordance knowledge transfer framework that extracts affordance

guiding the hand pose estimation. Such structural and contact priors can easily transfer to real

contrastive learning, AI alignment, AI capabilities, robotics, out-of-distribution (OOD) generalization, continual learning, transfer learning, uncertainty estimation / calibration

images and texts in a common feature space, which allows zero-shot transfer to a

text-to-image generation, and even edit prompts for style transfer! Finally, being a

most methods are currently either limited to specific editing types (e.g., object overlay, style transfer), or apply to

motion synthesis is already extensively studied and benchmarked, it remains challenging to transfer this success to

approaches to making the models more knowledge efficient such as multi-task learning, transfer learning, weakly supervised and

Abu Alhaija and 38 other authors View PDF HTML (experimental) Abstract: We introduce Cosmos-Transfer, a conditional

strong control capabilities across diverse tasks, including mesh-to-video generation, camera control, motion transfer, and object

and golden answers to evaluate models effectively on vision-centric tasks such as style transfer, a challenging

examples, malicious inputs crafted to fool trained models. Adversarial examples often exhibit black-box transfer, meaning that adversarial examples

strong control capabilities across diverse tasks, including mesh-to-video generation, camera control, motion transfer, and object

and 9 other authors View PDF Abstract: Prompt tuning is a new few-shot transfer learning technique that only

may well be from the use of power beaming to transfer energy and accelerate

paper, we present the first framework for one-shot 3D portrait style transfer, which can generate 3D

one of the sets and performs the only possible legal transfer of a

PDF Abstract: Learning to solve complex sequences of tasks--while both leveraging transfer and avoiding

a high-capacity model trained on this data, which we call RT-X, exhibits positive transfer and improves

and ethical considerations). Though foundation models are based on standard deep learning and transfer learning, their scale results

for various downstream tasks. We evaluate FeatureNeRF on tasks of 2D/3D semantic keypoint transfer and 2D

other authors View PDF HTML (experimental) Abstract: Reinforcement learning combined with sim-to-real transfer offers a general

real-time humanoid motion imitator in simulation using these refined motions and transfer it to the

the benefits of this approach in terms of value function transfer, learning within redundant action

PDF Abstract: Learning to solve complex sequences of tasks--while both leveraging transfer and avoiding

to different problems in machine learning, such as generative modeling and transfer learning. In this

type of covenant transaction that enforces a time-lock on the transfer of control

other authors View PDF HTML (experimental) Abstract: Reinforcement learning combined with sim-to-real transfer offers a general

dataset of 580k paired human images across 100k unique identities. For precise appearance transfer, we introduce a

train both levels of policies in simulation and perform Sim2Real transfer for real

is used to reference learned visual concepts (or describe new ones) enabling zero-shot transfer of the

a variety of observation modalities, on multiple embodiments, and further, exhibits positive transfer: the model

the simulator to align effectively with real-world dynamics. We evaluate ASAP across three transfer scenarios: IsaacGym to