Pioneering dual-encoder pre-training works (e.g., CLIP and ALIGN) have revealed the potential of aligning multi-modal representations with contrastive learning. However, these require a tremendous amount data computational resources billion-level web hundreds GPUs), which prevent researchers limited from reproduction further exploration. To this end, we propose stack novel methods, significantl...

In this paper, we define a framework for media resources manipulation in an adaptive content delivery system. We discuss the media resources manipulation in an adaptation and negotiation architecture in terms of binary relationship definition using a so-called "related resources" approach, resources extraction, adaptation and delivery. We show how different resources can be related by predefine...

We consider a simple restriction of the PRAM model (called PPRAM), where the input is arbitrarily partitioned between a xed set of p processors and the shared memory is restricted to m cells. This model allows for investigating the tradeoos/ bottlenecks with respect to the communication bandwidth (modeled by the shared memory size m) and the number of processors p. The model is quite simple and...

Peer punishment has been an important instrument in enabling social norms to emerge. However, it is usually assumed that unlimited resources are available to agents to cope with the resulting enforcement costs. In this paper, we use a modified version of the metanorm model [1] to investigate this, and show that it allows norm emergence only in limited cases under bounded resources.