In recent years, end-to-end learnt video codecs have demonstrated their potential to
compete with conventional coding algorithms in term of compression efficiency. However …

While contrastive pre-training is widely employed, its data efficiency problem has remained
relatively under-explored thus far. Existing methods often rely on static coreset selection …

In recent years, Dynamic Sparse Training (DST) has emerged as an alternative to post-
training pruning for generating efficient models. In principle, DST allows for a more memory …

Large language models (LLMs) have made significant strides in complex tasks, yet their
widespread adoption is impeded by substantial computational demands. With hundreds of …

The current landscape in time-series forecasting is dominated by Transformer-based
models. Their high parameter count and corresponding demand in computational resources …

In the field of deep reinforcement learning significant progress has been made, but it seems
we are missing the power of the scaling laws evident in large language models. This …

Dynamic Sparse Training (DST) is an effective approach for addressing the substantial
training resource requirements posed by the ever-increasing size of the Deep Neural …