12 accepted papers
While Mixture of Experts (MoE) models achieve remarkable efficiency by activating only subsets of parameters, they suffer from high memory access costs during inference. Memory-layer architectures offer an appealing alternative with very few memory access, but previous attempts like UltraMem have on…
Transformers have become the de facto architecture for a wide range of machine learning tasks, particularly in large language models (LLMs). Despite their remarkable performance, many challenges remain in training deep transformer networks, especially regarding the position of the layer normalizatio…
We present hyper-connections, a simple yet effective method that can serve as an alternative to residual connections. This approach specifically addresses common drawbacks observed in residual connection variants, such as the seesaw effect between gradient vanishing and representation collapse. Theo…
Empirical evidence indicates that LLMs exhibit spontaneous cross-lingual alignment. However, although LLMs show promising cross-lingual alignment in Information Extraction (IE), a significant imbalance across languages persists, highlighting an underlying deficiency. To address this, we propose Know…
Tokenization is a fundamental component of large language models (LLMs), yet its influence on model scaling and performance is not fully explored. In this paper, we introduce Over-Tokenized Transformers, a novel framework that decouples input and output vocabularies to improve language modeling perf…
Transformers have found extensive applications across various domains due to their powerful fitting capabilities. This success can be partially attributed to their inherent nonlinearity. Thus, in addition to the ReLU function employed in the original transformer architecture, researchers have explor…
The increasing application of multi-modal large language models (MLLMs) across various sectors has spotlighted the essence of their output reliability and accuracy, particularly their ability to produce content grounded in factual information (e.g. common and domain-specific knowledge). In this work…
The expanding computational costs and limited resources underscore the critical need for budgeted-iteration training, which aims to achieve optimal learning within predetermined iteration budgets. While learning rate schedules fundamentally govern the performance of different networks and tasks, par…
It is widely acknowledged that the performance of Transformer models is logarithmically related to their number of parameters and computational complexity. While approaches like Mixture of Experts (MoE) decouple parameter count from computational complexity, they still face challenges in inference d…
Temporal Knowledge Graph Question Answering (TKGQA) aims to answer questions with temporal intent over Temporal Knowledge Graphs (TKGs). The core challenge of this task lies in understanding the complex semantic information regarding multiple types of time constraints (e.g., before, first) in questi…
Reasoning over Temporal Knowledge Graphs (TKGs) that predicts temporal facts (e.g., events) in the future is crucial for many applications. The temporal facts in existing TKGs only contain their core entities (i.e., the entities playing core roles therein) and formulate them as quadruples, i.e., (su…
Event coreference resolution aims to classify all event mentions that refer to the same real-world event into the same group, which is necessary to information aggregation and many downstream applications. To resolve event coreference, existing methods usually calculate the similarities between even…