Related papers: Multipole Semantic Attention: A Fast Approximation…

While sparse attention mitigates the computational bottleneck of long-context LLM training, its distributed training process exhibits extreme heterogeneity in both \textit{1)} sequence length and \textit{2)} sparsity sensitivity, leading to…

Softmax Attention has a quadratic time complexity in sequence length, which becomes prohibitive to run at long contexts, even with highly optimized GPU kernels. For example, FlashAttention-2/3 (exact, GPU-optimized implementations of…

Multilingual pretraining typically lacks explicit alignment signals, leading to suboptimal cross-lingual alignment in the representation space. In this work, we show that training standard pretrained models for cross-lingual alignment with…

High-resource languages such as English, enables the pretraining of high-quality large language models (LLMs). The same can not be said for most other languages as LLMs still underperform for non-English languages, likely due to a gap in…

As Large Language Models (LLMs) scale to longer context windows, the computational cost of attention mechanisms, which traditionally grows quadratically with input length, presents a critical challenge for real-time and memory-constrained…

Efficient long-context understanding and reasoning are increasingly vital for large language model (LLM) applications such as multi-turn dialogue and program analysis. However, the core self-attention mechanism scales quadratically with…

This study introduces bifurcated attention, a method designed to enhance language model inference in shared-context batch decoding scenarios. Our approach addresses the challenge of redundant memory IO costs, a critical factor contributing…

Byte-based machine translation systems have shown significant potential in massively multilingual settings. Unicode encoding, which maps each character to specific byte(s), eliminates the emergence of unknown words, even in new languages.…

Prefilling computational costs pose a significant bottleneck for Large Language Models (LLMs) and Large Multimodal Models (LMMs) in long-context settings. While token pruning reduces sequence length, prior methods rely on heuristics that…

We present LongLoRA, an efficient fine-tuning approach that extends the context sizes of pre-trained large language models (LLMs), with limited computation cost. Typically, training LLMs with long context sizes is computationally expensive,…

Effective attention modules have played a crucial role in the success of Transformer-based large language models (LLMs), but the quadratic time and memory complexities of these attention modules also pose a challenge when processing long…

Prompt-tuning is an emerging strategy to adapt large language models (LLM) to downstream tasks by learning a (soft-)prompt parameter from data. Despite its success in LLMs, there is limited theoretical understanding of the power of…

Deploying Large Language Models (LLMs) on edge devices remains challenging due to their quadratically increasing computations with the sequence length. Existing studies for dynamic attention pruning are designed for hardware with massively…

Adapting Large Language Models (LLMs) to a continuous stream of tasks is a critical yet challenging endeavor. While Parameter-Efficient Fine-Tuning (PEFT) methods have become a standard for this, they face a fundamental dilemma in continual…

Learning semantically meaningful sentence embeddings is an open problem in natural language processing. In this work, we propose a sentence embedding learning approach that exploits both visual and textual information via a multimodal…

The large language models have achieved superior performance on various natural language tasks. One major drawback of such approaches is they are resource-intensive in fine-tuning new datasets. Soft-prompt tuning presents a…

Large language models (LLMs) increasingly operate in settings that require reliable long-context understanding, such as retrieval-augmented generation and multi-document reasoning. A common strategy is to fine-tune pretrained short-context…

The per-token cost of transformer inference scales with context length, preventing its application to lifelong in-context learning. Linear attention is an efficient alternative that maintains a constant memory footprint, even on infinite…

We compare the fast training and decoding speed of RETURNN of attention models for translation, due to fast CUDA LSTM kernels, and a fast pure TensorFlow beam search decoder. We show that a layer-wise pretraining scheme for recurrent…

Self-supervised learning (SSL) has greatly advanced speech representation learning, but multilingual SSL models remain constrained to languages encountered during pretraining. Retraining from scratch to incorporate new languages is…