Related papers: Expert-Choice Routing Enables Adaptive Computation…

Diffusion language models (DLMs) have emerged as a promising alternative to autoregressive (AR) models for language modeling, allowing flexible generation order and parallel generation of multiple tokens. However, this flexibility…

Mixture-of-Experts (MoE), while offering significant advantages as a Large Language Model (LLM) architecture, faces substantial challenges when deployed on low-cost edge devices with tight memory constraints. Expert offloading mitigates…

Masked diffusion models (MDMs) for text offer a compelling alternative to traditional autoregressive language models. Parallel generation makes them efficient, but their computational capabilities and the limitations inherent in their…

This paper systematically diagnoses the training failure modes of Token-Choice sparse Mixture-of-Experts (MoE) on video Diffusion Transformers. Starting from a pretrained dense model of about 5 billion parameters, we convert it into an MoE…

Large language models, such as OpenAI's ChatGPT, have demonstrated exceptional language understanding capabilities in various NLP tasks. Sparsely activated mixture-of-experts (MoE) has emerged as a promising solution for scaling models…

Detecting machine-generated text (MGT) has emerged as a critical challenge, driven by the rapid advancement of large language models (LLMs) capable of producing highly realistic, human-like content. However, the performance of current…

Diffusion Language models (DLMs) are a promising avenue for text generation due to their practical properties on tractable controllable generation. They also have the advantage of not having to predict text autoregressively. However,…

We propose Tensor-Trained Low-Rank Adaptation Mixture of Experts (TT-LoRA MoE), a novel computational framework integrating Parameter-Efficient Fine-Tuning (PEFT) with sparse MoE routing to address scalability challenges in large model…

Mixture-of-Experts layers achieve compute efficiency through weight sparsity: each token activates only a subset of experts. Data sparsity, where each expert processes only a subset of tokens, offers a complementary axis. Expert-choice…

Mixture-of-Experts (MoE) has recently emerged as the mainstream architecture for efficiently scaling large language models while maintaining near-constant computational cost. Expert parallelism distributes parameters by partitioning experts…

This paper shows how diffusion language models (DLMs) can be used as effective and efficient retrievers. Existing DLM-based retrievers (e.g., DiffEmbed) follow BERT-style encoding, representing each query or passage as a single mean-pooled…

Mixture-of-Experts (MoE) has become a prominent paradigm for scaling Large Language Models (LLMs). Parameter-efficient fine-tuning methods, such as LoRA, are widely adopted to adapt pretrained MoE LLMs to downstream tasks. However, existing…

Sparsely activated Mixture-of-Experts (MoE) models are widely adopted to scale up model capacity without increasing the computation budget. However, vanilla TopK routers are trained in a discontinuous, non-differentiable way, limiting their…

Mobile edge computing (MEC) emerges recently as a promising solution to relieve resource-limited mobile devices from computation-intensive tasks, which enables devices to offload workloads to nearby MEC servers and improve the quality of…

In the context of continuously rising global air traffic, efficient and safe Conflict Detection and Resolution (CD&R) is paramount for air traffic management. Although Deep Reinforcement Learning (DRL) offers a promising pathway for CD&R…

Diffusion Language Models (DLMs) offer a promising parallel generation paradigm but suffer from slow inference due to numerous refinement steps and the inability to use standard KV caching. We introduce CDLM (Consistency Diffusion Language…

Discrete diffusion models (DMs) have achieved strong performance in language and other discrete domains, offering a compelling alternative to autoregressive modeling. Yet this performance typically depends on large training datasets,…

We present Entropy Adaptive Decoding (EAD), a novel approach for efficient language model inference that dynamically switches between different-sized models based on prediction uncertainty. By monitoring rolling entropy in model logit…

Diffusion Large Language Models (dLLMs) have emerged as a promising alternative to purely autoregressive language models because they can decode multiple tokens in parallel. However, state-of-the-art block-wise dLLMs rely on a "remasking"…

Mobile edge computing (MEC) is a promising technique to improve the computational capacity of smart devices (SDs) in Internet of Things (IoT). However, the performance of MEC is restricted due to its fixed location and limited service…