相关论文: Peek2: Regex-free Byte-level Byte-Pair Encoding Pr…

Post-training quantization (PTQ) is a primary approach for deploying large language models without fine-tuning, and the quantized performance is often strongly affected by the calibration in PTQ. By contrast, in vision-language models…

Tokenization methods like Byte-Pair Encoding (BPE) enhance computational efficiency in large language models (LLMs) but often obscure internal character structures within tokens. This limitation hinders LLMs' ability to predict precise…

We present PLONQ, a progressive neural image compression scheme which pushes the boundary of variable bitrate compression by allowing quality scalable coding with a single bitstream. In contrast to existing learned variable bitrate…

Integrating vision models into large language models (LLMs) has sparked significant interest in creating vision-language foundation models, especially for video understanding. Recent methods often utilize memory banks to handle untrimmed…

We present a new dataset condensation framework termed Squeeze, Recover and Relabel (SRe$^2$L) that decouples the bilevel optimization of model and synthetic data during training, to handle varying scales of datasets, model architectures…

Modern Large Language Model serving system batches multiple requests to achieve high throughput, while batching attention operations is challenging, rendering memory bandwidth a critical bottleneck. The community relies on high-end GPUs…

In language processing, transformers benefit greatly from text being condensed. This is achieved through a larger vocabulary that captures word fragments instead of plain characters. This is often done with Byte Pair Encoding. In the…

Byte Pair Encoding (BPE) serves as a foundation method for text tokenization in the Natural Language Processing (NLP) field. Despite its wide adoption, the original BPE algorithm harbors an inherent flaw: it inadvertently introduces a…

With recent rapid growth of large language models (LLMs), discrete speech tokenization has played an important role for injecting speech into LLMs. However, this discretization gives rise to a loss of information, consequently impairing…

Neural Architecture Search (NAS) has enabled automatic discovery of more efficient neural network architectures, especially for mobile and embedded vision applications. Although recent research has proposed ways of quickly estimating…

Modern foundational models are often compressed via a combination of structured pruning and re-training to meet the strict compute, memory, and connectivity constraints of edge deployments. While state-of-the-art pruning schemes target the…

Diffusion Language Models (DLMs) have recently emerged as a strong alternative to autoregressive language models (LMs). DLMs offer comparable accuracy with faster inference speed via parallel decoding. However, standard DLM decoding…

Recent works have widely adopted large language model pretraining for source code, suggested source code-specific pretraining objectives and investigated the applicability of various Transformer-based language model architectures for source…

The adoption of Vision Transformers (ViTs) in resource-constrained applications necessitates improvements in inference throughput. To this end several token pruning and merging approaches have been proposed that improve efficiency by…

Mixture-of-Experts models have become a dominant architecture for scaling Large Language Models by activating only a sparse subset of experts per token. However, latency-critical MoE inference faces a fundamental tension: while expert…

Large language models (LLMs) enable long-context tasks but face efficiency challenges due to the growing key-value (KV) cache. We propose LeanK, a learning-based method that prunes unimportant key (K) cache channels by leveraging static…

Edge detection is a fundamental image analysis task that underpins numerous high-level vision applications. Recent advances in Transformer architectures have significantly improved edge quality by capturing long-range dependencies, but this…

Transformer-based large language models (LLMs) have already achieved remarkable results on long-text tasks, but the limited GPU memory (VRAM) resources struggle to accommodate the linearly growing demand for key-value (KV) cache as the…

We investigate preprocessing for vertex-subset problems on graphs. While the notion of kernelization, originating in parameterized complexity theory, is a formalization of provably effective preprocessing aimed at reducing the total…

Video Large Language Models (VideoLLMs) have made significant strides in video understanding but struggle with long videos due to the limitations of their backbone LLMs. Existing solutions rely on length extrapolation, which is…