相关论文: Lossless Anti-Distillation Sampling

Transfer learning promises to reduce the high sample complexity of deep reinforcement learning (RL), yet existing methods struggle with domain shift between source and target environments. Policy distillation provides powerful tactical…

The growing industrial demand for customized and cost-efficient large language models (LLMs) is fueled by the rise of vertical, domain-specific tasks and the need to optimize performance under constraints such as latency and budget.…

In recent years, the rapid development of deep neural networks has brought increased attention to the security and robustness of these models. While existing adversarial attack algorithms have demonstrated success in improving adversarial…

Diffusion models offer superior generation quality at the expense of extensive sampling steps. Distillation methods, with Distribution Matching Distillation (DMD) as a popular example, can mitigate this issue, but performance degradation…

Diffusion distillation, exemplified by Distribution Matching Distillation (DMD), has shown great promise in few-step generation but often sacrifices quality for sampling speed. While integrating Reinforcement Learning (RL) into distillation…

Huge amount of data is the key of the success of deep learning, however, redundant information impairs the generalization ability of the model and increases the burden of calculation. Dataset Distillation (DD) compresses the original…

Adaptive sampling results in dramatic improvements in the recovery of sparse signals in white Gaussian noise. A sequential adaptive sampling-and-refinement procedure called Distilled Sensing (DS) is proposed and analyzed. DS is a form of…

Autoregressive (AR) Large Language Models (LLMs) have demonstrated significant success across numerous tasks. However, the AR modeling paradigm presents certain limitations; for instance, contemporary autoregressive LLMs are trained to…

Privacy-sensitive users require deploying large language models (LLMs) within their own infrastructure (on-premises) to safeguard private data and enable customization. However, vulnerabilities in local environments can lead to unauthorized…

Machine learning as a Service (MLaaS) allows users to query the machine learning model in an API manner, which provides an opportunity for users to enjoy the benefits brought by the high-performance model trained on valuable data. This…

The quality of open-weight language models has dramatically improved in recent years. Sharing weights greatly facilitates model adoption by enabling their use across diverse hardware and software platforms. They also allow for more open…

Current federated learning algorithms take tens of communication rounds transmitting unwieldy model weights under ideal circumstances and hundreds when data is poorly distributed. Inspired by recent work on dataset distillation and…

Adversarial distillation in the standard min-max adversarial training framework aims to transfer adversarial robustness from a large, robust teacher network to a compact student. However, existing work often neglects to incorporate…

Adapting a pretrained diffusion model to new objectives at inference time remains an open problem in generative modeling. Existing steering methods suffer from inaccurate value estimation, especially at high noise levels, which biases…

Langevin dynamics sampling suffers from extremely low generation speed, fundamentally limited by numerous fine-grained iterations to converge to the target distribution. We introduce PID-controlled Langevin Dynamics (PIDLD), a novel…

Knowledge distillation from proprietary LLM APIs poses a growing threat to model providers, yet defenses against this attack remain fragmented and unevaluated. We present DistillGuard, a framework for systematically evaluating output-level…

Large language models (LLMs) exhibit severe multilingual safety misalignment: they possess strong safeguards in high-resource languages but remain highly vulnerable to jailbreak attacks in low-resource languages. Current safety alignment…

Diffusion Models have emerged as a leading class of generative models, yet their iterative sampling process remains computationally expensive. Timestep distillation is a promising technique to accelerate generation, but it often requires…

Although Score Distillation Sampling (SDS) has exhibited remarkable performance in conditional 3D content generation, a comprehensive understanding of its formulation is still lacking, hindering the development of 3D generation. In this…

Knowledge Distillation (KD) is a fundamental technique for compressing large language models (LLMs) into compact, efficient student models. However, existing white-box KD methods mainly focus on balancing ground truth and student-generated…