Related papers: ADWIN: Adaptive Windows for Horizon-Aware On-Polic…
Online Knowledge Distillation (OKD) methods streamline the distillation training process into a single stage, eliminating the need for knowledge transfer from a pretrained teacher network to a more compact student network. This paper…
Adapting vision-language models to remote sensing imagery remains challenging due to two key factors: limited semantic coverage in textual representations and insufficient adaptability of visual features. These issues are particularly…
On-policy self-distillation (self-OPD) densifies reinforcement learning with verifiable rewards (RLVR) by letting a policy teach itself under privileged context. We find that when this guidance spans the full response, all-token KL spends…
Decision transformer based sequential policies have emerged as a powerful paradigm in offline reinforcement learning (RL), yet their efficacy remains constrained by the quality of static datasets and inherent architectural limitations.…
Reinforcement learning from verifiable rewards (RLVR) suffers from sparse outcome signals, creating severe exploration bottlenecks on complex reasoning tasks. Recent on-policy self-distillation methods attempt to address this by utilizing…
Online Action Detection (OAD) in videos is proposed as a per-frame labeling task to address the real-time prediction tasks that can only obtain the previous and current video frames. This paper presents a novel learning-with-privileged…
The transfer of knowledge from one policy to another is an important tool in Deep Reinforcement Learning. This process, referred to as distillation, has been used to great success, for example, by enhancing the optimisation of agents,…
Knowledge distillation has attracted a great deal of interest recently to compress pre-trained language models. However, existing knowledge distillation methods suffer from two limitations. First, the student model simply imitates the…
In modern large language models (LLMs), LLM alignment is of crucial importance and is typically achieved through methods such as reinforcement learning from human feedback (RLHF) and direct preference optimization (DPO). However, in most…
In scenarios where obtaining real-time labels proves challenging, conventional approaches may result in sub-optimal performance. This paper presents an optimal strategy for streaming contexts with limited labeled data, introducing an…
Learning control policies in simulation enables rapid, safe, and cost-effective development of advanced robotic capabilities. However, transferring these policies to the real world remains difficult due to the sim-to-real gap, where…
While knowledge distillation has seen widespread use in pre-training and instruction tuning, its application to aligning language models with human preferences remains underexplored, particularly in the more realistic cross-tokenizer…
Reinforcement learning (RL) has been widely used to train LLM agents for multi-turn interactive tasks, but its sample efficiency is severely limited by sparse rewards and long horizons. On-policy self-distillation (OPSD) alleviates this by…
Knowledge distillation (KD) is a widely adopted technique for transferring knowledge from large language models to smaller student models; however, conventional supervised KD often suffers from a distribution mismatch between training and…
Current state-of-the-art object detectors are at the expense of high computational costs and are hard to deploy to low-end devices. Knowledge distillation, which aims at training a smaller student network by transferring knowledge from a…
Fine-tuning large pre-trained language models on downstream tasks is apt to suffer from overfitting when limited training data is available. While dropout proves to be an effective antidote by randomly dropping a proportion of units,…
Vision-based policies are widely applied in robotics for tasks such as manipulation and locomotion. On lightweight mobile robots, however, they face a trilemma of limited scene transferability, restricted onboard computation resources, and…
Knowledge distillation (KD) is a key technique for compressing large-scale language models (LLMs), yet prevailing logit-based methods typically employ static strategies that are misaligned with the dynamic learning process of student…
Spoken Language Understanding (SLU) is a core component of conversational systems, enabling machines to interpret user utterances. Despite its importance, developing effective SLU systems remains challenging due to the scarcity of labeled…
Speculative decoding accelerates large language model inference by pairing a target model with a lightweight draft model whose proposed tokens are verified in parallel. A common way to build draft models, like EAGLE3 or DFlash is supervised…