English
Related papers

Related papers: Visual-Advantage On-Policy Distillation for Vision…

200 papers

Knowledge distillation is a popular approach for enhancing the performance of ''student'' models, with lower representational capacity, by taking advantage of more powerful ''teacher'' models. Despite its apparent simplicity and widespread…

Machine Learning · Computer Science 2023-12-12 Mher Safaryan , Alexandra Peste , Dan Alistarh

Partial observability is a notorious challenge in reinforcement learning (RL), due to the need to learn complex, history-dependent policies. Recent empirical successes have used privileged expert distillation--which leverages availability…

Machine Learning · Computer Science 2025-10-06 Yuda Song , Dhruv Rohatgi , Aarti Singh , J. Andrew Bagnell

Training VideoLLMs for complex reasoning remains challenging due to sparse sequence level rewards and the lack of fine grained credit assignment over long, temporally grounded reasoning trajectories. While reinforcement learning with…

Computer Vision and Pattern Recognition · Computer Science 2026-05-25 Hao Lin , Kunyang Lv , Xu Jiang , Jingqi Tian , Zhongjing Du , Jiayu Ding , Qiaoman Zhang , Hongbo Jin

Large Vision Language Models (LVLMs) have achieved remarkable progress, yet they often suffer from language bias, producing answers without relying on visual evidence. While prior work attempts to mitigate this issue through decoding…

Computer Vision and Pattern Recognition · Computer Science 2026-05-22 Seulbi Lee , Sangheum Hwang

Large vision-language models (LVLMs) generally contain significantly more visual tokens than their textual counterparts, resulting in a considerable computational burden. Recent efforts have been made to tackle this issue by pruning visual…

Computer Vision and Pattern Recognition · Computer Science 2025-05-13 Qizhe Zhang , Aosong Cheng , Ming Lu , Renrui Zhang , Zhiyong Zhuo , Jiajun Cao , Shaobo Guo , Qi She , Shanghang Zhang

Training end-to-end policies from image data to directly predict navigation actions for robotic systems has proven inherently difficult. Existing approaches often suffer from either the sim-to-real gap during policy transfer or a limited…

Robotics · Computer Science 2026-03-17 Lazar Milikic , Manthan Patel , Jonas Frey

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…

Artificial Intelligence · Computer Science 2026-05-12 Jiaxuan Wang , Xuan Ouyang , Zhiyu Chen , Yulan Hu , Zheng Pan , Xin Li , Lan-Zhe Guo

In medical healthcare, obtaining detailed annotations is challenging, highlighting the need for robust Vision-Language Models (VLMs). Pretrained VLMs enable fine-tuning on small datasets or zero-shot inference, achieving performance…

Computer Vision and Pattern Recognition · Computer Science 2025-12-03 Phuc Pham , Nhu Pham , Ngoc Quoc Ly

The use of multi-camera views simultaneously has been shown to improve the generalization capabilities and performance of visual policies. However, the hardware cost and design constraints in real-world scenarios can potentially make it…

Robotics · Computer Science 2023-12-05 Cihan Acar , Kuluhan Binici , Alp Tekirdağ , Yan Wu

On-policy distillation (OPD) is an increasingly important paradigm for post-training language models. However, we identify a pervasive Scaling Law of Miscalibration: while OPD effectively improves task accuracy, it systematically traps…

Machine Learning · Computer Science 2026-04-21 Jiaxin Zhang , Xiangyu Peng , Qinglin Chen , Qinyuan Ye , Caiming Xiong , Chien-Sheng Wu

Large-scale vision-language mixture-of-experts (VL-MoE) models provide strong multimodal capability, but efficient deployment on memory-constrained platforms remains difficult. Existing MoE offloading systems are largely designed for…

Machine Learning · Computer Science 2026-05-08 Cheng Xu , Xiaofeng Hou , Jiacheng Liu , Chao Li

Vision-language models (VLMs) typically encode substantially more visual tokens than text tokens, resulting in significant token redundancy. Pruning uninformative visual tokens is therefore crucial for improving computational efficiency,…

Computer Vision and Pattern Recognition · Computer Science 2026-01-19 Kai Zhao , Wubang Yuan , Yuchen Lin , Liting Ruan , Xiaofeng Lu , Deng-Ping Fan , Ming-Ming Cheng , Dan Zeng

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…

Computation and Language · Computer Science 2026-01-21 Truong Nguyen , Phi Van Dat , Ngan Nguyen , Linh Ngo Van , Trung Le , Thanh Hong Nguyen

On-policy self-distillation has become a strong recipe for LLM reasoning, where a privileged teacher supervises the student's own rollouts while conditioning on the reference solution. A design choice shared by nearly all such methods,…

Artificial Intelligence · Computer Science 2026-05-28 Zihao Han , Tiangang Zhang , Huaibin Wang , Yilun Sun

Deep learning models may converge to suboptimal solutions despite strong validation accuracy, masking an optimization failure we term Trajectory Deviation. This is because as training proceeds, models can abandon high generalization states…

Machine Learning · Computer Science 2026-04-15 Eli Corn , Daphna Weinshall

Vision Large Language Models (VLLMs) incur high computational costs due to their reliance on hundreds of visual tokens to represent images. While token pruning offers a promising solution for accelerating inference, this paper, however,…

Computer Vision and Pattern Recognition · Computer Science 2026-03-10 Yahong Wang , Juncheng Wu , Zhangkai Ni , Longzhen Yang , Yihang Liu , Chengmei Yang , Ying Wen , Lianghua He , Xianfeng Tang , Hui Liu , Yuyin Zhou

Reinforcement learning has emerged as a powerful tool for improving diffusion-based text-to-image models, but existing methods are largely limited to single-task optimization. Extending RL to multiple tasks is challenging: joint…

Machine Learning · Computer Science 2026-05-15 Quanhao Li , Junqiu Yu , Kaixun Jiang , Yujie Wei , Zhen Xing , Pandeng Li , Ruihang Chu , Shiwei Zhang , Yu Liu , Zuxuan Wu

Knowledge distillation (KD) is widely used in audio tasks, such as speaker verification (SV), by transferring knowledge from a well-trained large model (the teacher) to a smaller, more compact model (the student) for efficiency and…

Audio and Speech Processing · Electrical Eng. & Systems 2024-09-17 Wenhao Yang , Jianguo Wei , Wenhuan Lu , Xugang Lu , Lei Li

Large-scale 3D vision-language models (VLMs) like LLaVA-3D offer strong spatial reasoning but are difficult to deploy due to high computational costs. We propose a knowledge distillation framework that transfers spatial reasoning from a 7B…

Computer Vision and Pattern Recognition · Computer Science 2026-05-12 Alaa Asfour , Christopher Indris , Leihan Chen , Tejas Vyas , Guanghui Wang

Traditional black-box distillation for Large Vision-Language Models (LVLMs) typically relies on a single teacher response per input, which often yields high-variance responses and format inconsistencies in multimodal or temporal scenarios.…

Computer Vision and Pattern Recognition · Computer Science 2026-03-13 Songlin Li , Xin Zhu , Zechao Guan , Peipeng Chen , Jian Yao