2 citations · 8 across the 17 of their papers we have counts for
21 papers
Route-and-Generate: KL-Regularized Router Improvement for Reasoning Routing in LLMs
Tien-Phat Nguyen, Truong Nguyen, Thanh-Thien Le +4
Large language models often apply complex reasoning indiscriminately, leading to overthinking on simple tasks and underthinking on harder ones. We present Route-and-Generate (Rouge…
Graph Sensitivity of Cartesian Products with Matched Bridges
Zhen-Mu Hong, Zi-Yi Wu, Zheng-Jiang Xia
For a graph , let denote the minimum of the maximum degree of an induced subgraph with vertices, where is the independence number, and write $f(G)=f_1(G…
Rationale-Guided Policy Optimization: Learning to Reason with Adaptive Rationale Scaffolding
Hoang Phan, Minh Pham, Chau Pham +3
On-policy reinforcement learning has become a central paradigm for improving the reasoning abilities of large language models. However, its effectiveness is often limited by reward…
Sharpening Tax in Post-Training
Changdae Oh, Qi Zeng, Qi Qi +7
An emerging hypothesis about reinforcement learning (RL) post-training of large language models (LLMs) is that it merely sharpens existing behaviors of a base model, improving sing…
Beyond Reasoning Gains: Mitigating General-Capability Forgetting in Large Reasoning Models
Hoang Phan, Xianjun Yang, Yuanshun Yao +6
Reinforcement learning with verifiable rewards (RLVR) has delivered impressive gains in mathematical and multimodal reasoning and has become a standard post-training paradigm for c…
Think Twice, Generate Once: Safeguarding by Progressive Self-Reflection
Hoang Phan, Victor Li, Qi Lei
Large language models (LLMs) have revolutionized natural language processing with their ability to generate coherent and contextually relevant text. However, their deployment raise…
Toward a Holistic Approach to Continual Model Merging
Hoang Phan, Sungmin Cha, Tung Lam Tran +1
We present a holistic framework for Continual Model Merging (CMM) that intervenes at three critical stages: pre-merging, during merging, and post-merging-to address two fundamental…
Preserving Clusters in Prompt Learning for Unsupervised Domain Adaptation
Tung-Long Vuong, Hoang Phan, Vy Vo +4
Recent approaches leveraging multi-modal pre-trained models like CLIP for Unsupervised Domain Adaptation (UDA) have shown significant promise in bridging domain gaps and improving…
Personalized Large Vision-Language Models
Chau Pham, Hoang Phan, David Doermann +1
The personalization model has gained significant attention in image generation yet remains underexplored for large vision-language models (LVLMs). Beyond generic ones, with persona…
DiMSUM: Diffusion Mamba -- A Scalable and Unified Spatial-Frequency Method for Image Generation
Hao Phung, Quan Dao, Trung Dao +3
We introduce a novel state-space architecture for diffusion models, effectively harnessing spatial and frequency information to enhance the inductive bias towards local features in…
Leveraging Hierarchical Taxonomies in Prompt-based Continual Learning
Quyen Tran, Hoang Phan, Minh Le +6
Humans perceive the world as a series of sequential events, which can be hierarchically organized with different levels of abstraction based on conceptual knowledge. Drawing inspir…
Sketchy Moment Matching: Toward Fast and Provable Data Selection for Finetuning
Yijun Dong, Hoang Phan, Xiang Pan +1
We revisit data selection in a modern context of finetuning from a fundamental perspective. Extending the classical wisdom of variance minimization in low dimensions to high-dimens…
Enhancing Domain Adaptation through Prompt Gradient Alignment
Hoang Phan, Lam Tran, Quyen Tran +1
Prior Unsupervised Domain Adaptation (UDA) methods often aim to train a domain-invariant feature extractor, which may hinder the model from learning sufficiently discriminative fea…
Controllable Prompt Tuning For Balancing Group Distributional Robustness
Hoang Phan, Andrew Gordon Wilson, Qi Lei
Models trained on data composed of different groups or domains can suffer from severe performance degradation under distribution shifts. While recent methods have largely focused o…
KOPPA: Improving Prompt-based Continual Learning with Key-Query Orthogonal Projection and Prototype-based One-Versus-All
Quyen Tran, Hoang Phan, Lam Tran +4
Drawing inspiration from prompt tuning techniques applied to Large Language Models, recent methods based on pre-trained ViT networks have achieved remarkable results in the field o…
Sharpness & Shift-Aware Self-Supervised Learning
Ngoc N. Tran, Son Duong, Hoang Phan +3
Self-supervised learning aims to extract meaningful features from unlabeled data for further downstream tasks. In this paper, we consider classification as a downstream task in pha…
Flat Seeking Bayesian Neural Networks
Van-Anh Nguyen, Tung-Long Vuong, Hoang Phan +3
Bayesian Neural Networks (BNNs) provide a probabilistic interpretation for deep learning models by imposing a prior distribution over model parameters and inferring a posterior dis…
An Optimal Transport-driven Approach for Cultivating Latent Space in Online Incremental Learning
Quyen Tran, Hai Nguyen, Hoang Phan +6
In online incremental learning, data continuously arrives with substantial distributional shifts, creating a significant challenge because previous samples have limited replay valu…
Beyond Losses Reweighting: Empowering Multi-Task Learning via the Generalization Perspective
Hoang Phan, Lam Tran, Quyen Tran +6
Multi-task learning (MTL) trains deep neural networks to optimize several objectives simultaneously using a shared backbone, which leads to reduced computational costs, improved da…
Stochastic Multiple Target Sampling Gradient Descent
Hoang Phan, Ngoc Tran, Trung Le +3
Sampling from an unnormalized target distribution is an essential problem with many applications in probabilistic inference. Stein Variational Gradient Descent (SVGD) has been show…
Global-Local Regularization Via Distributional Robustness
Hoang Phan, Trung Le, Trung Phung +3
Despite superior performance in many situations, deep neural networks are often vulnerable to adversarial examples and distribution shifts, limiting model generalization ability in…