68 citations · 501 across the 111 of their papers we have counts for
23 papers · 1 filter
From Hindsight to Foresight: Self-Encouraged Hindsight Distillation for Knowledge-based Visual Question Answering
Yu Zhao, Ying Zhang, Xuhui Sui +4
Knowledge-based Visual Question Answering (KBVQA) necessitates external knowledge incorporation beyond cross-modal understanding. Existing KBVQA methods either utilize implicit kno…
Sparse Model Inversion: Efficient Inversion of Vision Transformers for Data-Free Applications
Zixuan Hu, Yongxian Wei, Li Shen +4
Model inversion, which aims to reconstruct the original training data from pre-trained discriminative models, is especially useful when the original training data is unavailable du…
Adaptive Defense against Harmful Fine-Tuning for Large Language Models via Bayesian Data Scheduler
Zixuan Hu, Li Shen, Zhenyi Wang +2
Harmful fine-tuning poses critical safety risks to fine-tuning-as-a-service for large language models. Existing defense strategies preemptively build robustness via attack simulati…
Network Sparsity Unlocks the Scaling Potential of Deep Reinforcement Learning
Guozheng Ma, Lu Li, Zilin Wang +3
Effectively scaling up deep reinforcement learning models has proven notoriously difficult due to network pathologies during training, motivating various targeted interventions suc…
Hyper-modal Imputation Diffusion Embedding with Dual-Distillation for Federated Multimodal Knowledge Graph Completion
Ying Zhang, Yu Zhao, Xuhui Sui +5
With the increasing multimodal knowledge privatization requirements, multimodal knowledge graphs in different institutes are usually decentralized, lacking of effective collaborati…
Mastering Massive Multi-Task Reinforcement Learning via Mixture-of-Expert Decision Transformer
Yilun Kong, Guozheng Ma, Qi Zhao +4
Despite recent advancements in offline multi-task reinforcement learning (MTRL) have harnessed the powerful capabilities of the Transformer architecture, most approaches focus on a…
Decision Flow Policy Optimization
Jifeng Hu, Sili Huang, Siyuan Guo +6
In recent years, generative models have shown remarkable capabilities across diverse fields, including images, videos, language, and decision-making. By applying powerful generativ…
Multimodal Reasoning Agent for Zero-Shot Composed Image Retrieval
Rong-Cheng Tu, Wenhao Sun, Hanzhe You +4
Zero-Shot Composed Image Retrieval (ZS-CIR) aims to retrieve target images given a compositional query, consisting of a reference image and a modifying text-without relying on anno…
R1-ShareVL: Incentivizing Reasoning Capability of Multimodal Large Language Models via Share-GRPO
Huanjin Yao, Qixiang Yin, Jingyi Zhang +8
In this work, we aim to incentivize the reasoning ability of Multimodal Large Language Models (MLLMs) via reinforcement learning (RL) and develop an effective approach that mitigat…
Ada-R1: Hybrid-CoT via Bi-Level Adaptive Reasoning Optimization
Haotian Luo, Haiying He, Yibo Wang +6
Recently, long-thought reasoning models achieve strong performance on complex reasoning tasks, but often incur substantial inference overhead, making efficiency a critical concern.…
Cross-Domain Diffusion with Progressive Alignment for Efficient Adaptive Retrieval
Junyu Luo, Yusheng Zhao, Xiao Luo +5
Unsupervised efficient domain adaptive retrieval aims to transfer knowledge from a labeled source domain to an unlabeled target domain, while maintaining low storage cost and high…
Analytic Energy-Guided Policy Optimization for Offline Reinforcement Learning
Jifeng Hu, Sili Huang, Zhejian Yang +6
Conditional decision generation with diffusion models has shown powerful competitiveness in reinforcement learning (RL). Recent studies reveal the relation between energy-function-…
OptMerge: Unifying Multimodal LLM Capabilities and Modalities via Model Merging
Yongxian Wei, Runxi Cheng, Weike Jin +7
Foundation models update slowly due to resource-intensive training, whereas domain-specific models evolve rapidly between releases. Model merging seeks to combine multiple expert m…
R1-Compress: Long Chain-of-Thought Compression via Chunk Compression and Search
Yibo Wang, Haotian Luo, Huanjin Yao +8
Chain-of-Thought (CoT) reasoning enhances large language models (LLMs) by enabling step-by-step problem-solving, yet its extension to Long-CoT introduces substantial computational…
Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities
Zhiwei Hao, Jianyuan Guo, Li Shen +6
Large language models (LLMs) have achieved impressive performance across various domains. However, the substantial hardware resources required for their training present a signific…
Combatting Dimensional Collapse in LLM Pre-Training Data via Diversified File Selection
Ziqing Fan, Siyuan Du, Shengchao Hu +5
Selecting high-quality pre-training data for large language models (LLMs) is crucial for enhancing their overall performance under limited computation budget, improving both traini…
Neuron-level Balance between Stability and Plasticity in Deep Reinforcement Learning
Jiahua Lan, Sen Zhang, Haixia Pan +3
In contrast to the human ability to continuously acquire knowledge, agents struggle with the stability-plasticity dilemma in deep reinforcement learning (DRL), which refers to the…
SeWA: Selective Weight Average via Probabilistic Masking
Peng Wang, Shengchao Hu, Zerui Tao +5
Weight averaging has become a standard technique for enhancing model performance. However, methods such as Stochastic Weight Averaging (SWA) and Latest Weight Averaging (LAWA) ofte…
Near-Optimal Online Learning for Multi-Agent Submodular Coordination: Tight Approximation and Communication Efficiency
Qixin Zhang, Zongqi Wan, Yu Yang +2
Coordinating multiple agents to collaboratively maximize submodular functions in unpredictable environments is a critical task with numerous applications in machine learning, robot…
Safety Reasoning with Guidelines
Haoyu Wang, Zeyu Qin, Li Shen +3
Training safe LLMs remains a critical challenge. The most widely used method, Refusal Training (RT), struggles to generalize against various Out-of-Distribution (OOD) jailbreaking…
O1-Pruner: Length-Harmonizing Fine-Tuning for O1-Like Reasoning Pruning
Haotian Luo, Li Shen, Haiying He +6
Recently, long-thought reasoning LLMs, such as OpenAI's O1, adopt extended reasoning processes similar to how humans ponder over complex problems. This reasoning paradigm significa…
Merging Models on the Fly Without Retraining: A Sequential Approach to Scalable Continual Model Merging
Anke Tang, Enneng Yang, Li Shen +4
Deep model merging represents an emerging research direction that combines multiple fine-tuned models to harness their specialized capabilities across different tasks and domains.…
Panacea: Mitigating Harmful Fine-tuning for Large Language Models via Post-fine-tuning Perturbation
Yibo Wang, Tiansheng Huang, Li Shen +6
Harmful fine-tuning attack introduces significant security risks to the fine-tuning services. Main-stream defenses aim to vaccinate the model such that the later harmful fine-tunin…