14 citations · 21 across the 21 of their papers we have counts for
25 papers
Beyond K/V Caches: Cross-Model State Handoff in Hybrid Linear-Attention LLMs
Jiacheng Cai, Tao Zhang, Siqi Liang +2
Production deployments often use different models to balance cost and quality through cascading, routing, and switching models during long conversations. Context accumulates across…
Sparse Attention for Video Generation Acceleration via Growing Sparsity and Reduced Search
Yun Dai, Yuyun Chen, Jiarui Wen +3
Diffusion Transformers (DiTs) dominate video generation but suffer from high inference cost due to the quadratic complexity of self-attention. Sparse attention mitigates this by sk…
PARK: Accurate Block Retrieval for Sparse Attention in Video Diffusion Transformers
Yun Dai, Jiarui Wen, Huiping Zhuang +2
Diffusion Transformers (DiTs) have become a dominant architecture for video generation, but their efficiency is limited by the quadratic complexity of full attention. Sparse attent…
EgoSafe: A First-Person Mobile-Captured Benchmark for Visual Safety Understanding
Yuyun Chen, Tianao Li, TianQuan Feng +4
Reliable visual safety understanding in real-world scenarios demands more than just object recognition; it requires causal reasoning under epistemic uncertainty. While Large Vision…
MixKVQ: Query-Aware Mixed-Precision KV Cache Quantization for Long-Context Reasoning
Tao Zhang, Ziqian Zeng, Hao Peng +2
Long Chain-of-Thought (CoT) reasoning has significantly advanced the capabilities of Large Language Models (LLMs), but this progress is accompanied by substantial memory and latenc…
ARGUS: Defending Against Multimodal Indirect Prompt Injection via Steering Instruction-Following Behavior
Weikai Lu, Ziqian Zeng, Kehua Zhang +5
Multimodal Large Language Models (MLLMs) are increasingly vulnerable to multimodal Indirect Prompt Injection (IPI) attacks, which embed malicious instructions in images, videos, or…
RCP-Merging: Merging Long Chain-of-Thought Models with Domain-Specific Models by Considering Reasoning Capability as Prior
Junyao Yang, Jianwei Wang, Huiping Zhuang +2
Large Language Models (LLMs) with long chain-of-thought (CoT) capability, termed Reasoning Models, demonstrate superior intricate problem-solving abilities through multi-step long…
SynAdapt: Learning Adaptive Reasoning in Large Language Models via Synthetic Continuous Chain-of-Thought
Jianwei Wang, Ziming Wu, Fuming Lai +2
While Chain-of-Thought (CoT) reasoning improves model performance, it incurs significant time costs due to the generation of discrete CoT tokens (DCoT). Continuous CoT (CCoT) offer…
Decompose, Plan in Parallel, and Merge: A Novel Paradigm for Large Language Models based Planning with Multiple Constraints
Zhengdong Lu, Weikai Lu, Yiling Tao +6
Despite significant advances in Large Language Models (LLMs), planning tasks still present challenges for LLM-based agents. Existing planning methods face two key limitations: heav…
L3A: Label-Augmented Analytic Adaptation for Multi-Label Class Incremental Learning
Xiang Zhang, Run He, Jiao Chen +5
Class-incremental learning (CIL) enables models to learn new classes continually without forgetting previously acquired knowledge. Multi-label CIL (MLCIL) extends CIL to a real-wor…
Semantic Shift Estimation via Dual-Projection and Classifier Reconstruction for Exemplar-Free Class-Incremental Learning
Run He, Di Fang, Yicheng Xu +5
Exemplar-Free Class-Incremental Learning (EFCIL) aims to sequentially learn from distinct categories without retaining exemplars but easily suffers from catastrophic forgetting of…
RewardDS: Privacy-Preserving Fine-Tuning for Large Language Models via Reward Driven Data Synthesis
Jianwei Wang, Chengming Shi, Junyao Yang +5
The success of large language models (LLMs) has attracted many individuals to fine-tune them for domain-specific tasks by uploading their data. However, in sensitive areas like hea…
SEA: Low-Resource Safety Alignment for Multimodal Large Language Models via Synthetic Embeddings
Weikai Lu, Hao Peng, Huiping Zhuang +2
Multimodal Large Language Models (MLLMs) have serious security vulnerabilities.While safety alignment using multimodal datasets consisting of text and data of additional modalities…
X-SGS: Safe and Generalizable Gaussian Splatting with X-dimensional Watermarks
Zihang Cheng, Wentao Bao, Huiping Zhuang +6
3D Gaussian Splatting (3DGS) has been widely used in 3D reconstruction and 3D generation. However, the rapid adoption of 3D Gaussian Splatting raises growing concerns about informa…
AIR: Analytic Imbalance Rectifier for Continual Learning
Di Fang, Yinan Zhu, Runze Fang +4
Continual learning (CL) agents incrementally learn from sequentially arriving data and adapt to the dynamic, ever-changing nature of real-world environments. However, many existing…
GenderAlign: An Alignment Dataset for Mitigating Gender Bias in Large Language Models
Tao Zhang, Ziqian Zeng, Yuxiang Xiao +4
Large Language Models (LLMs) are prone to generating content that exhibits gender biases, raising significant ethical concerns. Alignment, the process of fine-tuning LLMs to better…
PrivacyRestore: Privacy-Preserving Inference in Large Language Models via Privacy Removal and Restoration
Ziqian Zeng, Jianwei Wang, Junyao Yang +4
The widespread usage of online Large Language Models (LLMs) inference services has raised significant privacy concerns about the potential exposure of private information in user i…
Online Analytic Exemplar-Free Continual Learning with Large Models for Imbalanced Autonomous Driving Task
Huiping Zhuang, Di Fang, Kai Tong +4
In autonomous driving, even a meticulously trained model can encounter failures when facing unfamiliar scenarios. One of these scenarios can be formulated as an online continual le…
Eraser: Jailbreaking Defense in Large Language Models via Unlearning Harmful Knowledge
Weikai Lu, Ziqian Zeng, Jianwei Wang +5
Jailbreaking attacks can enable Large Language Models (LLMs) to bypass the safeguard and generate harmful content. Existing jailbreaking defense methods have failed to address the…
DS-AL: A Dual-Stream Analytic Learning for Exemplar-Free Class-Incremental Learning
Huiping Zhuang, Run He, Kai Tong +3
Class-incremental learning (CIL) under an exemplar-free constraint has presented a significant challenge. Existing methods adhering to this constraint are prone to catastrophic for…
On the use of Silver Standard Data for Zero-shot Classification Tasks in Information Extraction
Jianwei Wang, Tianyin Wang, Ziqian Zeng
The superior performance of supervised classification methods in the information extraction (IE) area heavily relies on a large amount of gold standard data. Recent zero-shot class…
F-OAL: Forward-only Online Analytic Learning with Fast Training and Low Memory Footprint in Class Incremental Learning
Huiping Zhuang, Yuchen Liu, Run He +5
Online Class Incremental Learning (OCIL) aims to train models incrementally, where data arrive in mini-batches, and previous data are not accessible. A major challenge in OCIL is C…
GACL: Exemplar-Free Generalized Analytic Continual Learning
Huiping Zhuang, Yizhu Chen, Di Fang +5
Class incremental learning (CIL) trains a network on sequential tasks with separated categories in each task but suffers from catastrophic forgetting, where models quickly lose pre…
Chimera: A Lossless Decoding Method for Accelerating Large Language Models Inference by Fusing all Tokens
Ziqian Zeng, Jiahong Yu, Qianshi Pang +4
Large language models (LLMs) have demonstrated remarkable capabilities across various tasks. However, their widespread application is hindered by the resource-intensive decoding pr…