no code implementations • ECCV 2020 • Wenyu Sun, Chen Tang, Weigui Li, Zhuqing Yuan, Huazhong Yang, Yongpan Liu
This paper proposes a deep video compression method to simultaneously encode multiple frames with Frame-Conv3D and differential modulation.
no code implementations • 4 Jun 2025 • Kun Zhao, Bohao Yang, Chen Tang, Siyuan Dai, Haoteng Tang, Chenghua Lin, Liang Zhan
To leverage their complementary strengths, we introduce SLIDE (Small and Large Integrated for Dialogue Evaluation), a method integrating SLMs and LLMs via adaptive weighting.
no code implementations • CVPR 2025 • Chen Tang, Xinzhu Ma, Encheng Su, Xiufeng Song, Xiaohong Liu, Wei-Hong Li, Lei Bai, Wanli Ouyang, Xiangyu Yue
Traditional spatiotemporal models generally rely on task-specific architectures, which limit their generalizability and scalability across diverse tasks due to domain-specific design requirements.
no code implementations • 19 Mar 2025 • Zhong Ji, Ci Liu, Jingren Liu, Chen Tang, Yanwei Pang, Xuelong Li
Central to this approach is the Optimal Transport Adapter (OTA), which employs a cross-modal attention mechanism to enrich textual representations and facilitate subsequent better information interaction.
1 code implementation • 20 Feb 2025 • Xun Liang, Jiawei Yang, Yezhaohui Wang, Chen Tang, Zifan Zheng, Simin Niu, Shichao Song, Hanyu Wang, Bo Tang, Feiyu Xiong, Keming Mao, Zhiyu Li
Large Language Models (LLMs) have demonstrated exceptional comprehension capabilities and a vast knowledge base, suggesting that LLMs can serve as efficient tools for automated survey generation.
no code implementations • 4 Feb 2025 • Qihao Lin, Chen Tang, Lan Zhang, Junyang Zhang, Xiangyang Li
In this paper, we derive the watermark embedding distribution based on the logits of LLMs and propose a formal inequality to segment the text optimally for watermark embedding.
no code implementations • 14 Jan 2025 • Chen Tang, Bo Lv, Zifan Zheng, Bohao Yang, Kun Zhao, Ning Liao, Xiaoxing Wang, Feiyu Xiong, Zhiyu Li, Nayu Liu, Jingchi Jiang
Additionally, this study explores a novel recurrent routing strategy that may inspire further advancements in enhancing the reasoning capabilities of language models.
no code implementations • 9 Jan 2025 • Mingzi Wang, Yuan Meng, Chen Tang, Weixiang Zhang, Yijian Qin, Yang Yao, Yingxin Li, Tongtong Feng, Xin Wang, Xun Guan, Zhi Wang, Wenwu Zhu
The co-design of neural network architectures, quantization precisions, and hardware accelerators offers a promising approach to achieving an optimal balance between performance and efficiency, particularly for model deployment on resource-constrained edge devices.
2 code implementations • CVPR 2025 • Weixiang Zhang, Shuzhao Xie, Chengwei Ren, Siyi Xie, Chen Tang, Shijia Ge, Mingzi Wang, Zhi Wang
We propose EVOlutionary Selector (EVOS), an efficient training paradigm for accelerating Implicit Neural Representation (INR).
no code implementations • 10 Dec 2024 • Bo Lv, Chen Tang, Yanan Zhang, Xin Liu, Yue Yu, Ping Luo
In this paper, we propose SpecFuse, a novel ensemble framework that outputs the fused result by iteratively producing the next segment through collaboration among LLMs.
no code implementations • 8 Dec 2024 • Shuzhao Xie, Jiahang Liu, Weixiang Zhang, Shijia Ge, Sicheng Pan, Chen Tang, Yunpeng Bai, Zhi Wang
Using the size estimator and MPQ, we develop a calibrated algorithm to identify optimal hyperparameters in just 10 minutes, achieving a 1. 69$\times$ efficiency increase with quality comparable to state-of-the-art methods.
no code implementations • 7 Dec 2024 • Jiacheng Jiang, Yuan Meng, Chen Tang, Han Yu, Qun Li, Zhi Wang, Wenwu Zhu
To address this limitation, we propose a novel Gradient-Adaptive Quantization-Aware Training (GAQAT) framework for DG.
no code implementations • 15 Sep 2024 • Shuzhao Xie, Weixiang Zhang, Chen Tang, Yunpeng Bai, Rongwei Lu, Shijia Ge, Zhi Wang
More specifically, we first replace rotation quaternions with Euler angles; then, we apply region adaptive hierarchical transform to key attributes to reduce entropy.
no code implementations • 12 Sep 2024 • Weixiang Zhang, Shuzhao Xie, Shijia Ge, Wei Yao, Chen Tang, Zhi Wang
Neural Radiance Field (NeRF) has achieved remarkable success in creating immersive media representations through its exceptional reconstruction capabilities.
no code implementations • 11 Aug 2024 • Nanyang Du, Chen Tang, Yuxiao Jiang, Yuan Meng, Zhi Wang
To address these limitations, we propose efficient unsupervised domain adaptation with ReTraining-Free Quantization (RTF-Q).
no code implementations • 7 Aug 2024 • Chen Tang, Ben Abbatematteo, Jiaheng Hu, Rohan Chandra, Roberto Martín-Martín, Peter Stone
Reinforcement learning (RL), particularly its combination with deep neural networks referred to as deep RL (DRL), has shown tremendous promise across a wide range of applications, suggesting its potential for enabling the development of sophisticated robotic behaviors.
1 code implementation • 6 Jul 2024 • Ye Li, Chen Tang, Yuan Meng, Jiajun Fan, Zenghao Chai, Xinzhu Ma, Zhi Wang, Wenwu Zhu
We introduce PRANCE, a Vision Transformer compression framework that jointly optimizes the activated channels and reduces tokens, based on the characteristics of inputs.
no code implementations • 1 Jul 2024 • Zhan Chen, Chen Tang, Lu Xiong
Additionally, to enhance the temporal consistency and causal relationships of the predictions, we propose a Time Series Memory framework to learn the conditional distribution models of the prediction outputs at future time steps from multivariate time series.
no code implementations • 28 Jun 2024 • Chen Tang, Bohao Yang, Kun Zhao, Bo Lv, Chenghao Xiao, Frank Guerin, Chenghua Lin
Named entity recognition (NER) stands as a fundamental and pivotal task within the realm of Natural Language Processing.
1 code implementation • CVPR 2025 • Lei Chen, Yuan Meng, Chen Tang, Xinzhu Ma, Jingyan Jiang, Xin Wang, Zhi Wang, Wenwu Zhu
Recent advancements in diffusion models, particularly the architectural transformation from UNet-based models to Diffusion Transformers (DiTs), significantly improve the quality and scalability of image and video generation.
1 code implementation • 25 Jun 2024 • Bohao Yang, Dong Liu, Chenghao Xiao, Kun Zhao, Chen Tang, Chao Li, Lin Yuan, Guang Yang, Lanxiao Huang, Chenghua Lin
Large Language Models (LLMs) demonstrate remarkable ability to comprehend instructions and generate human-like text, enabling sophisticated agent simulation beyond basic behavior replication.
no code implementations • 24 Jun 2024 • Yuxin Chen, Chen Tang, Chenran Li, Ran Tian, Wei Zhan, Peter Stone, Masayoshi Tomizuka
Instead of inferring the complete human behavior characteristics, MEReQ infers a residual reward function that captures the discrepancy between the human expert's and the prior policy's underlying reward functions.
1 code implementation • 15 Jun 2024 • Yijun Liu, Yuan Meng, Fang Wu, Shenhao Peng, Hang Yao, Chaoyu Guan, Chen Tang, Xinzhu Ma, Zhi Wang, Wenwu Zhu
Based on this benchmark, we conduct extensive experiments with two well-known LLMs (English and Chinese) and four quantization algorithms to investigate this topic in-depth, yielding several counter-intuitive and valuable findings, e. g., models quantized using a calibration set with the same distribution as the test data are not necessarily optimal.
1 code implementation • 7 Jun 2024 • Zenghao Chai, Chen Tang, Yongkang Wong, Mohan Kankanhalli
The creation of 4D avatars (i. e., animated 3D avatars) from text description typically uses text-to-image (T2I) diffusion models to synthesize 3D avatars in the canonical space and subsequently applies animation with target motions.
no code implementations • 3 Jun 2024 • Xuehui Yu, Yi Guan, Rujia Shen, Xin Li, Chen Tang, Jingchi Jiang
To tackle these issues, we introduce the Causal Prompting Reinforcement Learning (CPRL) framework, designed for highly suboptimal and resource-constrained online scenarios.
no code implementations • 30 May 2024 • Ke Yi, Yuhui Xu, Heng Chang, Chen Tang, Yuan Meng, Tong Zhang, Jia Li
Large Language Models (LLMs) have advanced rapidly but face significant memory demands.
1 code implementation • 24 May 2024 • Kun Zhao, Bohao Yang, Chen Tang, Chenghua Lin, Liang Zhan
Our approach introduces several techniques: (1) Contrastive learning to differentiate between robust and non-robust response embeddings; (2) A novel metric for semantic sensitivity that combines embedding cosine distances with similarity learned through neural networks, and (3) a strategy for incorporating the evaluation results from both the SLM and LLMs.
no code implementations • 15 Apr 2024 • Haojun Sun, Chen Tang, Zhi Wang, Yuan Meng, Jingyan Jiang, Xinzhu Ma, Wenwu Zhu
Diffusion models have emerged as preeminent contenders in the realm of generative models.
no code implementations • 8 Apr 2024 • Qun Li, Yuan Meng, Chen Tang, Jiacheng Jiang, Zhi Wang
Quantization is a promising technique for reducing the bit-width of deep models to improve their runtime performance and storage efficiency, and thus becomes a fundamental step for deployment.
1 code implementation • 1 Apr 2024 • Bohao Yang, Kun Zhao, Chen Tang, Dong Liu, Liang Zhan, Chenghua Lin
Trainable evaluation metrics, typically trained with true positive and randomly selected negative responses, tend to assign higher scores to responses that share greater content similarity with a given context.
no code implementations • 20 Mar 2024 • Tyler Loakman, Chen Tang, Chenghua Lin
Previous work in phonologically and phonetically grounded language generation has mainly focused on domains such as puns and poetry.
no code implementations • 22 Feb 2024 • Catherine Weaver, Chen Tang, Ce Hao, Kenta Kawamoto, Masayoshi Tomizuka, Wei Zhan
Thus, we propose BeTAIL: Behavior Transformer Adversarial Imitation Learning, which combines a Behavior Transformer (BeT) policy from human demonstrations with online AIL.
1 code implementation • CVPR 2024 • Chen Tang, Yuan Meng, Jiacheng Jiang, Shuzhao Xie, Rongwei Lu, Xinzhu Ma, Zhi Wang, Wenwu Zhu
Conversely, mixed-precision quantization (MPQ) is advocated to compress the model effectively by allocating heterogeneous bit-width for layers.
1 code implementation • 19 Nov 2023 • Chen Tang, Tyler Loakman, Chenghua Lin
These results underscore the effectiveness of our model in leveraging context and event features to improve the quality of generated narratives.
no code implementations • 14 Nov 2023 • Hongyang Jiang, Mengdi Gao, Zirong Liu, Chen Tang, Xiaoqing Zhang, Shuai Jiang, Wu Yuan, Jiang Liu
In this work, we propose a human-in-the-loop, label-free early DR diagnosis framework called GlanceSeg, based on SAM.
no code implementations • 13 Nov 2023 • Rongwei Lu, Yutong Jiang, Yinan Mao, Chen Tang, Bin Chen, Laizhong Cui, Zhi Wang
Recognizing the computational limitations of mobile devices, we propose the DAGC-A, which is computationally less demanding and enhances the robustness of compression in non-IID scenarios.
no code implementations • 3 Nov 2023 • Tommaso Benciolini, Chen Tang, Marion Leibold, Catherine Weaver, Masayoshi Tomizuka, Wei Zhan
In the exploration, a MPC collects diverse data by balancing the racing objectives and the exploration criterion; then the GP is re-trained.
no code implementations • 31 Oct 2023 • Chen Tang, Frank Guerin, Chenghua Lin
This paper presents a tool called ``ACL Anthology Helper''.
1 code implementation • 24 Oct 2023 • Tomas Goldsack, Zhihao Zhang, Chen Tang, Carolina Scarton, Chenghua Lin
Previous approaches for automatic lay summarisation are exclusively reliant on the source article that, given it is written for a technical audience (e. g., researchers), is unlikely to explicitly define all technical concepts or state all of the background information that is relevant for a lay audience.
1 code implementation • 24 Oct 2023 • Chen Tang, Shun Wang, Tomas Goldsack, Chenghua Lin
Abstracts derived from biomedical literature possess distinct domain-specific characteristics, including specialised writing styles and biomedical terminologies, which necessitate a deep understanding of the related literature.
no code implementations • 11 Oct 2023 • Yuxin Chen, Chen Tang, Ran Tian, Chenran Li, Jinning Li, Masayoshi Tomizuka, Wei Zhan
We observe that, generally, a more diverse set of co-play agents during training enhances the generalization performance of the ego agent; however, this improvement varies across distinct scenarios and environments.
1 code implementation • 22 Sep 2023 • Bohao Yang, Chen Tang, Kun Zhao, Chenghao Xiao, Chenghua Lin
Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of natural language processing tasks.
1 code implementation • 19 Sep 2023 • Bohao Yang, Chen Tang, Chenghua Lin
In this paper, We propose a novel framework that models dialogues between patients and healthcare professionals using AMR graphs, where the neural networks incorporate textual and graphical knowledge with a dual attention mechanism.
1 code implementation • 18 Sep 2023 • Yiheng Li, Seth Z. Zhao, Chenfeng Xu, Chen Tang, Chenran Li, Mingyu Ding, Masayoshi Tomizuka, Wei Zhan
Accumulating substantial volumes of real-world driving data proves pivotal in the realm of trajectory forecasting for autonomous driving.
no code implementations • 18 Sep 2023 • Jinning Li, Xinyi Liu, Banghua Zhu, Jiantao Jiao, Masayoshi Tomizuka, Chen Tang, Wei Zhan
GOLD distills an offline DT policy into a lightweight policy network through guided online safe RL training, which outperforms both the offline DT policy and online safe RL algorithms.
1 code implementation • 28 Jun 2023 • Chen Tang, Hongbo Zhang, Tyler Loakman, Chenghua Lin, Frank Guerin
Further analysis also shows that our representation learning framework can fill the semantic gap by coagulating representations of both text and graph knowledge.
no code implementations • NeurIPS 2023 • Chenran Li, Chen Tang, Haruki Nishimura, Jean Mercat, Masayoshi Tomizuka, Wei Zhan
Specifically, we formulate the customization problem as a Markov Decision Process (MDP) with a reward function that combines 1) the inherent reward of the demonstration; and 2) the add-on reward specified by the downstream task.
no code implementations • 14 Jun 2023 • Ce Hao, Catherine Weaver, Chen Tang, Kenta Kawamoto, Masayoshi Tomizuka, Wei Zhan
Our Skill-Critic algorithm optimizes both the low-level and high-level policies; these policies are initialized and regularized by the latent space learned from offline demonstrations to guide the parallel policy optimization.
1 code implementation • 6 Jun 2023 • Tyler Loakman, Chen Tang, Chenghua Lin
Previous work in phonetically-grounded language generation has mainly focused on domains such as lyrics and poetry.
no code implementations • 12 May 2023 • Kai Ouyang, Chen Tang, Wenhao Zheng, Xiangjin Xie, Xuanji Xiao, Jian Dong, Hai-Tao Zheng, Zhi Wang
To address this issue, we propose using knowledge soft integration to balance the utilization of multimodal features and the curse of knowledge problem it brings about.
1 code implementation • 10 May 2023 • Hongbo Zhang, Chen Tang, Tyler Loakman, Bohao Yang, Stefan Goetze, Chenghua Lin
Commonsense knowledge is crucial to many natural language processing tasks.
no code implementations • 25 Apr 2023 • Hongyang Jiang, Jingqi Huang, Chen Tang, Xiaoqing Zhang, Mengdi Gao, Jiang Liu
Concretely, the HITL CAD system was implemented on the multiple instance learning (MIL), where eye-tracking gaze maps were beneficial to cherry-pick diagnosis-related instances.
no code implementations • 3 Apr 2023 • Kai Ouyang, Wenhao Zheng, Chen Tang, Xuanji Xiao, Hai-Tao Zheng
To tackle this issue, we argue that a trade-off should be achieved between the introduction of large amounts of auxiliary information and the protection of valuable information related to CVR.
no code implementations • 24 Mar 2023 • Wei-Jer Chang, Chen Tang, Chenran Li, Yeping Hu, Masayoshi Tomizuka, Wei Zhan
To ensure that autonomous vehicles take safe and efficient maneuvers in different interactive traffic scenarios, we should be able to evaluate autonomous vehicles against reactive agents with different social characteristics in the simulation environment.
1 code implementation • ICCV 2023 • Chen Tang, Li Lyna Zhang, Huiqiang Jiang, Jiahang Xu, Ting Cao, Quanlu Zhang, Yuqing Yang, Zhi Wang, Mao Yang
However, prior supernet training methods that rely on uniform sampling suffer from the gradient conflict issue: the sampled subnets can have vastly different model sizes (e. g., 50M vs. 2G FLOPs), leading to different optimization directions and inferior performance.
no code implementations • 14 Feb 2023 • Chen Tang, Kai Ouyang, Zenghao Chai, Yunpeng Bai, Yuan Meng, Zhi Wang, Wenwu Zhu
This general and dataset-independent property makes us search for the MPQ policy over a rather small-scale proxy dataset and then the policy can be directly used to quantize the model trained on a large-scale dataset.
1 code implementation • 27 Oct 2022 • Chen Tang, Hongbo Zhang, Tyler Loakman, Chenghua Lin, Frank Guerin
In this paper, we propose a novel framework to improve medical dialogue generation by considering features centered on domain-specific terminology.
1 code implementation • 22 Oct 2022 • Chen Tang, Chenghua Lin, Henglin Huang, Frank Guerin, Zhihao Zhang
One of the key challenges of automatic story generation is how to generate a long narrative that can maintain fluency, relevance, and coherence.
1 code implementation • 19 Oct 2022 • Chen Tang, Zhihao Zhang, Tyler Loakman, Chenghua Lin, Frank Guerin
To improve the performance of long text generation, recent studies have leveraged automatically planned event structures (i. e. storylines) to guide story generation.
1 code implementation • 19 Oct 2022 • Henglin Huang, Chen Tang, Tyler Loakman, Frank Guerin, Chenghua Lin
In spite of the success of prior works with the application of pre-trained models, current neural models for Chinese stories still struggle to generate high-quality long text narratives.
1 code implementation • 21 Apr 2022 • Chenfeng Xu, Tian Li, Chen Tang, Lingfeng Sun, Kurt Keutzer, Masayoshi Tomizuka, Alireza Fathi, Wei Zhan
It is hard to replicate these approaches in trajectory forecasting due to the lack of adequate trajectory data (e. g., 34K samples in the nuScenes dataset).
no code implementations • 21 Apr 2022 • Chen Tang, Haoyu Zhai, Kai Ouyang, Zhi Wang, Yifei Zhu, Wenwu Zhu
We propose to feed different data samples with varying quantization schemes to achieve a data-dependent dynamic inference, at a fine-grained layer level.
no code implementations • 19 Apr 2022 • Chen Tang, Wei Zhan, Masayoshi Tomizuka
Moreover, to properly evaluate an IBP model with offline datasets, we propose a Shapley-value-based metric to verify if the prediction model satisfies the inherent temporal independence of an interventional distribution.
no code implementations • 28 Mar 2022 • Lingfeng Sun, Chen Tang, Yaru Niu, Enna Sachdeva, Chiho Choi, Teruhisa Misu, Masayoshi Tomizuka, Wei Zhan
To address these issues, we propose a novel approach to avoid KL vanishing and induce an interpretable interactive latent space with pseudo labels.
1 code implementation • 16 Mar 2022 • Chen Tang, Kai Ouyang, Zhi Wang, Yifei Zhu, YaoWei Wang, Wen Ji, Wenwu Zhu
For example, MPQ search on ResNet18 with our indicators takes only 0. 06 s, which improves time efficiency exponentially compared to iterative search methods.
no code implementations • 6 Mar 2022 • Chen Tang, Frank Guerin, Chenghua Lin
In recent years, considerable research has been dedicated to the application of neural models in the field of natural language generation (NLG).
no code implementations • NeurIPS 2021 • Chen Tang, Wei Zhan, Masayoshi Tomizuka
In this work, we argue that one of the typical formulations of VAEs in multi-agent modeling suffers from an issue we refer to as social posterior collapse, i. e., the model is prone to ignoring historical social context when predicting the future trajectory of an agent.
no code implementations • 9 Nov 2021 • Jinning Li, Chen Tang, Masayoshi Tomizuka, Wei Zhan
Reinforcement Learning (RL) has been shown effective in domains where the agent can learn policies by actively interacting with its operating environment.
no code implementations • 23 Feb 2021 • Chen Tang, Nishan Srishankar, Sujitha Martin, Masayoshi Tomizuka
Explainability is essential for autonomous vehicles and other robotics systems interacting with humans and other objects during operation.
no code implementations • 21 Oct 2020 • Chen Tang, Wenyu Sun, Zhuqing Yuan, Yongpan Liu
To accelerate deep CNN models, this paper proposes a novel spatially adaptive framework that can dynamically generate pixel-wise sparsity according to the input image.
no code implementations • 11 Nov 2019 • Chen Tang, Jianyu Chen, Masayoshi Tomizuka
Current methods for long-term trajectory prediction cannot guarantee the physical feasibility of predicted distribution.
2 code implementations • 26 Oct 2019 • Daniel Seita, David Chan, Roshan Rao, Chen Tang, Mandi Zhao, John Canny
Learning from demonstrations is a popular tool for accelerating and reducing the exploration requirements of reinforcement learning.
no code implementations • 7 Dec 2018 • Zhuo Xu, Chen Tang, Masayoshi Tomizuka
Although deep reinforcement learning (deep RL) methods have lots of strengths that are favorable if applied to autonomous driving, real deep RL applications in autonomous driving have been slowed down by the modeling gap between the source (training) domain and the target (deployment) domain.