Liping Jing
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Liping Jing has a total of 147 co-authors. Here are the top co-authors by number of shared publications:
Jian Yu
Beijing Key Lab of Traffic Data Analysis and Mining, Beijing Jiaotong University, Beijing, 100044, China
13 shared publications- Jian Yu
Beijing Key Lab of Traffic Data Analysis and Mining Beijing Jiaotong University Beijing China
12 shared publications - Huafeng Liu
Beijing Jiaotong University, Beijing, China
11 shared publications - Michael KwokPo Ng
Department of Mathematics, Hong Kong Baptist University, Kowloon Tong, Hong Kong
11 shared publications - Mingyang Song
Beijing Key Lab of Traffic Data Analysis and Mining Beijing Jiaotong University Beijing China
8 shared publications - Yilin Lyu
School of Computer Science and Technology Beijing Key Laboratory of Traffic Data Mining and Embodied Intelligence Beijing Jiaotong University State Key Laboratory of Advanced Rail Autonomous Operation
8 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 3463-3472
Learning Universal Semantic Simplex for Jailbreaking Large Vision-Language Models
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 3463-3472
Open AccessLearning Universal Semantic Simplex for Jailbreaking Large Vision-Language Models
- 17 October 2026
- Published by Springer Nature
- p. 251-260
Typographic Attack Detection in Large Vision-Language Models via Attention-Level Evidential Conflict
- 17 October 2026
- Published by Springer Nature
- p. 251-260
Typographic Attack Detection in Large Vision-Language Models via Attention-Level Evidential Conflict
- Open Access
- 21 July 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
P4Q: Learning to Prompt for Quantization in Low-Bit CLIP
Large-scale pre-trained Vision-Language Models (VLMs) have demonstrated remarkable performance across various visual and multimodal tasks. However, deploying these models on downstream application platforms remains challenging due to computational demands and domain gaps. Quantization offers a promising solution by significantly reducing these costs, making VLMs more feasible for deployment in such environments. There are two prevailing paradigms: Quantization-Aware Training (QAT), which preserves model performance but incurs substantial training costs; and Post-Training Quantization (PTQ), which offers greater efficiency but introduces multimodal gaps and leads to performance degradation on downstream tasks. To reduce computational costs and bridge domain gaps, we propose the “Prompt for Quantization” (P4Q) by integrating PTQ with Parameter-Efficient Fine-Tuning (PEFT) techniques. P4Q compresses model parameters and activations via PTQ, introducing learnable prompts and low-bit adapters to enhance performance on downstream tasks. The learnable prompts embed downstream knowledge to mitigate domain gaps, while the low-bit adapters realign the distributions of image and text features, thereby mitigating multimodal gaps. We also introduce a distillation loss based on cosine similarity predictions to distill the quantized model using a full-precision teacher model. Extensive experiments on thirteen datasets demonstrate that P4Q significantly enhances the performance of low-bit CLIP while reducing deployment costs. For instance, an 8-bit P4Q compressed CLIP-ViT/B-32 achieves 66.94% Top-1 accuracy on ImageNet, surpassing the prompt fine-tuned full-precision counterpart by 2.24% while reducing model size by 4 \(\times\) . The source code is publicly available at https://github.com/HuixinSun/P4Q_official .- 21 July 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
Open AccessP4Q: Learning to Prompt for Quantization in Low-Bit CLIP
Large-scale pre-trained Vision-Language Models (VLMs) have demonstrated remarkable performance across various visual and multimodal tasks. However, deploying these models on downstream application platforms remains challenging due to computational demands and domain gaps. Quantization offers a promising solution by significantly reducing these costs, making VLMs more feasible for deployment in such environments. There are two prevailing paradigms: Quantization-Aware Training (QAT), which preserves model performance but incurs substantial training costs; and Post-Training Quantization (PTQ), which offers greater efficiency but introduces multimodal gaps and leads to performance degradation on downstream tasks. To reduce computational costs and bridge domain gaps, we propose the “Prompt for Quantization” (P4Q) by integrating PTQ with Parameter-Efficient Fine-Tuning (PEFT) techniques. P4Q compresses model parameters and activations via PTQ, introducing learnable prompts and low-bit adapters to enhance performance on downstream tasks. The learnable prompts embed downstream knowledge to mitigate domain gaps, while the low-bit adapters realign the distributions of image and text features, thereby mitigating multimodal gaps. We also introduce a distillation loss based on cosine similarity predictions to distill the quantized model using a full-precision teacher model. Extensive experiments on thirteen datasets demonstrate that P4Q significantly enhances the performance of low-bit CLIP while reducing deployment costs. For instance, an 8-bit P4Q compressed CLIP-ViT/B-32 achieves 66.94% Top-1 accuracy on ImageNet, surpassing the prompt fine-tuned full-precision counterpart by 2.24% while reducing model size by 4 \(\times\) . The source code is publicly available at https://github.com/HuixinSun/P4Q_official . - Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 11062-11070
Maximum Redundancy Pruning: A Principle-Driven Layerwise Sparsity Allocation for LLMs
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 11062-11070
Open AccessMaximum Redundancy Pruning: A Principle-Driven Layerwise Sparsity Allocation for LLMs
- 23 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48  (1) , 675-692
End-to-End Autonomous Driving Without Costly Modularization and 3D Manual Annotation
We propose UAD, an end-to-end framework with Unsupervised pretext task for vision-based Autonomous Driving, achieving the best open-loop evaluation performance in nuScenes, meanwhile showing robust closed-loop driving quality in CARLA. Our motivation stems from the observation that current end-to-end autonomous driving (E2EAD) models still mimic the modular architecture in typical driving stacks, with carefully designed supervised perception and prediction subtasks to provide environment information for oriented planning. Although achieving groundbreaking progress, such design has certain drawbacks: 1) preceding subtasks require massive high-quality 3D annotations as supervision, posing a significant impediment to scaling the training data; and 2) each submodule entails substantial computation overhead in both training and inference. To this end, we propose UAD, an E2EAD framework with an unsupervised1 proxy to address all these issues. Firstly, we design a novel Angular Perception Pretext to eliminate the annotation requirement. The pretext perceives the driving scene by predicting the angular-wise spatial objectness and temporal dynamics, without manual annotation. Secondly, a self-supervised training strategy, which learns the consistency of the predicted trajectories under different augment views, is proposed to enhance the planning robustness in steering scenarios. Our UAD achieves 38.7% relative improvements over UniAD on the average collision rate of nuScenes open-loop evaluation and obtains the route completion score of 98.5% in closed-loop evaluation of CARLA’s Town05 Long benchmark, which outperforms the recent work VADv2. Moreover, the proposed method consumes only 44.3% training resources of UniAD and runs$3.4\times$ faster in inference when employing the same backbone network. Our innovative design not only for the first time demonstrates unarguable performance advantages over supervised counterparts, but also enjoys unprecedented efficiency in data, training, and inference.- 23 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48  (1) , 675-692
End-to-End Autonomous Driving Without Costly Modularization and 3D Manual Annotation
We propose UAD, an end-to-end framework with Unsupervised pretext task for vision-based Autonomous Driving, achieving the best open-loop evaluation performance in nuScenes, meanwhile showing robust closed-loop driving quality in CARLA. Our motivation stems from the observation that current end-to-end autonomous driving (E2EAD) models still mimic the modular architecture in typical driving stacks, with carefully designed supervised perception and prediction subtasks to provide environment information for oriented planning. Although achieving groundbreaking progress, such design has certain drawbacks: 1) preceding subtasks require massive high-quality 3D annotations as supervision, posing a significant impediment to scaling the training data; and 2) each submodule entails substantial computation overhead in both training and inference. To this end, we propose UAD, an E2EAD framework with an unsupervised1 proxy to address all these issues. Firstly, we design a novel Angular Perception Pretext to eliminate the annotation requirement. The pretext perceives the driving scene by predicting the angular-wise spatial objectness and temporal dynamics, without manual annotation. Secondly, a self-supervised training strategy, which learns the consistency of the predicted trajectories under different augment views, is proposed to enhance the planning robustness in steering scenarios. Our UAD achieves 38.7% relative improvements over UniAD on the average collision rate of nuScenes open-loop evaluation and obtains the route completion score of 98.5% in closed-loop evaluation of CARLA’s Town05 Long benchmark, which outperforms the recent work VADv2. Moreover, the proposed method consumes only 44.3% training resources of UniAD and runs$3.4\times$ faster in inference when employing the same backbone network. Our innovative design not only for the first time demonstrates unarguable performance advantages over supervised counterparts, but also enjoys unprecedented efficiency in data, training, and inference.- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 615-625
Efficient Sphere-Effect Based Information Diffusion Prediction on Large-scale Social Networks
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 615-625
Open AccessEfficient Sphere-Effect Based Information Diffusion Prediction on Large-scale Social Networks
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 1799-1810
Learning OOD Robust Neural Operator with Risk-Averse Stochastic Optimization
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 1799-1810
Open AccessLearning OOD Robust Neural Operator with Risk-Averse Stochastic Optimization
- Open Access
Continual Gradient Low-Rank Projection Fine-Tuning for LLMs
Continual fine-tuning of Large Language Models (LLMs) is hampered by the trade-off between efficiency and expressiveness. Low-Rank Adaptation (LoRA) offers efficiency but constrains the model's ability to learn new tasks and transfer knowledge due to its low-rank nature and reliance on explicit parameter constraints. We propose GORP (Gradient LOw Rank Projection) for Continual Learning, a novel training strategy that overcomes these limitations by synergistically combining full and low-rank parameters and jointly updating within a unified low-rank gradient subspace. GORP expands the optimization space while preserving efficiency and mitigating catastrophic forgetting. Extensive experiments on continual learning benchmarks demonstrate GORP's superior performance compared to existing state-of-the-art approaches. Code is available at https://github.com/Wcxwcxw/GORP.
Open AccessContinual Gradient Low-Rank Projection Fine-Tuning for LLMs
Continual fine-tuning of Large Language Models (LLMs) is hampered by the trade-off between efficiency and expressiveness. Low-Rank Adaptation (LoRA) offers efficiency but constrains the model's ability to learn new tasks and transfer knowledge due to its low-rank nature and reliance on explicit parameter constraints. We propose GORP (Gradient LOw Rank Projection) for Continual Learning, a novel training strategy that overcomes these limitations by synergistically combining full and low-rank parameters and jointly updating within a unified low-rank gradient subspace. GORP expands the optimization space while preserving efficiency and mitigating catastrophic forgetting. Extensive experiments on continual learning benchmarks demonstrate GORP's superior performance compared to existing state-of-the-art approaches. Code is available at https://github.com/Wcxwcxw/GORP.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 14815-14829
Continual Gradient Low-Rank Projection Fine-Tuning for LLMs
Chenxu Wang, Yilin Lyu, Zicheng Sun, Liping Jing. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 14815-14829
Open AccessContinual Gradient Low-Rank Projection Fine-Tuning for LLMs
Chenxu Wang, Yilin Lyu, Zicheng Sun, Liping Jing. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.
- Open Access
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 8972-8981
Align2Concept: Language Guided Interpretable Image Recognition by Visual Prototype and Textual Concept Alignment
Most works of interpretable neural networks strive for learning the semantics concepts merely from single modal information such as images. However, humans usually learn semantic concepts from multiple modalities and the semantics is encoded by the brain from fused multi-modal information. Inspired by cognitive science and vision-language learning, we propose a Prototype-Concept Alignment Network (ProCoNet) for learning visual prototypes under the guidance of textual concepts. In the ProCoNet, we have designed a visual encoder to decompose the input image into regional features of prototypes, while also developing a prompt generation strategy that incorporates in-context learning to prompt large language models to generate textual concepts. To align visual prototypes with textual concepts, we leverage the multimodal space provided by the pre-trained CLIP as a bridge. Specifically, the regional features from the vision space and the cropped regions of prototypes encoded by CLIP reside on different but semantically highly correlated manifolds, i.e. follow a multi-manifold distribution. We transform the multi-manifold distribution alignment problem into optimizing the projection matrix by Cayley transform on the Stiefel manifold. Through the learned projection matrix, visual prototypes can be projected into the multimodal space to align with semantically similar textual concept features encoded by CLIP. We conducted two case studies on the CUB-200-2011 and Oxford Flower dataset. Our experiments show that the ProCoNet provides higher accuracy and better interpretability compared to the single-modality interpretable model. Furthermore, ProCoNet offers a level of interpretability not previously available in other interpretable methods.
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 8972-8981
Open AccessAlign2Concept: Language Guided Interpretable Image Recognition by Visual Prototype and Textual Concept Alignment
Most works of interpretable neural networks strive for learning the semantics concepts merely from single modal information such as images. However, humans usually learn semantic concepts from multiple modalities and the semantics is encoded by the brain from fused multi-modal information. Inspired by cognitive science and vision-language learning, we propose a Prototype-Concept Alignment Network (ProCoNet) for learning visual prototypes under the guidance of textual concepts. In the ProCoNet, we have designed a visual encoder to decompose the input image into regional features of prototypes, while also developing a prompt generation strategy that incorporates in-context learning to prompt large language models to generate textual concepts. To align visual prototypes with textual concepts, we leverage the multimodal space provided by the pre-trained CLIP as a bridge. Specifically, the regional features from the vision space and the cropped regions of prototypes encoded by CLIP reside on different but semantically highly correlated manifolds, i.e. follow a multi-manifold distribution. We transform the multi-manifold distribution alignment problem into optimizing the projection matrix by Cayley transform on the Stiefel manifold. Through the learned projection matrix, visual prototypes can be projected into the multimodal space to align with semantically similar textual concept features encoded by CLIP. We conducted two case studies on the CUB-200-2011 and Oxford Flower dataset. Our experiments show that the ProCoNet provides higher accuracy and better interpretability compared to the single-modality interpretable model. Furthermore, ProCoNet offers a level of interpretability not previously available in other interpretable methods.