Author name cluster

Qiang Yan

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

2 author rows

ICML Conference 2025 Conference Paper

Cape: Context-Aware Prompt Perturbation Mechanism with Differential Privacy

Haoqi Wu
Wei Dai
Li Wang
Qiang Yan

Large Language Models (LLMs) have gained significant popularity due to their remarkable capabilities in text understanding and generation. However, despite their widespread deployment in inference services such as ChatGPT, concerns about the potential leakage of sensitive user data have arisen. Existing solutions primarily rely on privacy-enhancing technologies to mitigate such risks, facing the trade-off among efficiency, privacy, and utility. To narrow this gap, we propose Cape, a context-aware prompt perturbation mechanism based on differential privacy, to enable efficient inference with an improved privacy-utility trade-off. Concretely, we introduce a hybrid utility function that better captures the token similarity. Additionally, we propose a bucketized sampling mechanism to handle large sampling space, which might lead to long-tail phenomenons. Extensive experiments across multiple datasets, along with ablation studies, demonstrate that Cape achieves a better privacy-utility trade-off compared to prior state-of-the-art works.

Details

NeurIPS Conference 2025 Conference Paper

ObCLIP: Oblivious CLoud-Device Hybrid Image Generation with Privacy Preservation

Haoqi Wu
Wei Dai
Ming Xu
Wang Li
Qiang Yan

Diffusion Models have gained significant popularity due to their remarkable capabilities in image generation, albeit at the cost of intensive computation requirement. Meanwhile, despite their widespread deployment in inference services such as Midjourney, concerns about the potential leakage of sensitive information in uploaded user prompts have arisen. Existing solutions either fail to strike an effective balance between utility and efficiency, or lack rigorous privacy guarantees. To bridge this gap, we propose ObCLIP, a plug-and-play safeguard that enables oblivious cloud-device hybrid generation scheme. By oblivious, each input prompt is transformed into a set of semantically similar candidate prompts that differ only in sensitive attributes (e. g. , gender, ethnicity). The cloud server processes all candidate prompts without knowing which one is the real one, thus preventing any prompt leakage. To mitigate server cost, only a small portion of denoising steps is performed upon the large cloud model. The resulting intermediate latents are then transmitted back to the device, which selects the targeted latent and completes the remaining denoising using a small local model to obtain the final image. Additionally, we analyze and incorporate several cache-based accelerations that leverage temporal and batch redundancy, effectively reducing computation cost with minimal utility degradation. Extensive experiments across multiple datasets demonstrate that ObCLIP provides rigorous privacy and comparable utility to large cloud models with slightly increased server computation.

PDF Details

IJCAI Conference 2022 Conference Paper

MGAD: Learning Descriptional Representation Distilled from Distributional Semantics for Unseen Entities

Yuanzheng Wang
Xueqi Cheng
Yixing Fan
Xiaofei Zhu
Huasheng Liang
Qiang Yan
Jiafeng Guo

Entity representation plays a central role in building effective entity retrieval models. Recent works propose to learn entity representations based on entity-centric contexts, which achieve SOTA performances on many tasks. However, these methods lead to poor representations for unseen entities since they rely on a multitude of occurrences for each entity to enable accurate representations. To address this issue, we propose to learn enhanced descriptional representations for unseen entities by distilling knowledge from distributional semantics into descriptional embeddings. Specifically, we infer enhanced embeddings for unseen entities based on descriptions by aligning the descriptional embedding space to the distributional embedding space with different granularities, i. e. , element-level, batch-level and space-level alignment. Experimental results on four benchmark datasets show that our approach improves the performance over all baseline methods. In particular, our approach can achieve the effectiveness of the teacher model on almost all entities, and maintain such high performance on unseen entities.

PDF Details DOI