Weiwei Deng
Research Subjects
Research Interests
Weiwei Deng has a total of 72 co-authors. Here are the top co-authors by number of shared publications:
- Qi Zhang
Microsoft, Beijing, China
8 shared publications - Feng Sun
Microsoft, Beijing, China
5 shared publications - Hao Sun
Microsoft, Beijing, China
5 shared publications - Haizhen Huang
Microsoft Corporation
3 shared publications - Senzhang Wang
Central South University, Changsha, China
3 shared publications - Weihao Han
Microsoft, Beijing, China
3 shared publications
Latest Publications
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 4469-4473
HybridSparse: An End-to-End Hybrid Framework for Efficient Large-Scale Retrieval
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 4469-4473
Open AccessHybridSparse: An End-to-End Hybrid Framework for Efficient Large-Scale Retrieval
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 2985-2995
LettinGo: Explore User Profile Generation for Recommendation System
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 2985-2995
Open AccessLettinGo: Explore User Profile Generation for Recommendation System
- Open Access
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 384-393
Towards Distributionally Balanced End-to-End Generative Retrieval
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 384-393
Open AccessTowards Distributionally Balanced End-to-End Generative Retrieval
- Open Access
- 22 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 216-227
Unleash LLMs Potential for Sequential Recommendation by Coordinating Dual Dynamic Index Mechanism
- 22 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 216-227
Open AccessUnleash LLMs Potential for Sequential Recommendation by Coordinating Dual Dynamic Index Mechanism
- Open Access
- 29 November 2024
- Published by Qeios Ltd in Qeios Preprints
StreamAdapter: Efficient Test Time Adaptation from Contextual Streams
In-context learning (ICL) allows large language models (LLMs) to adapt to new tasks directly from the given demonstrations without requiring gradient updates. While recent advances have expanded context windows to accommodate more demonstrations, this approach increases inference costs without necessarily improving performance. To mitigate these issues, We propose StreamAdapter, a novel approach that directly updates model parameters from context at test time, eliminating the need for explicit in-context demonstrations. StreamAdapter employs context mapping and weight absorption mechanisms to dynamically transform ICL demonstrations into parameter updates with minimal additional parameters. By reducing reliance on numerous in-context examples, StreamAdapter significantly reduce inference costs and allows for efficient inference with constant time complexity, regardless of demonstration count. Extensive experiments across diverse tasks and model architectures demonstrate that StreamAdapter achieves comparable or superior adaptation capability to ICL while requiring significantly fewer demonstrations. The superior task adaptation and context encoding capabilities of StreamAdapter on both language understanding and generation tasks provides a new perspective for adapting LLMs at test time using context, allowing for more efficient adaptation across scenarios and more cost-effective inference.
- 29 November 2024
- Published by Qeios Ltd in Qeios Preprints
Open AccessStreamAdapter: Efficient Test Time Adaptation from Contextual Streams
In-context learning (ICL) allows large language models (LLMs) to adapt to new tasks directly from the given demonstrations without requiring gradient updates. While recent advances have expanded context windows to accommodate more demonstrations, this approach increases inference costs without necessarily improving performance. To mitigate these issues, We propose StreamAdapter, a novel approach that directly updates model parameters from context at test time, eliminating the need for explicit in-context demonstrations. StreamAdapter employs context mapping and weight absorption mechanisms to dynamically transform ICL demonstrations into parameter updates with minimal additional parameters. By reducing reliance on numerous in-context examples, StreamAdapter significantly reduce inference costs and allows for efficient inference with constant time complexity, regardless of demonstration count. Extensive experiments across diverse tasks and model architectures demonstrate that StreamAdapter achieves comparable or superior adaptation capability to ICL while requiring significantly fewer demonstrations. The superior task adaptation and context encoding capabilities of StreamAdapter on both language understanding and generation tasks provides a new perspective for adapting LLMs at test time using context, allowing for more efficient adaptation across scenarios and more cost-effective inference.
- Open Access
- 13 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 393-402
OneSparse: A Unified System for Multi-index Vector Search
Multi-index vector search has become the cornerstone for many applications, such as recommendation systems. Efficient search in such a multi-modal hybrid vector space is challenging since no single index design performs well for all kinds of vector data. Existing approaches to processing multi-index hybrid queries either suffer from algorithmic limitations or processing inefficiency. In this paper, we propose OneSparse, a unified multi-vector index query system that incorporates multiple posting-based vector indices, which enables highly efficient retrieval of multi-modal data-sets. OneSparse introduces a novel multi-index query engine design of inter-index intersection push-down. It also optimizes the vector posting format to expedite multi-index queries. Our experiments show OneSparse achieves more than 6x search performance improvement while maintaining comparable accuracy. OneSparse has already been integrated into Microsoft online web search and advertising systems with 5x+ latency gain for Bing web search and 2.0% Revenue Per Mille (RPM) gain for Bing sponsored search.
- 13 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 393-402
Open AccessOneSparse: A Unified System for Multi-index Vector Search
Multi-index vector search has become the cornerstone for many applications, such as recommendation systems. Efficient search in such a multi-modal hybrid vector space is challenging since no single index design performs well for all kinds of vector data. Existing approaches to processing multi-index hybrid queries either suffer from algorithmic limitations or processing inefficiency. In this paper, we propose OneSparse, a unified multi-vector index query system that incorporates multiple posting-based vector indices, which enables highly efficient retrieval of multi-modal data-sets. OneSparse introduces a novel multi-index query engine design of inter-index intersection push-down. It also optimizes the vector posting format to expedite multi-index queries. Our experiments show OneSparse achieves more than 6x search performance improvement while maintaining comparable accuracy. OneSparse has already been integrated into Microsoft online web search and advertising systems with 5x+ latency gain for Bing web search and 2.0% Revenue Per Mille (RPM) gain for Bing sponsored search.
- Open Access
- 4 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 1280-1290
Constraint-aware and Ranking-distilled Token Pruning for Efficient Transformer Inference
Deploying pre-trained transformer models like BERT on downstream tasks in resource-constrained scenarios is challenging due to their high inference cost, which grows rapidly with input sequence length. In this work, we propose a constraint-aware and ranking-distilled token pruning method ToP, which selectively removes unnecessary tokens as input sequence passes through layers, allowing the model to improve online inference speed while preserving accuracy. ToP overcomes the limitation of inaccurate token importance ranking in the conventional self-attention mechanism through a ranking-distilled token distillation technique, which distills effective token rankings from the final layer of unpruned models to early layers of pruned models. Then, ToP introduces a coarse-to-fine pruning approach that automatically selects the optimal subset of transformer layers and optimizes token pruning decisions within these layers through improved L0 regularization. Extensive experiments on GLUE benchmark and SQuAD tasks demonstrate that ToP outperforms state-of-the-art token pruning and model compression methods with improved accuracy and speedups. ToP reduces the average FLOPs of BERT by 8.1X while achieving competitive accuracy on GLUE, and provides a real latency speedup of up to 7.4X on an Intel CPU. Code is available at https://github.com/microsoft/Moonlit/tree/main/ToP
- 4 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 1280-1290
Open AccessConstraint-aware and Ranking-distilled Token Pruning for Efficient Transformer Inference
Deploying pre-trained transformer models like BERT on downstream tasks in resource-constrained scenarios is challenging due to their high inference cost, which grows rapidly with input sequence length. In this work, we propose a constraint-aware and ranking-distilled token pruning method ToP, which selectively removes unnecessary tokens as input sequence passes through layers, allowing the model to improve online inference speed while preserving accuracy. ToP overcomes the limitation of inaccurate token importance ranking in the conventional self-attention mechanism through a ranking-distilled token distillation technique, which distills effective token rankings from the final layer of unpruned models to early layers of pruned models. Then, ToP introduces a coarse-to-fine pruning approach that automatically selects the optimal subset of transformer layers and optimizes token pruning decisions within these layers through improved L0 regularization. Extensive experiments on GLUE benchmark and SQuAD tasks demonstrate that ToP outperforms state-of-the-art token pruning and model compression methods with improved accuracy and speedups. ToP reduces the average FLOPs of BERT by 8.1X while achieving competitive accuracy on GLUE, and provides a real latency speedup of up to 7.4X on an Intel CPU. Code is available at https://github.com/microsoft/Moonlit/tree/main/ToP
- Open Access
- 4 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 5039-5050
BERT4CTR: An Efficient Framework to Combine Pre-trained Language Model with Non-textual Features for CTR Prediction
Although deep pre-trained language models have shown promising benefit in a large set of industrial scenarios, including Click-Through-Rate (CTR) prediction, how to integrate pre-trained language models that handle only textual signals into a prediction pipeline with non-textual features is challenging. Up to now, two directions have been explored to integrate multi-modal inputs in fine-tuning of pre-trained language models. One consists of fusing the outcome of language models and non-textual features through an aggregation layer, resulting into ensemble framework, where the cross-information between textual and non-textual inputs are learned only in the aggregation layer. The second one consists of splitting and transforming non-textual features into fine-grained tokens that are fed, along with textual tokens, directly into the transformer layers of language models. However, by adding additional tokens, this approach increases the complexity of the learning and inference. We propose in this paper, a novel framework, BERT4CTR, that addresses these limitations. The new framework leverages Uni-Attention mechanism to benefit from the interactions between non-textual and textual features, while maintaining low training and inference time-costs, through a dimensionality reduction. We demonstrate through comprehensive experiments on both public and commercial data that BERT4CTR outperforms significantly the state-of-the-art approaches to handle multi-modal inputs and is applicable to CTR prediction. In comparison with ensemble framework, BERT4CTR brings more than 0.4% AUC gain on both tested data sets with only 7% increase on latency.
- 4 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 5039-5050
Open AccessBERT4CTR: An Efficient Framework to Combine Pre-trained Language Model with Non-textual Features for CTR Prediction
Although deep pre-trained language models have shown promising benefit in a large set of industrial scenarios, including Click-Through-Rate (CTR) prediction, how to integrate pre-trained language models that handle only textual signals into a prediction pipeline with non-textual features is challenging. Up to now, two directions have been explored to integrate multi-modal inputs in fine-tuning of pre-trained language models. One consists of fusing the outcome of language models and non-textual features through an aggregation layer, resulting into ensemble framework, where the cross-information between textual and non-textual inputs are learned only in the aggregation layer. The second one consists of splitting and transforming non-textual features into fine-grained tokens that are fed, along with textual tokens, directly into the transformer layers of language models. However, by adding additional tokens, this approach increases the complexity of the learning and inference. We propose in this paper, a novel framework, BERT4CTR, that addresses these limitations. The new framework leverages Uni-Attention mechanism to benefit from the interactions between non-textual and textual features, while maintaining low training and inference time-costs, through a dimensionality reduction. We demonstrate through comprehensive experiments on both public and commercial data that BERT4CTR outperforms significantly the state-of-the-art approaches to handle multi-modal inputs and is applicable to CTR prediction. In comparison with ensemble framework, BERT4CTR brings more than 0.4% AUC gain on both tested data sets with only 7% increase on latency.
- Open Access
- 4 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 4924-4936
PASS: Personalized Advertiser-aware Sponsored Search
The nucleus of online sponsored search systems lies in measuring the relevance between the search intents of users and the advertising purposes of advertisers. Existing conventional doublet-based (query-keyword) relevance models solely rely on short queries and keywords to uncover such intents, which ignore the diverse and personalized preferences of participants (i.e., users and advertisers), resulting in undesirable advertising performance. In this paper, we investigate the novel problem of Personalized A dvertiser-aware Sponsored Search (PASS). Our motivation lies in incorporating the portraits of users and advertisers into relevance models to facilitate the modeling of intrinsic search intents and advertising purposes, leading to a quadruple-based (i.e., user-query-keyword-advertiser) task. Various types of historical behaviors are explored in the format of hypergraphs to provide abundant signals on identifying the preferences of participants. A novel heterogeneous textual hypergraph transformer is further proposed to deeply fuse the textual semantics and the high-order hypergraph topology. Our proposal is extensively evaluated over real industry datasets, and experimental results demonstrate its superiority.
- 4 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 4924-4936
Open AccessPASS: Personalized Advertiser-aware Sponsored Search
The nucleus of online sponsored search systems lies in measuring the relevance between the search intents of users and the advertising purposes of advertisers. Existing conventional doublet-based (query-keyword) relevance models solely rely on short queries and keywords to uncover such intents, which ignore the diverse and personalized preferences of participants (i.e., users and advertisers), resulting in undesirable advertising performance. In this paper, we investigate the novel problem of Personalized A dvertiser-aware Sponsored Search (PASS). Our motivation lies in incorporating the portraits of users and advertisers into relevance models to facilitate the modeling of intrinsic search intents and advertising purposes, leading to a quadruple-based (i.e., user-query-keyword-advertiser) task. Various types of historical behaviors are explored in the format of hypergraphs to provide abundant signals on identifying the preferences of participants. A novel heterogeneous textual hypergraph transformer is further proposed to deeply fuse the textual semantics and the high-order hypergraph topology. Our proposal is extensively evaluated over real industry datasets, and experimental results demonstrate its superiority.
- Open Access
- 18 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 2189-2193
Multi-Grained Topological Pre-Training of Language Models in Sponsored Search
Relevance models measure the semantic closeness between queries and the candidate ads, widely recognized as the nucleus of sponsored search systems. Conventional relevance models solely rely on the textual data within the queries and ads, whose performance is hindered by the scarce semantic information in these short texts. Recently, user behavior graphs have been incorporated to provide complementary information beyond pure textual semantics.Despite the promising performance, behavior-enhanced models suffer from exhausting resource costs due to the extra computations introduced by explicit topological aggregations. In this paper, we propose a novel Multi-Grained Topological Pre-Training paradigm, MGTLM, to teach language models to understand multi-grained topological information in behavior graphs, which contributes to eliminating explicit graph aggregations and avoiding information loss. Extensive experimental results over online and offline settings demonstrate the superiority of our proposal.
- 18 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 2189-2193
Open AccessMulti-Grained Topological Pre-Training of Language Models in Sponsored Search
Relevance models measure the semantic closeness between queries and the candidate ads, widely recognized as the nucleus of sponsored search systems. Conventional relevance models solely rely on the textual data within the queries and ads, whose performance is hindered by the scarce semantic information in these short texts. Recently, user behavior graphs have been incorporated to provide complementary information beyond pure textual semantics.Despite the promising performance, behavior-enhanced models suffer from exhausting resource costs due to the extra computations introduced by explicit topological aggregations. In this paper, we propose a novel Multi-Grained Topological Pre-Training paradigm, MGTLM, to teach language models to understand multi-grained topological information in behavior graphs, which contributes to eliminating explicit graph aggregations and avoiding information loss. Extensive experimental results over online and offline settings demonstrate the superiority of our proposal.