Yozen Liu
Yozen Liu has a total of 50 co-authors. Here are the top co-authors by number of shared publications:
- Neil Shah
Snap Inc., Bellevue, WA, USA
8 shared publications - Tong Zhao
Snap Inc., Bellevue, WA, USA
7 shared publications - Shubham Vij
Snap Inc., Bellevue, WA, USA
4 shared publications - Jun Yu
Snap Inc., Bellevue, WA, USA
3 shared publications - Maarten W. Bos
Snap Inc., United States
3 shared publications - Ron Dotsch
Snap Inc., Santa Monica, CA, USA
3 shared publications
Latest Publications
- Open Access
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 4414-4418
Beyond Unimodal Perspectives: Generative Retrieval with Multimodal Semantics
Generative retrieval (GR) has revolutionized recommendation systems by integrating the content of the items into semantic identifiers. However, existing frameworks predominantly isolate modalities (e.g., relying solely on text), overlooking the inherently multimodal nature of real-world items. This work addresses the underexplored challenge of Multimodal Generative Retrieval (MGR). Through a systematic analysis of Early and Late Fusion strategies, we reveal that naive integration fails due to two critical limitations: modality sensitivity, where one modality dominates the representation, and modality correspondence, where the model fails to align distinct semantic IDs across modalities. To overcome these challenges, we introduce MGR-LF++, an enhanced late fusion framework. MGR-LF++ employs contrastive modality alignment to enforce cross-modal consistency and incorporates special tokens to preserve autoregressive integrity. Extensive experiments demonstrate that MGR-LF++ achieves performance improvements of over 20% compared to unimodal and naive multimodal alternatives.
- 19 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 4414-4418
Open AccessBeyond Unimodal Perspectives: Generative Retrieval with Multimodal Semantics
Generative retrieval (GR) has revolutionized recommendation systems by integrating the content of the items into semantic identifiers. However, existing frameworks predominantly isolate modalities (e.g., relying solely on text), overlooking the inherently multimodal nature of real-world items. This work addresses the underexplored challenge of Multimodal Generative Retrieval (MGR). Through a systematic analysis of Early and Late Fusion strategies, we reveal that naive integration fails due to two critical limitations: modality sensitivity, where one modality dominates the representation, and modality correspondence, where the model fails to align distinct semantic IDs across modalities. To overcome these challenges, we introduce MGR-LF++, an enhanced late fusion framework. MGR-LF++ employs contrastive modality alignment to enforce cross-modal consistency and incorporates special tokens to preserve autoregressive integrity. Extensive experiments demonstrate that MGR-LF++ achieves performance improvements of over 20% compared to unimodal and naive multimodal alternatives.
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 5225-5236
GiGL: Large-Scale Graph Neural Networks at Snapchat
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 5225-5236
Open AccessGiGL: Large-Scale Graph Neural Networks at Snapchat
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 6241-6242
Training Industry-scale GNNs with GiGL
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 6241-6242
Open AccessTraining Industry-scale GNNs with GiGL
- Open Access
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 2521-2530
Improving Out-of-Vocabulary Hashing in Recommendation Systems
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 2521-2530
Open AccessImproving Out-of-Vocabulary Hashing in Recommendation Systems
- Open Access
- 22 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 357-369
GraphHash: Graph Clustering Enables Parameter Efficiency in Recommender Systems
- 22 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 357-369
Open AccessGraphHash: Graph Clustering Enables Parameter Efficiency in Recommender Systems
- Open Access
- 10 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 2431-2436
General-Purpose User Modeling with Behavioral Logs: A Snapchat Case Study
Learning general-purpose user representations based on user behavioral logs is an increasingly popular user modeling approach. It benefits from easily available, privacy-friendly yet expressive data, and does not require extensive re-tuning of the upstream user model for different downstream tasks. While this approach has shown promise in search engines and e-commerce applications, its fit for instant messaging platforms, a cornerstone of modern digital communication, remains largely uncharted. We explore this research gap using Snapchat data as a case study. Specifically, we implement a Transformer-based user model with customized training objectives and show that the model can produce high-quality user representations across a broad range of evaluation tasks, among which we introduce three new downstream tasks that concern pivotal topics in user research: user safety, engagement and churn. We also tackle the challenge of efficient extrapolation of long sequences at inference time, by applying a novel positional encoding method.
- 10 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 2431-2436
Open AccessGeneral-Purpose User Modeling with Behavioral Logs: A Snapchat Case Study
Learning general-purpose user representations based on user behavioral logs is an increasingly popular user modeling approach. It benefits from easily available, privacy-friendly yet expressive data, and does not require extensive re-tuning of the upstream user model for different downstream tasks. While this approach has shown promise in search engines and e-commerce applications, its fit for instant messaging platforms, a cornerstone of modern digital communication, remains largely uncharted. We explore this research gap using Snapchat data as a case study. Specifically, we implement a Transformer-based user model with customized training objectives and show that the model can produce high-quality user representations across a broad range of evaluation tasks, among which we introduce three new downstream tasks that concern pivotal topics in user research: user safety, engagement and churn. We also tackle the challenge of efficient extrapolation of long sequences at inference time, by applying a novel positional encoding method.
- Open Access
- 10 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 2930-2934
Improving Embedding-Based Retrieval in Friend Recommendation with ANN Query Expansion
Embedding-based retrieval in graph-based recommendation has shown great improvements over traditional graph walk retrieval methods, and has been adopted in large-scale industry applications such as friend recommendations [16]. However, it is not without its challenges: retraining graph embeddings frequently due to changing data is slow and costly, and producing high recall of approximate nearest neighbor search (ANN) on such embeddings is challenging due to the power law distribution of the indexed users. In this work, we address theses issues by introducing a simple query expansion method in ANN, called FriendSeedSelection, where for each node query, we construct a set of 1-hop embeddings and run ANN search. We highlight our approach does not require any model-level tuning, and is inferred from the data at test-time. This design choice effectively enables our recommendation system to adapt to the changing graph distribution without frequent heavy model retraining. We also discuss how we design our system to efficiently construct such queries online to support 10k+ QPS. For friend recommendation, our method shows improvements of recall, and 11% relative friend reciprocated communication metric gains, now serving over 800 million monthly active users at Snapchat.
- 10 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 2930-2934
Open AccessImproving Embedding-Based Retrieval in Friend Recommendation with ANN Query Expansion
Embedding-based retrieval in graph-based recommendation has shown great improvements over traditional graph walk retrieval methods, and has been adopted in large-scale industry applications such as friend recommendations [16]. However, it is not without its challenges: retraining graph embeddings frequently due to changing data is slow and costly, and producing high recall of approximate nearest neighbor search (ANN) on such embeddings is challenging due to the power law distribution of the indexed users. In this work, we address theses issues by introducing a simple query expansion method in ANN, called FriendSeedSelection, where for each node query, we construct a set of 1-hop embeddings and run ANN search. We highlight our approach does not require any model-level tuning, and is inferred from the data at test-time. This design choice effectively enables our recommendation system to adapt to the changing graph distribution without frequent heavy model retraining. We also discuss how we design our system to efficiently construct such queries online to support 10k+ QPS. For friend recommendation, our method shows improvements of recall, and 11% relative friend reciprocated communication metric gains, now serving over 800 million monthly active users at Snapchat.
- Open Access
- 4 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 2036-2048
CARL-G: Clustering-Accelerated Representation Learning on Graphs
Self-supervised learning on graphs has made large strides in achieving great performance in various downstream tasks. However, many state-of-the-art methods suffer from a number of impediments, which prevent them from realizing their full potential. For instance, contrastive methods typically require negative sampling, which is often computationally costly. While non-contrastive methods avoid this expensive step, most existing methods either rely on overly complex architectures or dataset-specific augmentations. In this paper, we ask: Can we borrow from classical unsupervised machine learning literature in order to overcome those obstacles? Guided by our key insight that the goal of distance-based clustering closely resembles that of contrastive learning: both attempt to pull representations of similar items together and dissimilar items apart. As a result, we propose CARL-G - a novel clustering-based framework for graph representation learning that uses a loss inspired by Cluster Validation Indices (CVIs), i.e., internal measures of cluster quality (no ground truth required). CARL-G is adaptable to different clustering methods and CVIs, and we show that with the right choice of clustering method and CVI, CARL-G outperforms node classification baselines on 4/5 datasets with up to a 79× training speedup compared to the best-performing baseline. CARL-G also performs at par or better than baselines in node clustering and similarity search tasks, training up to 1,500× faster than the best-performing baseline. Finally, we also provide theoretical foundations for the use of CVI-inspired losses in graph representation learning.
- 4 August 2023
- Published by Association for Computing Machinery (ACM)
- p. 2036-2048
Open AccessCARL-G: Clustering-Accelerated Representation Learning on Graphs
Self-supervised learning on graphs has made large strides in achieving great performance in various downstream tasks. However, many state-of-the-art methods suffer from a number of impediments, which prevent them from realizing their full potential. For instance, contrastive methods typically require negative sampling, which is often computationally costly. While non-contrastive methods avoid this expensive step, most existing methods either rely on overly complex architectures or dataset-specific augmentations. In this paper, we ask: Can we borrow from classical unsupervised machine learning literature in order to overcome those obstacles? Guided by our key insight that the goal of distance-based clustering closely resembles that of contrastive learning: both attempt to pull representations of similar items together and dissimilar items apart. As a result, we propose CARL-G - a novel clustering-based framework for graph representation learning that uses a loss inspired by Cluster Validation Indices (CVIs), i.e., internal measures of cluster quality (no ground truth required). CARL-G is adaptable to different clustering methods and CVIs, and we show that with the right choice of clustering method and CVI, CARL-G outperforms node classification baselines on 4/5 datasets with up to a 79× training speedup compared to the best-performing baseline. CARL-G also performs at par or better than baselines in node clustering and similarity search tasks, training up to 1,500× faster than the best-performing baseline. Finally, we also provide theoretical foundations for the use of CVI-inspired losses in graph representation learning.
- Open Access
- 18 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 3330-3334
Embedding Based Retrieval in Friend Recommendation
Friend recommendation systems in online social and professional networks such as Snapchat helps users find friends and build connections, leading to better user engagement and retention. Traditional friend recommendation systems take advantage of the principle of locality and use graph traversal to retrieve friend candidates, e.g. Friends-of-Friends (FoF). While this approach has been adopted and shown efficacy in companies with large online networks such as Linkedin and Facebook, it suffers several challenges: (i) discrete graph traversal offers limited reach in cold-start settings, (ii) it is expensive and infeasible in realtime settings beyond 1 or 2 hop requests owing to latency constraints, and (iii) it cannot well-capture the complexity of graph topology or connection strengths, forcing one to resort to other mechanisms to rank and find top-K candidates. In this paper, we proposed a new Embedding Based Retrieval (EBR) system for retrieving friend candidates, which complements the traditional FoF retrieval by retrieving candidates beyond 2-hop, and providing a natural way to rank FoF candidates. Through online A/B test, we observe statistically significant improvements in the number of friendships made with EBR as an additional retrieval source in both low- and high-density network markets. Our contributions in this work include deploying a novel retrieval system to a large-scale friend recommendation system at Snapchat, generating embeddings for billions of users using Graph Neural Networks, and building EBR infrastructure in production to support Snapchat scale.
- 18 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 3330-3334
Open AccessEmbedding Based Retrieval in Friend Recommendation
Friend recommendation systems in online social and professional networks such as Snapchat helps users find friends and build connections, leading to better user engagement and retention. Traditional friend recommendation systems take advantage of the principle of locality and use graph traversal to retrieve friend candidates, e.g. Friends-of-Friends (FoF). While this approach has been adopted and shown efficacy in companies with large online networks such as Linkedin and Facebook, it suffers several challenges: (i) discrete graph traversal offers limited reach in cold-start settings, (ii) it is expensive and infeasible in realtime settings beyond 1 or 2 hop requests owing to latency constraints, and (iii) it cannot well-capture the complexity of graph topology or connection strengths, forcing one to resort to other mechanisms to rank and find top-K candidates. In this paper, we proposed a new Embedding Based Retrieval (EBR) system for retrieving friend candidates, which complements the traditional FoF retrieval by retrieving candidates beyond 2-hop, and providing a natural way to rank FoF candidates. Through online A/B test, we observe statistically significant improvements in the number of friendships made with EBR as an additional retrieval source in both low- and high-density network markets. Our contributions in this work include deploying a novel retrieval system to a large-scale friend recommendation system at Snapchat, generating embeddings for billions of users using Graph Neural Networks, and building EBR infrastructure in production to support Snapchat scale.
- Open Access
- 19 April 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-14
Reciprocity, Homophily, and Social Network Effects in Pictorial Communication: A Case Study of Bitmoji Stickers
- 19 April 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-14
Open AccessReciprocity, Homophily, and Social Network Effects in Pictorial Communication: A Case Study of Bitmoji Stickers