Liang Wu

Liang Wu

Auto-generated profile, based on publicly available information

Liang Wu

LinkedIn Inc., Sunnyvale, CA, USA

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 May 2024
Verified
Conference Paper
Open access
YZYaochen Zhu
Yaochen Zhu
LWLiang Wu
Liang Wu
QGQi Guo
Qi Guo
LHLiangjie Hong
Liangjie Hong
JLJundong Li
Jundong Li
Recently, there has been growing interest in developing the next-generation recommender systems (RSs) based on pretrained large language models (LLMs). However, the semantic gap between natural language and recommendation tasks is still not well addressed, leading to multiple issues such as spuriously correlated user/item descriptors, ineffective language modeling on user/item data, inefficient recommendations via auto-regression, etc. In this paper, we propose CLLM4Rec, the first generative RS that tightly integrates the LLM paradigm and ID paradigm of RSs, aiming to address the above challenges simultaneously. We first extend the vocabulary of pretrained LLMs with user/item ID tokens to faithfully model user/item collaborative and content semantics. Accordingly, a novel soft+hard prompting strategy is proposed to effectively learn user/item collaborative/content token embeddings via language modeling on RS-specific corpora, where each document is split into a prompt consisting of heterogeneous soft (user/item) tokens and hard (vocab) tokens and a main text consisting of homogeneous item tokens or vocab tokens to facilitate stable and effective language modeling. In addition, a novel mutual regularization strategy is introduced to encourage CLLM4Rec to capture recommendation-related information from noisy user/item content. Finally, we propose a novel recommendation-oriented finetuning strategy for CLLM4Rec, where an item prediction head with multinomial likelihood is added to the pretrained CLLM4Rec backbone to predict hold-out items based on soft+hard prompts established from masked user-item interaction history, where recommendations of multiple items can be generated efficiently without hallucination.

18 May 2018 in Language Resources and Evaluation
Verified
Journal Article
Closed access
PDF file
LWLiang Wu
Liang Wu
FMFred Morstatter
Fred Morstatter
HLHuan Liu
Huan Liu

5 August 2025 in ACM Transactions on Intelligent Systems and Technology
Verified
Journal Article
Open access
XWXinyuan Wang
Xinyuan Wang
LWLiang Wu
Liang Wu
LHLiangjie Hong
Liangjie Hong
HLHao Liu
Hao Liu
Yanjie FuYanjie Fu
Yanjie Fu
Graph recommendation methods, representing a connected interaction perspective, reformulate user-item interactions as graphs to leverage graph structure and topology to recommend and have proved practical effectiveness at scale. Large language models, representing a textual generative perspective, excel at modeling user languages, understanding behavioral contexts, capturing user-item semantic relationships, analyzing textual sentiments, and generating coherent and contextually relevant texts as recommendations. However, there is a gap between the connected graph perspective and the text generation perspective as the task formulations are different. A research question arises: how can we effectively integrate the two perspectives for more personalized recsys? To fill this gap, we propose to incorporate graph-edge information into LLMs via prompt and attention innovations. We reformulate recommendations as a probabilistic generative problem using prompts. We develop a framework to incorporate graph edge information from the prompt and attention mechanisms for graph-structured LLM recommendations. We develop a new prompt design that brings in both first-order and second-order graph relationships; we devise an improved LLM attention mechanism to embed direct the spatial and connectivity information of edges. Our evaluation of real-world datasets demonstrates the framework's ability to understand connectivity information in graph data and to improve the relevance and quality of recommendation results. Our code is released at: https://github.com/anord-wang/LLM4REC.git .

1 October 2015
Verified
Conference Paper
Closed access
LWLiang Wu
Liang Wu
TMTeng-Sheng Moh
Teng-Sheng Moh
NKNatalia Khuri
Natalia Khuri
Although rigorous clinical studies are required before a drug is placed on the market, it is impossible to predict all side effects for the approved medication. The United States Food and Drug Administration actively monitors approved drugs to identify adverse events. The FDA Adverse Event Reporting System (FAERS) contains a database of adverse drug events reported by the healthcare providers and consumers. The ubiquitous online social networks, such as Twitter, can provide complementary information about adverse drug events. Short Twitter postings, or tweets, are often used to express an opinion about drugs, as well as solicit and receive feedback from consumers of a drug. Thus, adverse drug events can be discovered by extracting from tweets users' opinions about drugs. Here, we developed a computational pipeline for collecting, processing, and analyzing tweets to find signals about adverse drug reactions, defined as drug side effects caused by a drug at a normal dose during normal use. Manual examination of processed tweets identified several known side effects of four drugs.

20 July 2025
Verified
Conference Paper
Open access
PLPing Liu
Ping Liu
HWHaichao Wei
Haichao Wei
XHXiaochen Hou
Xiaochen Hou
JSJianqiang Shen
Jianqiang Shen
SHShihai He
Shihai He
WZWenjing Zhang
Wenjing Zhang

4 August 2023
Verified
Conference Paper
Open access
YZYaochen Zhu
Yaochen Zhu
JMJing Ma
Jing Ma
LWLiang Wu
Liang Wu
QGQi Guo
Qi Guo
LHLiangjie Hong
Liangjie Hong
JLJundong Li
Jundong Li
Recommender systems (RSs) have become an indispensable part of online platforms. With the growing concerns of algorithmic fairness, RSs are not only expected to deliver high-quality personalized content, but are also demanded not to discriminate against users based on their demographic information. However, existing RSs could capture undesirable correlations between sensitive features and observed user behaviors, leading to biased recommendations. Most fair RSs tackle this problem by completely blocking the influences of sensitive features on recommendations. But since sensitive features may also affect user interests in a fair manner (e.g., race on culture-based preferences), indiscriminately eliminating all the influences of sensitive features inevitably degenerate the recommendations quality and necessary diversities. To address this challenge, we propose a path-specific fair RS (PSF-RS) for recommendations. Specifically, we summarize all fair and unfair correlations between sensitive features and observed ratings into two latent proxy mediators, where the concept of path-specific bias (PS-Bias) is defined based on path-specific counterfactual inference. Inspired by Pearl's minimal change principle, we address the PS-Bias by minimally transforming the biased factual world into a hypothetically fair world, where a fair RS model can be learned accordingly by solving a constrained optimization problem. For the technical part, we propose a feasible implementation of PSF-RS, i.e., PSF-VAE, with weakly-supervised variational inference, which robustly infers the latent mediators such that unfairness can be mitigated while necessary recommendation diversities can be maximally preserved simultaneously. Experiments conducted on semi-simulated and real-world datasets demonstrate the effectiveness of PSF-RS.