Jiaming Liang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jiaming Liang has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
Yimin Zhu
School of Business, Sun Yat-Sen University, Guangzhou, China
2 shared publications- Ashwin Padaki
University of Pennsylvania
1 shared publication Erik Waingarten
University of Pennsylvania
1 shared publication- Guoliang Li
Tsinghua University
1 shared publication - Jiemin Zhang
Business School, Hohai University Department of Marketing
1 shared publication Jifei Wu
1 shared publication
Latest Publications
- Open Access
Prune, Don't Rebuild: Efficiently Tuning $α$-Reachable Graphs for Nearest Neighbor Search
Vector similarity search is an essential primitive in modern AI and ML applications. Most vector databases adopt graph-based approximate nearest neighbor (ANN) search algorithms, such as DiskANN (Subramanya et al., 2019), which have demonstrated state-of-the-art empirical performance. DiskANN's graph construction is governed by a reachability parameter $α$, which gives a trade-off between construction time, query time, and accuracy. However, adaptively tuning this trade-off typically requires rebuilding the index for different $α$ values, which is prohibitive at scale. In this work, we propose RP-Tuning, an efficient post-hoc routine, based on DiskANN's pruning step, to adjust the $α$ parameter without reconstructing the full index. Within the $α$-reachability framework of prior theoretical works (Indyk and Xu, 2023; Gollapudi et al., 2025), we prove that pruning an initially $α$-reachable graph with RP-Tuning preserves worst-case reachability guarantees in general metrics and improved guarantees in Euclidean metrics. Empirically, we show that RP-Tuning accelerates DiskANN tuning on four public datasets by up to $43\times$ with negligible overhead.Open AccessPrune, Don't Rebuild: Efficiently Tuning $α$-Reachable Graphs for Nearest Neighbor Search
Vector similarity search is an essential primitive in modern AI and ML applications. Most vector databases adopt graph-based approximate nearest neighbor (ANN) search algorithms, such as DiskANN (Subramanya et al., 2019), which have demonstrated state-of-the-art empirical performance. DiskANN's graph construction is governed by a reachability parameter $α$, which gives a trade-off between construction time, query time, and accuracy. However, adaptively tuning this trade-off typically requires rebuilding the index for different $α$ values, which is prohibitive at scale. In this work, we propose RP-Tuning, an efficient post-hoc routine, based on DiskANN's pruning step, to adjust the $α$ parameter without reconstructing the full index. Within the $α$-reachability framework of prior theoretical works (Indyk and Xu, 2023; Gollapudi et al., 2025), we prove that pruning an initially $α$-reachable graph with RP-Tuning preserves worst-case reachability guarantees in general metrics and improved guarantees in Euclidean metrics. Empirically, we show that RP-Tuning accelerates DiskANN tuning on four public datasets by up to $43\times$ with negligible overhead. - 22 January 2026
- Published by SAGE Publications in Journal of Hospitality & Tourism Research
Technology or Market? How Innovation Drivers Shape Consumer Responses to AI-Designed Services
The rapid advancement of Artificial Intelligence (AI) is substantially transforming service design in the hospitality and tourism sector. Despite extensive discussion about the potential benefits of AI in this field, consumers’ reactions to human-designed services compared with AI-designed services have not been sufficiently explored. Drawing on stereotype theory, this research proposes that human- and AI-designed services elicit divergent consumer responses under different innovation-driving contexts. Specifically, consumers respond more positively to human-designed services in market-driven innovation contexts, whereas they favor AI-designed services in technology-driven innovation contexts. Through three empirical studies, this research demonstrates a significant interaction effect between design sources and innovation drivers on consumer response, mediated by perceived uncertainty. The findings not only enhance theoretical understanding of AI’s role in service design and new service development, but also provide practical guidance for hospitality and tourism firms on whether to disclose AI design in design of new services.
- 22 January 2026
- Published by SAGE Publications in Journal of Hospitality & Tourism Research
Technology or Market? How Innovation Drivers Shape Consumer Responses to AI-Designed Services
The rapid advancement of Artificial Intelligence (AI) is substantially transforming service design in the hospitality and tourism sector. Despite extensive discussion about the potential benefits of AI in this field, consumers’ reactions to human-designed services compared with AI-designed services have not been sufficiently explored. Drawing on stereotype theory, this research proposes that human- and AI-designed services elicit divergent consumer responses under different innovation-driving contexts. Specifically, consumers respond more positively to human-designed services in market-driven innovation contexts, whereas they favor AI-designed services in technology-driven innovation contexts. Through three empirical studies, this research demonstrates a significant interaction effect between design sources and innovation drivers on consumer response, mediated by perceived uncertainty. The findings not only enhance theoretical understanding of AI’s role in service design and new service development, but also provide practical guidance for hospitality and tourism firms on whether to disclose AI design in design of new services.
- 18 December 2025
- Published by Emerald Publishing in Journal of Research in Interactive Marketing
- Vol. 20 (8) , 1146-1162
Doing bad when facing AI: the influence of interactions with AI or human agents on unethical consumer behaviour
Purpose: As artificial intelligence (AI) becomes an increasingly prevalent social agent in consumers' lives, instances of unethical consumer behaviour towards it are emerging. This study investigates how perceiving AI as lacking responsibility reduces consumers' anticipatory guilt and thereby encourages unethical behaviour towards it. Design/methodology/approach: This study conducted three scenario-based experimental studies with a total of 510 participants, to examine the effect of the service-agent type (human vs AI) on unethical consumer behaviour. Specifically, it investigated the sequential mediation of responsibility attribution and anticipatory guilt, as well as the moderating role of social distance between consumer and the service agent. Findings: The results showed that consumers may feel less anticipatory guilt for their unethical behaviours towards AI agents because AI is regarded as less responsible for the losses resulting from the unethical behaviour, which in turn drives them to engage more. But this effect is attenuated when the social distance between consumer and the AI agent is close. Originality/value: Our findings reveal the underlying mechanisms of responsibility attribution and anticipatory guilt in unethical consumer behaviour towards AI agents and enrich the application of moral disengagement theory in consumer–AI interactions.
- 18 December 2025
- Published by Emerald Publishing in Journal of Research in Interactive Marketing
- Vol. 20 (8) , 1146-1162
Doing bad when facing AI: the influence of interactions with AI or human agents on unethical consumer behaviour
Purpose: As artificial intelligence (AI) becomes an increasingly prevalent social agent in consumers' lives, instances of unethical consumer behaviour towards it are emerging. This study investigates how perceiving AI as lacking responsibility reduces consumers' anticipatory guilt and thereby encourages unethical behaviour towards it. Design/methodology/approach: This study conducted three scenario-based experimental studies with a total of 510 participants, to examine the effect of the service-agent type (human vs AI) on unethical consumer behaviour. Specifically, it investigated the sequential mediation of responsibility attribution and anticipatory guilt, as well as the moderating role of social distance between consumer and the service agent. Findings: The results showed that consumers may feel less anticipatory guilt for their unethical behaviours towards AI agents because AI is regarded as less responsible for the losses resulting from the unethical behaviour, which in turn drives them to engage more. But this effect is attenuated when the social distance between consumer and the AI agent is close. Originality/value: Our findings reveal the underlying mechanisms of responsibility attribution and anticipatory guilt in unethical consumer behaviour towards AI agents and enrich the application of moral disengagement theory in consumer–AI interactions.
- Open Access
RITA: Group Attention is All You Need for Timeseries Analytics
Timeseries analytics is of great importance in many real-world applications. Recently, the Transformer model, popular in natural language processing, has been leveraged to learn high quality feature embeddings from timeseries, core to the performance of various timeseries analytics tasks. However, the quadratic time and space complexities limit Transformers' scalability, especially for long timeseries. To address these issues, we develop a timeseries analytics tool, RITA, which uses a novel attention mechanism, named group attention, to address this scalability issue. Group attention dynamically clusters the objects based on their similarity into a small number of groups and approximately computes the attention at the coarse group granularity. It thus significantly reduces the time and space complexity, yet provides a theoretical guarantee on the quality of the computed attention. The dynamic scheduler of RITA continuously adapts the number of groups and the batch size in the training process, ensuring group attention always uses the fewest groups needed to meet the approximation quality requirement. Extensive experiments on various timeseries datasets and analytics tasks demonstrate that RITA outperforms the state-of-the-art in accuracy and is significantly faster -- with speedups of up to 63X.
Open AccessRITA: Group Attention is All You Need for Timeseries Analytics
Timeseries analytics is of great importance in many real-world applications. Recently, the Transformer model, popular in natural language processing, has been leveraged to learn high quality feature embeddings from timeseries, core to the performance of various timeseries analytics tasks. However, the quadratic time and space complexities limit Transformers' scalability, especially for long timeseries. To address these issues, we develop a timeseries analytics tool, RITA, which uses a novel attention mechanism, named group attention, to address this scalability issue. Group attention dynamically clusters the objects based on their similarity into a small number of groups and approximately computes the attention at the coarse group granularity. It thus significantly reduces the time and space complexity, yet provides a theoretical guarantee on the quality of the computed attention. The dynamic scheduler of RITA continuously adapts the number of groups and the batch size in the training process, ensuring group attention always uses the fewest groups needed to meet the approximation quality requirement. Extensive experiments on various timeseries datasets and analytics tasks demonstrate that RITA outperforms the state-of-the-art in accuracy and is significantly faster -- with speedups of up to 63X.