Bo Pang

Bo Pang

Auto-generated profile, based on publicly available information

Bo Pang

Department of Computer Science and Technology, Shanghai Jiao Tong University, Shanghai, China

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

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

Representative Publications

View All Publications
27 April 2020 in Nature Machine Intelligence
Verified
Journal Article
Closed access
PDF file
BPBo Pang
Bo Pang
KZKaiwen Zha
Kaiwen Zha
HCHanwen Cao
Hanwen Cao
JTJiajun Tang
Jiajun Tang
MYMinghui Yu
Minghui Yu
CLCewu Lu
Cewu Lu

30 October 2008
Verified
Conference Paper
Open access
PDF file
XWXuerui Wang
Xuerui Wang
ABAndrei Broder
Andrei Broder
EGEvgeniy Gabrilovich
Evgeniy Gabrilovich
VJVanja Josifovski
Vanja Josifovski
BPBo Pang
Bo Pang
The non-English Web is growing at breakneck speed, but available language processing tools are mostly English based. Taxonomies are a case in point: while there are plenty of commercial and non-commercial taxonomies for the English Web, taxonomies for other languages are either not available or of very limited quality. Given that building taxonomies in all non-English languages is prohibitively expensive, it is natural to ask whether existing English taxonomies can be leveraged, possibly via machine translation, to enable information processing tasks in other languages. Preliminary results presented in this paper indicate that the answer is affirmative with respect to query classification, a task which is essential both for understanding the user intent and thus provide better search results, and for better targeting of search-based advertising, the economic underpinning of commercial Web search engines. We propose a robust method for classifying non-English queries against an English taxonomy and classifier using widely available, off-the-shelf machine translation systems. In particular, we show that by viewing the search results in the query's original language as independent sources of information, we can alleviate the impact of poor quality or erroneous machine translations. Empirical results for Chinese queries show that we achieve remarkably encouraging results.

12 July 2024 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
BPBo Pang
Bo Pang
GPGao Peng
Gao Peng
YLYizhuo Li
Yizhuo Li
CLCewu Lu
Cewu Lu
The computational complexity of video models increases linearly with the square number of frames. Thus, constrained bycomputational resources, training video models to learn long-term temporal semantics end-to-end is quite a challenge. Currently, the main-stream method is to split a raw video into clips, leading to incomplete fragmentary temporal information flow and failure of modeling long-term semantics. In this paper, we design the Markov Progressive framework (MaPro), a theoretical framework consisting of the progressive modeling method and a paradigm model tailored for it. Thecore idea of MaPro is to find a paradigm model consisting of proposed Markov operators which can be trained in multiple sequential steps and ensure that the multi-step progressive modeling is equivalent to the conventional end-to-endmodeling. By training the paradigm model under the progressive method, we are able to model long videos end-to-endwith limited resources and ensure the effective transmission of long-term temporal information. We provide implementations of this theoretical system on the mainstream CNN- and Transformer-based models, where they are modified to conform to the Markov paradigm. As a general and robust training method, we experimentally demonstrate that it yields significant performance improvements on different backbones and datasets. As an illustrative example, the proposed method improves the SlowOnly network by 4.1 mAP on Charades and 2.5 top-1 accuracy on Kinetics. And for TimeSformer, MaPro improves its performance on Kinetics by 2.0 top-1 accuracy. Importantly, all these improvements areachieved with a little parameter and computation overhead.

9 January 2025 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
BPBo Pang
Bo Pang
ZWZhenyu Wei
Zhenyu Wei
JLJingli Lin
Jingli Lin
CLCewu Lu
Cewu Lu
Contrastive learning, a discriminative self-learning framework, is one of the most popular representation learning methods which has a wide range of application scenarios. Although relative techniques have been continuously updated in recent years, designing and seeking positive pairs are still inevitable. Just because of the requirement of explicit positive pairs, the utilization of contrastive learning is restricted in dense, multi-modal, and other scenarios where positive pairs are difficult to obtain. To solve this problem, in this paper, we design an auto-pairing mechanism called Implicit Relation Circulation (IRC) for discriminative self-learning frameworks. Its core idea is to conduct a random walk among multiple feature groups we want to contrast but without explicit matchup, which we call the complex task (Task C). By linking the head and tail of the random walk to form a circulation with a simple task (task S) containing easy-obtaining pairs, we can apply cycle consistency as supervision guidance to gradually learn the wanted positive pairs among the random walk of feature groups automatically. We provide several amazing applications of IRC: we can learn 1) effective dense image pixel relations and representation with only image-level pairs; 2) 3D temporal point-level multi-modal point cloud relations and representation; and 3) even image representation with the help of language without off-the-shelf vision-language pairs. As an easy-to-use plug-and-play mechanism, we evaluate its universality and robustness with multiple self-learning algorithms, tasks, and datasets, achieving stable and significant improvements. As an illustrative example, IRC improves the SOTA performance by about 3.0 mIoU on image semantic segmentation, 1.5 mIoU on 3D segmentation, 1.3 mAP on 3D detection, and an average of 1.2 top1 accuracy on image classification with the help of the auto-learned positive pairs. Importantly, these improvements are achieved with little parameter and computation overhead. We hope IRC can provide the community with new insight into discriminative self-learning.