Peng Jiang

Peng Jiang

Auto-generated profile, based on publicly available information

Peng Jiang

Kuaishou Technology, Beijing, China

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

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

Representative Publications

View All Publications
8 August 2026
Verified
Conference Paper
Open access
KGKaiFeng Guo
KaiFeng Guo
YYYiming Yang
Yiming Yang
JGJingtong Gao
Jingtong Gao
GZGuolei Zeng
Guolei Zeng
FYFukang Yang
Fukang Yang
XZXiangyu Zhao
Xiangyu Zhao
Generative recommenders select items by autoregressively decoding semantic identifiers (SIDs), whose token positions induce a coarse-to-fine hierarchy over the item space. In practice, SID decoders are trained via supervised next-token prediction, which imitates logged trajectories rather than directly optimizing downstream utility. This motivates post-training with outcome feedback to guide decoding toward higher utility. However, logged feedback is only observed for the final exposed item, causing most post-training methods to operate at the item level and broadcast the same terminal signal across all SID tokens. As a result, token-level credit assignment becomes sparse, high-variance, and layer-dependent. To this end, we propose Hierarchical Residual Policy Optimization (HRPO), a post-training framework that converts item-level outcomes into dense, token-aligned learning signals for conservative token-wise improvement. Specifically, HRPO first estimates SID prefix-level utilities via group-wise reward smoothing over feature-based user clusters. It then decomposes these utilities into residual token credits and accumulates them into credit-to-go signals. Finally, Residual-Return Policy Optimization (RRPO) optimizes the residual credits using clipped updates, group-normalized advantages, and KL regularization to preserve stability. Experiments on a public dataset and an online A/B test in a large-scale commercial system show consistent gains in session-level utility and key business metrics. Source code and the archived artifact are available for reproduction.

8 August 2026
Verified
Conference Paper
Open access
YLYewen Li
Yewen Li
JGJingtong Gao
Jingtong Gao
PJPeng Jiang
Peng Jiang
RARuyi An
Ruyi An
XZXiangyu Zhao
Xiangyu Zhao
KGKun Gai
Kun Gai

8 August 2026
Verified
Conference Paper
Open access
SYShengtian Yang
Shengtian Yang
YLYewen Li
Yewen Li
PJPeng Jiang
Peng Jiang
ZLZhiyi Lyu
Zhiyi Lyu
BABo An
Bo An
LFLei Feng
Lei Feng

19 July 2026
Verified
Conference Paper
Open access
LWLixiang Wang
Lixiang Wang
SSShaoyun Shi
Shaoyun Shi
PWPeng Wang
Peng Wang
WWWenjin Wu
Wenjin Wu
PJPeng Jiang
Peng Jiang

8 August 2026
Verified
Conference Paper
Open access
BXBen Xue
Ben Xue
DLDan Liu
Dan Liu
LWLixiang Wang
Lixiang Wang
MSMingjie Sun
Mingjie Sun
PWPeng Wang
Peng Wang
PJPeng Jiang
Peng Jiang

6 November 2026
Verified
Conference Paper
Open access
SYSiyun Yang
Siyun Yang
SYShixiao Yang
Shixiao Yang
JWJian Wang
Jian Wang
DFDi Fan
Di Fan
KCKehe Cai
Kehe Cai
PJPeng Jiang
Peng Jiang