Dongsheng Hou
About Me
I am Dongsheng Hou, currently an undergraduate student in the Department of Computer Science and Engineering at Southern University of Science and Technology (SUSTech). I am a member of the Software Quality Lab, Research Institute of Trustworthy Autonomous Systems (RITAS), under the supervision of Professor Yepang Liu.
Research Interests
Background
Education & Lab
Southern University of Science and Technology (SUSTech)
Undergraduate • Class of 2028
Shenzhen Middle School
High School • Class of 2024
SUSTech Software Quality Lab
Current Researcher
Interests
Beyond Code.
Current Research
SHAPLEYCONTEXTPRUNING: A Cooperative Game Perspective for Context Reranking and Pruning
Submitted: NeurIPS 2026Framing context reranking and pruning as a cooperative game to surface high-value evidence with Shapley-style contributions.
Collaborators: Yanqiao Chen, Yuhan Rui
From Spatial to Spectral: An Efficient, Frequency-Guided Feature Representation Learner for Small Object Detection
Submitted: NeurIPS 2026TL;DR: From Spatial to Spectral, we introduce a plug-and-play Decompose–Enhance–Reconstruct (DER) operator instantiated in the backbone/neck/head to preserve high-frequency cues for efficient small-object detection across diverse detectors.
BCPPO: Bachelier-Inspired Constrained Proximal Policy Optimization for Tail-Risk-Aware Safe Reinforcement Learning
Under Review: AAAI 2027BCPPO combines constrained PPO with bootstrapped cost-critic disagreement and a Bachelier-inspired expected-excess penalty, encouraging caution around uncertain cost predictions while preserving standard critic learning and policy-only deployment.
WebCover: Marginal Coverage Gain-Guided Submodular RL for Efficient Web Testing
Under Review: ICSE 2027WebCover frames automated web GUI testing as submodular reinforcement learning, rewarding marginal coverage gains and prioritizing untried actions to reduce redundant interactions and improve coverage within fixed testing budgets.