Reinforcement Learning with Decomposed Subtasks
本文针对多技能任务中环境反馈稀疏延迟的问题，提出RLDS方法，通过分解子任务奖励来优化策略更新。
0 citationsRead paper
Selected work
Representative Papers
GraphMatch: Fusing Language and Graph Representations in a Dynamic Two-Sided Work Marketplace
To address the challenge of matching recommendations in dynamic, text-rich two-sided job markets, this paper proposes an end-to-end joint learning framework that pioneers deep integration of pretrained language models (PLMs) and temporal graph neural networks (GNNs). Specifically, PLMs capture fine-grained semantic evolution from job postings and candidate profiles, while moment-wise subgraph sampling and adversarial negative sampling jointly model structural and temporal dynamics of interaction graphs. The method overcomes longstanding bottlenecks in synergistic language–graph representation learning and incorporates low-latency inference optimizations for scalable real-time deployment. Evaluated on the real-world Upwork dataset, it significantly outperforms unimodal baselines—both pure language- and pure graph-based—achieving superior matching accuracy and computational efficiency. The system has been successfully deployed in production.
0 citationsRead paper
Recent publications
Latest Papers
Reinforcement Learning with Decomposed Subtasks
本文针对多技能任务中环境反馈稀疏延迟的问题，提出RLDS方法，通过分解子任务奖励来优化策略更新。
0 citationsRead paper
GraphMatch: Fusing Language and Graph Representations in a Dynamic Two-Sided Work Marketplace
To address the challenge of matching recommendations in dynamic, text-rich two-sided job markets, this paper proposes an end-to-end joint learning framework that pioneers deep integration of pretrained language models (PLMs) and temporal graph neural networks (GNNs). Specifically, PLMs capture fine-grained semantic evolution from job postings and candidate profiles, while moment-wise subgraph sampling and adversarial negative sampling jointly model structural and temporal dynamics of interaction graphs. The method overcomes longstanding bottlenecks in synergistic language–graph representation learning and incorporates low-latency inference optimizations for scalable real-time deployment. Evaluated on the real-world Upwork dataset, it significantly outperforms unimodal baselines—both pure language- and pure graph-based—achieving superior matching accuracy and computational efficiency. The system has been successfully deployed in production.
0 citationsRead paper