Chengcheng Yang

Lingyang Zeng, Guangze Chen, Kaichen Yu et al.

Personalized agents are increasingly applied to assist users across a wide range of tasks. Effective personalized assistance requires not only retrieving explicit facts from past interactions stored in agent memory, but also inferring abstract personal characteristics. However, existing memory benchmarks primarily evaluate whether an agent can retrieve information explicitly stated in conversational histories, failing to provide an effective assessment of deeper user understanding. In this work, we propose Setoka, a benchmark for evaluating memory-augmented personalized agents with hierarchical user understanding from heterogeneous data. Grounded in theories from cognitive and personality psychology, Setoka defines four levels of user understanding, i.e., semantic memory, episodic memory, behavior pattern, and personality trait. Moreover, to enable realistic yet privacy-preserving evaluation, we design a psychometrics-based pipeline that synthesizes diverse, coherent heterogeneous user data and queries at scale. Finally, we leverage Setoka to evaluate 3 language models combined with 5 memory systems for 10 synthetic users. Our comprehensive evaluation reveals that while existing systems perform well on semantic memory retrieval, their performance declines on episodic memory. Moreover, when dealing with behavior pattern and personality trait understanding tasks that require integrating heterogeneous and fragmented information dispersed over time, performance declines even further. These findings demonstrate that user understanding cannot be handled by simple fact retrieval, motivating the design of memory mechanisms for cross-source integration and abstraction over long-term user behavior.

Jin-Hu Liu, Zi-Ke Zhang, Chengcheng Yang et al.

In this paper, based on the gravity principle of classical physics, we propose a tunable gravity-based model, which considers tag usage pattern to weigh both the mass and distance of network nodes. We then apply this model in solving the problems of information filtering and network evolving. Experimental results on two real-world data sets, \emph{Del.icio.us} and \emph{MovieLens}, show that it can not only enhance the algorithmic performance, but can also better characterize the properties of real networks. This work may shed some light on the in-depth understanding of the effect of gravity model.

Jinhu Liu, Chengcheng Yang, Zi-Ke Zhang

Recommender systems can change our life a lot and help us select suitable and favorite items much more conveniently and easily. As a consequence, various kinds of algorithms have been proposed in last few years to improve the performance. However, all of them face one critical problem: data sparsity. In this paper, we proposed a two-step recommendation algorithm via iterative local least squares (ILLS). Firstly, we obtain the ratings matrix which is constructed via users' behavioral records, and it is normally very sparse. Secondly, we preprocess the "ratings" matrix through ProbS which can convert the sparse data to a dense one. Then we use ILLS to estimate those missing values. Finally, the recommendation list is generated. Experimental results on the three datasets: MovieLens, Netflix, RYM, suggest that the proposed method can enhance the algorithmic accuracy of AUC. Especially, it performs much better in dense datasets. Furthermore, since this methods can improve those missing value more accurately via iteration which might show light in discovering those inactive users' purchasing intention and eventually solving cold-start problem.