arXiv subjects

Zhenfeng Liu

Publications and source records attributed to Zhenfeng Liu.

4 recordsLinked to original sources

arXiv2026

RecRM-Bench: Benchmarking Multidimensional Reward Modeling for Agentic Recommender Systems

The integration of Large Language Model (LLM) agents is transforming recommender systems from simple query-item matching towards deeply personalized and interactive recommendations. Reinforcement Learning (RL) provides an essential framework for the optimization of these agents in recommendation tasks. However, current methodologies remain limited by a reliance on single dimensional outcome-based rewards that focus exclusively on final user interactions, overlooking critical intermediate capabilities, such as instruction following and complex intent understanding. Despite the necessity for designing multi-dimensional reward, the field lacks a standardized benchmark to facilitate this development. To bridge this gap, we introduce RecRM-Bench, the largest and most comprehensive benchmark to date for agentic recommender systems. It comprises over 1 million structured entries across four core evaluation dimensions: instruction following, factual consistency, query-item relevance, and fine-grained user behavior prediction. By supporting comprehensive assessment from syntactic compliance to complex intent grounding and preference modeling, RecRM-Bench provides a foundational dataset for training sophisticated reward models. Furthermore, we propose a systematic framework for the construction of multi-dimensional reward models and the integration of a hybrid reward function, establishing a robust foundation for developing reliable and highly capable agentic recommender systems. The complete RecRM-Bench dataset is publicly available at https://huggingface.co/datasets/wwzeng/RecRM-Bench.

cs.IR↗

arXiv2025

LocalSearchBench: Benchmarking Agentic Search in Real-World Local Life Services

Recent advances in large reasoning models LRMs have enabled agentic search systems to perform complex multi-step reasoning across multiple sources. However, most studies focus on general information retrieval and rarely explores vertical domains with unique challenges. In this work, we focus on local life services and introduce LocalSearchBench, which encompass diverse and complex business scenarios. Real-world queries in this domain are often ambiguous and require multi-hop reasoning across merchants and products, remaining challenging and not fully addressed. As the first comprehensive benchmark for agentic search in local life services, LocalSearchBench comprises a database of over 1.3M merchant entries across 6 service categories and 9 major cities, and 900 multi-hop QA tasks from real user queries that require multi-step reasoning. We also developed LocalPlayground, a unified environment integrating multiple tools for LRMs interaction. Experiments show that even state-of-the-art LRMs struggle on LocalSearchBench: the best model (DeepSeek-V3.2) achieves only 35.60% correctness, and most models have issues with completeness (average 60.32%) and faithfulness (average 30.72%). This highlights the need for specialized benchmarks and domain-specific agent training in local life services. Code, Benchmark, and Leaderboard are available at https://localsearchbench.github.io/.

cs.AI↗

arXiv2024

Dual-frequency optical-microwave atomic clocks based on cesium atoms

$^{133}$Cs, which is the only stable cesium (Cs) isotope, is one of the most investigated elements in atomic spectroscopy and was used to realize the atomic clock in 1955. Among all atomic clocks, the cesium atomic clock has a special place, since the current unit of time is based on a microwave transition in the Cs atom. In addition, the long lifetime of the $6{\text{P}}_{3/2}$ state and simple preparation technique of Cs vapor cells have great relevance to quantum and atom optics experiments, which suggests the use of the $6{\text{S}} - 6{\text{P}}$ D2 transition as an optical frequency standard. In this work, using one laser as the local oscillator and Cs atoms as the quantum reference, we realized two atomic clocks in the optical and microwave frequencies, respectively. Both clocks could be freely switched or simultaneously output. The optical clock based on the vapor cell continuously operated with a frequency stability of $3.89 \times {10^{ - 13}}$ at 1 s, decreasing to $2.17 \times {10^{ - 13}}$ at 32 s, which was frequency stabilized by modulation transfer spectroscopy and estimated by an optical comb. Then, applying this stabilized laser for an optically pumped Cs beam atomic clock to reduce the laser frequency noise, we obtained a microwave clock with a frequency stability of $1.84 \times {10^{ - 12}}/\sqrt τ$, reaching $5.99 \times {10^{ - 15}}$ at $10^5$ s. This study demonstrates an attractive feature for the commercialization and deployment of optical and microwave clocks and will guide further development of integrated atomic clocks with better stability. Thus, this study lays the groundwork for future quantum metrology and laser physics.

physics.atom-ph↗

arXiv2023

Realistic Rainy Weather Simulation for LiDARs in CARLA Simulator

Employing data augmentation methods to enhance perception performance in adverse weather has attracted considerable attention recently. Most of the LiDAR augmentation methods post-process the existing dataset by physics-based models or machine-learning methods. However, due to the limited environmental annotations and the fixed vehicle trajectories in the existing dataset, it is challenging to edit the scene and expand the diversity of traffic flow and scenario. To this end, we propose a simulator-based physical modeling approach to augment LiDAR data in rainy weather in order to improve the perception performance of LiDAR in this scenario. We complete the modeling task of the rainy weather in the CARLA simulator and establish a pipeline for LiDAR data collection. In particular, we pay special attention to the spray and splash rolled up by the wheels of surrounding vehicles in rain and complete the simulation of this special scenario through the Spray Emitter method we developed. In addition, we examine the influence of different weather conditions on the intensity of the LiDAR echo, develop a prediction network for the intensity of the LiDAR echo, and complete the simulation of 4-feat LiDAR point cloud data. In the experiment, we observe that the model augmented by the synthetic data improves the object detection task's performance in the rainy sequence of the Waymo Open Dataset. Both the code and the dataset will be made publicly available at https://github.com/PJLab-ADG/PCSim#rainypcsim.

cs.RO↗