arXiv subjects

Zhefan Wang

Publications and source records attributed to Zhefan Wang.

12 recordsLinked to original sources

arXiv2026

Personalized Turn-Level User Conversation Satisfaction Benchmark

User satisfaction with AI assistants is highly personalized: the same response may satisfy one user but disappoint another depending on what each user expects and what they have asked for before. Existing automatic evaluation methods mostly measure generic response quality, making it difficult to judge whether a response satisfies a user at a specific turn. We study this problem as personalized turn-level user conversation satisfaction evaluation. We build a conversation satisfaction evaluator that combines compact user memories with target-turn context to produce satisfaction scores and dissatisfaction-oriented rationales. Meta-evaluation against human satisfaction annotations shows that personalized memory and post-hoc score calibration improve ordinal agreement and dissatisfied-turn detection over supervised, retrieval-based, and generic LLM-as-a-judge baselines. We further introduce PersTurnBench, a personalized turn-level user conversation satisfaction benchmark that uses the verified evaluator to assess generation models via replay. By holding the replay state fixed, PersTurnBench enables controlled comparison of generic generation models and memory-augmented personalized systems without new human labels for every candidate model. The evaluator and benchmark let researchers compare candidate generation models on personalized satisfaction without collecting new user feedback for every model.

cs.CL↗

arXiv2026

Knowledge-to-Verification: Exploring RLVR for LLMs in Knowledge-Intensive Domains

Reinforcement learning with verifiable rewards (RLVR) has demonstrated promising potential to enhance the reasoning capabilities of large language models (LLMs) in domains such as mathematics and coding. However, its applications on knowledge-intensive domains have not been effectively explored due to the scarcity of high-quality verifiable data. Furthermore, current RLVR focuses solely on the correctness of final answers, leading to the limitations of flawed reasoning and sparse reward signals. In this work, we propose Knowledge-to-Verification (K2V), a framework that extends RLVR to knowledge-intensive domains through automated verifiable data synthesis, while enabling verification of the LLM's reasoning process. Extensive experiments demonstrate that K2V enhances the reasoning of LLM in knowledge-intensive domains without significantly compromising the model's general capabilities. This study also suggests that integrating automated data synthesis with reasoning verification is a promising direction to enhance model capabilities in these broader domains. Code is available at https://github.com/SeedScientist/K2V.

cs.CL↗

arXiv2025

Human vs. Agent in Task-Oriented Conversations

Task-oriented conversational systems are essential for efficiently addressing diverse user needs, yet their development requires substantial amounts of high-quality conversational data that is challenging and costly to obtain. While large language models (LLMs) have demonstrated potential in generating synthetic conversations, the extent to which these agent-generated interactions can effectively substitute real human conversations remains unclear. This work presents the first systematic comparison between LLM-simulated users and human users in personalized task-oriented conversations. We propose a comprehensive analytical framework encompassing three key aspects (conversation strategy, interaction style, and conversation evaluation) and ten distinct dimensions for evaluating user behaviors, and collect parallel conversational datasets from both human users and LLM agent users across four representative scenarios under identical conditions. Our analysis reveals significant behavioral differences between the two user types in problem-solving approaches, question broadness, user engagement, context dependency, feedback polarity and promise, language style, and hallucination awareness. We found consistency in the agent users and human users across the depth-first or breadth-first dimensions, as well as the usefulness dimensions. These findings provide critical insights for advancing LLM-based user simulation. Our multi-dimensional taxonomy constructed a generalizable framework for analyzing user behavior patterns, offering insights from LLM agent users and human users. By this work, we provide perspectives on rethinking how to use user simulation in conversational systems in the future.

cs.IR↗

arXiv2025

Generative Sequential Recommendation via Hierarchical Behavior Modeling

Recommender systems in multi-behavior domains, such as advertising and e-commerce, aim to guide users toward high-value but inherently sparse conversions. Leveraging auxiliary behaviors (e.g., clicks, likes, shares) is therefore essential. Recent progress on generative recommendations has brought new possibilities for multi-behavior sequential recommendation. However, existing generative approaches face two significant challenges: 1) Inadequate Sequence Modeling: capture the complex, cross-level dependencies within user behavior sequences, and 2) Lack of Suitable Datasets: publicly available multi-behavior recommendation datasets are almost exclusively derived from e-commerce platforms, limiting the validation of feasibility in other domains, while also lacking sufficient side information for semantic ID generation. To address these issues, we propose a novel generative framework, GAMER (Generative Augmentation and Multi-lEvel behavior modeling for Recommendation), built upon a decoder-only backbone. GAMER introduces a cross-level interaction layer to capture hierarchical dependencies among behaviors and a sequential augmentation strategy that enhances robustness in training. To further advance this direction, we collect and release ShortVideoAD, a large-scale multi-behavior dataset from a mainstream short-video platform, which differs fundamentally from existing e-commerce datasets and provides pretrained semantic IDs for research on generative methods. Extensive experiments show that GAMER consistently outperforms both discriminative and generative baselines across multiple metrics.

cs.IR↗

arXiv2025

StepTool: Enhancing Multi-Step Tool Usage in LLMs via Step-Grained Reinforcement Learning

Despite their powerful text generation capabilities, large language models (LLMs) still struggle to effectively utilize external tools to solve complex tasks, a challenge known as tool learning. Existing methods primarily rely on supervised fine-tuning, treating tool learning as a text generation problem while overlooking the decision-making complexities inherent in multi-step contexts. In this work, we propose modeling tool learning as a dynamic decision-making process and introduce StepTool, a novel step-grained reinforcement learning framework that enhances LLMs' capabilities in multi-step tool use. StepTool comprises two key components: Step-grained Reward Shaping, which assigns rewards to each tool interaction based on its invocation success and contribution to task completion; and Step-grained Optimization, which applies policy gradient methods to optimize the model across multiple decision steps. Extensive experiments across diverse benchmarks show that StepTool consistently outperforms both SFT-based and RL-based baselines in terms of task Pass Rate and Recall of relevant tools. Furthermore, our analysis suggests that StepTool helps models discover new tool-use strategies rather than merely re-weighting prior knowledge. These results highlight the importance of fine-grained decision modeling in tool learning and establish StepTool as a general and robust solution for enhancing multi-step tool use in LLMs. Code and data are available at https://github.com/yuyq18/StepTool.

cs.CL↗

arXiv2025

ROGRAG: A Robustly Optimized GraphRAG Framework

Large language models (LLMs) commonly struggle with specialized or emerging topics which are rarely seen in the training corpus. Graph-based retrieval-augmented generation (GraphRAG) addresses this by structuring domain knowledge as a graph for dynamic retrieval. However, existing pipelines involve complex engineering workflows, making it difficult to isolate the impact of individual components. It is also challenging to evaluate the retrieval effectiveness due to the overlap between the pretraining and evaluation datasets. In this work, we introduce ROGRAG, a Robustly Optimized GraphRAG framework. Specifically, we propose a multi-stage retrieval mechanism that integrates dual-level with logic form retrieval methods to improve retrieval robustness without increasing computational cost. To further refine the system, we incorporate various result verification methods and adopt an incremental database construction approach. Through extensive ablation experiments, we rigorously assess the effectiveness of each component. Our implementation includes comparative experiments on SeedBench, where Qwen2.5-7B-Instruct initially underperformed. ROGRAG significantly improves the score from 60.0% to 75.0% and outperforms mainstream methods. Experiments on domain-specific datasets reveal that dual-level retrieval enhances fuzzy matching, while logic form retrieval improves structured reasoning, highlighting the importance of multi-stage retrieval.ROGRAG is released as an open-source resource and supports installation with pip.

cs.IR↗

arXiv2025

SeedBench: A Multi-task Benchmark for Evaluating Large Language Models in Seed Science

Seed science is essential for modern agriculture, directly influencing crop yields and global food security. However, challenges such as interdisciplinary complexity and high costs with limited returns hinder progress, leading to a shortage of experts and insufficient technological support. While large language models (LLMs) have shown promise across various fields, their application in seed science remains limited due to the scarcity of digital resources, complex gene-trait relationships, and the lack of standardized benchmarks. To address this gap, we introduce SeedBench -- the first multi-task benchmark specifically designed for seed science. Developed in collaboration with domain experts, SeedBench focuses on seed breeding and simulates key aspects of modern breeding processes. We conduct a comprehensive evaluation of 26 leading LLMs, encompassing proprietary, open-source, and domain-specific fine-tuned models. Our findings not only highlight the substantial gaps between the power of LLMs and the real-world seed science problems, but also make a foundational step for research on LLMs for seed design.

cs.CL↗

arXiv2025

Scalable Multi-Agent Reinforcement Learning for Residential Load Scheduling under Data Governance

As a data-driven approach, multi-agent reinforcement learning (MARL) has made remarkable advances in solving cooperative residential load scheduling problems. However, centralized training, the most common paradigm for MARL, limits large-scale deployment in communication-constrained cloud-edge environments. As a remedy, distributed training shows unparalleled advantages in real-world applications but still faces challenge with system scalability, e.g., the high cost of communication overhead during coordinating individual agents, and needs to comply with data governance in terms of privacy. In this work, we propose a novel MARL solution to address these two practical issues. Our proposed approach is based on actor-critic methods, where the global critic is a learned function of individual critics computed solely based on local observations of households. This scheme preserves household privacy completely and significantly reduces communication cost. Simulation experiments demonstrate that the proposed framework achieves comparable performance to the state-of-the-art actor-critic framework without data governance and communication constraints.

cs.MA↗

arXiv2024

MACRec: a Multi-Agent Collaboration Framework for Recommendation

LLM-based agents have gained considerable attention for their decision-making skills and ability to handle complex tasks. Recognizing the current gap in leveraging agent capabilities for multi-agent collaboration in recommendation systems, we introduce MACRec, a novel framework designed to enhance recommendation systems through multi-agent collaboration. Unlike existing work on using agents for user/item simulation, we aim to deploy multi-agents to tackle recommendation tasks directly. In our framework, recommendation tasks are addressed through the collaborative efforts of various specialized agents, including Manager, User/Item Analyst, Reflector, Searcher, and Task Interpreter, with different working flows. Furthermore, we provide application examples of how developers can easily use MACRec on various recommendation tasks, including rating prediction, sequential recommendation, conversational recommendation, and explanation generation of recommendation results. The framework and demonstration video are publicly available at https://github.com/wzf2000/MACRec.

cs.IR↗

arXiv2024

An Embarrassingly Simple Approach to Enhance Transformer Performance in Genomic Selection for Crop Breeding

Genomic selection (GS), as a critical crop breeding strategy, plays a key role in enhancing food production and addressing the global hunger crisis. The predominant approaches in GS currently revolve around employing statistical methods for prediction. However, statistical methods often come with two main limitations: strong statistical priors and linear assumptions. A recent trend is to capture the non-linear relationships between markers by deep learning. However, as crop datasets are commonly long sequences with limited samples, the robustness of deep learning models, especially Transformers, remains a challenge. In this work, to unleash the unexplored potential of attention mechanism for the task of interest, we propose a simple yet effective Transformer-based framework that enables end-to-end training of the whole sequence. Via experiments on rice3k and wheat3k datasets, we show that, with simple tricks such as k-mer tokenization and random masking, Transformer can achieve overall superior performance against seminal methods on GS tasks of interest.

cs.LG↗

arXiv2024

To Recommend or Not: Recommendability Identification in Conversations with Pre-trained Language Models

Most current recommender systems primarily focus on what to recommend, assuming users always require personalized recommendations. However, with the widely spread of ChatGPT and other chatbots, a more crucial problem in the context of conversational systems is how to minimize user disruption when we provide recommendation services for users. While previous research has extensively explored different user intents in dialogue systems, fewer efforts are made to investigate whether recommendations should be provided. In this paper, we formally define the recommendability identification problem, which aims to determine whether recommendations are necessary in a specific scenario. First, we propose and define the recommendability identification task, which investigates the need for recommendations in the current conversational context. A new dataset is constructed. Subsequently, we discuss and evaluate the feasibility of leveraging pre-trained language models (PLMs) for recommendability identification. Finally, through comparative experiments, we demonstrate that directly employing PLMs with zero-shot results falls short of meeting the task requirements. Besides, fine-tuning or utilizing soft prompt techniques yields comparable results to traditional classification methods. Our work is the first to study recommendability before recommendation and provides preliminary ways to make it a fundamental component of the future recommendation system.

cs.IR↗

arXiv2023

Intent-aware Ranking Ensemble for Personalized Recommendation

Ranking ensemble is a critical component in real recommender systems. When a user visits a platform, the system will prepare several item lists, each of which is generally from a single behavior objective recommendation model. As multiple behavior intents, e.g., both clicking and buying some specific item category, are commonly concurrent in a user visit, it is necessary to integrate multiple single-objective ranking lists into one. However, previous work on rank aggregation mainly focused on fusing homogeneous item lists with the same objective while ignoring ensemble of heterogeneous lists ranked with different objectives with various user intents. In this paper, we treat a user's possible behaviors and the potential interacting item categories as the user's intent. And we aim to study how to fuse candidate item lists generated from different objectives aware of user intents. To address such a task, we propose an Intent-aware ranking Ensemble Learning~(IntEL) model to fuse multiple single-objective item lists with various user intents, in which item-level personalized weights are learned. Furthermore, we theoretically prove the effectiveness of IntEL with point-wise, pair-wise, and list-wise loss functions via error-ambiguity decomposition. Experiments on two large-scale real-world datasets also show significant improvements of IntEL on multiple behavior objectives simultaneously compared to previous ranking ensemble models.

cs.IR↗