arXiv subjects

Xueyang Feng

Publications and source records attributed to Xueyang Feng.

14 recordsLinked to original sources

arXiv2026

Towards Fast Domain Adaptation and Fine-Grained User Simulation for Evaluating Conversational Recommender Systems

Conversational Recommender Systems (CRSs) enhance user experience through multi-turn interactions, yet evaluating their performance remains challenging. While Large Language Model (LLM) based user simulators are effective, they suffer from three key limitations: (1) Lack of Domain Adaptability: Reliance on fixed prompts and predefined action spaces hinders transfer to novel domains; (2) Limited User Modeling: Inability to accurately replicate subtle linguistic styles and dynamic preferences; (3) Insufficient Evaluation Validity: Existing simulators fail to adequately assess fundamental capabilities and system robustness. To overcome these, we propose AdaptSim, an Adaptive domain and automatic prompt tuning User Simulator. AdaptSim offers an efficient framework for evaluating CRSs by enabling realistic behavior modeling and diverse style generation. It leverages automatic prompt generation and an open action mechanism to reduce manual effort and improve cross-domain flexibility. For response generation, we employ controlled text generation with a "think-then-respond" strategy for fine-grained control over language style. For CRS evaluation, AdaptSim incorporates a novel Breadth-First Search (BFS)-based, turn-level pairwise comparison framework for comprehensive assessment. Extensive experiments across three domains and four LLMs demonstrate that AdaptSim generates realistic dialogues, enabling a highly effective and reliable evaluation of CRS capabilities and robustness.

cs.IR↗

arXiv2026

Topological and morphological signatures of disorder in a self-assembled, soft matter sponge network

Many soft matter systems exhibit ordered, polycontinuous network morphologies, such as the cubic (double) gyroid or diamond, as well as disordered network morphologies known generically as ``random sponges". While presumed to share similar local packing geometry, the structural relationship between these ordered and disordered network morphologies has remained obscure. We use slice and view scanning electron microscopy to analyze and compare multi-scale morphological features of an ordered double-gyroid morphology to the amorphous sponge morphology formed in the same block copolymer sample. We find that node valence of the minority component network of the sponge is mostly gyroidal (trivalent), with a small fraction of diamond-like (tetravalent) connections. We analyze mesoatoms -- space-filling volumes occupied by chains around each network node -- finding significant differences in shape and size between ordered and amorphous regions. Local block thickness and inter-domain curvature within mesoatomic units of the disordered sponge exhibits a surprisingly similar degree of dispersity to the ordered double-gyroid. The mean differences in local packing geometry derive from topological distinction: loops of the minority networks of the ordered double-gyroid are intercatenated, while loops of the disordered sponge are not. In this way, the sponge may be viewed as disordered variant of a single-gyroidal morphology. We exploit these topological differences to demarcate the boundary region between ordered and disordered networks and highlight modulations of the mesoatom motifs at the boundary. These observations point to new questions about potential metastability of disordered networks and their possible role as kinetic precursors to long-range ordered network morphologies.

cond-mat.soft↗

arXiv2026

How Does Personalized Memory Shape LLM Behavior? Benchmarking Rational Preference Utilization in Personalized Assistants

Large language model (LLM)-powered assistants have recently integrated memory mechanisms that record user preferences, leading to more personalized and user-aligned responses. However, irrelevant personalized memories are often introduced into the context, interfering with the LLM's intent understanding. To comprehensively investigate the dual effects of personalization, we develop RPEval, a benchmark comprising a personalized intent reasoning dataset and a multi-granularity evaluation protocol. RPEval reveals the widespread phenomenon of irrational personalization in existing LLMs and, through error pattern analysis, illustrates its negative impact on user experience. Finally, we introduce RP-Reasoner, which treats memory utilization as a pragmatic reasoning process, enabling the selective integration of personalized information. Experimental results demonstrate that our method significantly outperforms carefully designed baselines on RPEval, and resolves 80% of the bad cases observed in a large-scale commercial personalized assistant, highlighting the potential of pragmatic reasoning to mitigate irrational personalization. Our benchmark is publicly available at https://github.com/XueyangFeng/RPEval.

cs.CL↗

arXiv2025

Interactive Recommendation Agent with Active User Commands

Traditional recommender systems rely on passive feedback mechanisms that limit users to simple choices such as like and dislike. However, these coarse-grained signals fail to capture users' nuanced behavior motivations and intentions. In turn, current systems cannot also distinguish which specific item attributes drive user satisfaction or dissatisfaction, resulting in inaccurate preference modeling. These fundamental limitations create a persistent gap between user intentions and system interpretations, ultimately undermining user satisfaction and harming system effectiveness. To address these limitations, we introduce the Interactive Recommendation Feed (IRF), a pioneering paradigm that enables natural language commands within mainstream recommendation feeds. Unlike traditional systems that confine users to passive implicit behavioral influence, IRF empowers active explicit control over recommendation policies through real-time linguistic commands. To support this paradigm, we develop RecBot, a dual-agent architecture where a Parser Agent transforms linguistic expressions into structured preferences and a Planner Agent dynamically orchestrates adaptive tool chains for on-the-fly policy adjustment. To enable practical deployment, we employ simulation-augmented knowledge distillation to achieve efficient performance while maintaining strong reasoning capabilities. Through extensive offline and long-term online experiments, RecBot shows significant improvements in both user satisfaction and business outcomes.

cs.IR↗

arXiv2025

RecUserSim: A Realistic and Diverse User Simulator for Evaluating Conversational Recommender Systems

Conversational recommender systems (CRS) enhance user experience through multi-turn interactions, yet evaluating CRS remains challenging. User simulators can provide comprehensive evaluations through interactions with CRS, but building realistic and diverse simulators is difficult. While recent work leverages large language models (LLMs) to simulate user interactions, they still fall short in emulating individual real users across diverse scenarios and lack explicit rating mechanisms for quantitative evaluation. To address these gaps, we propose RecUserSim, an LLM agent-based user simulator with enhanced simulation realism and diversity while providing explicit scores. RecUserSim features several key modules: a profile module for defining realistic and diverse user personas, a memory module for tracking interaction history and discovering unknown preferences, and a core action module inspired by Bounded Rationality theory that enables nuanced decision-making while generating more fine-grained actions and personalized responses. To further enhance output control, a refinement module is designed to fine-tune final responses. Experiments demonstrate that RecUserSim generates diverse, controllable outputs and produces realistic, high-quality dialogues, even with smaller base LLMs. The ratings generated by RecUserSim show high consistency across different base LLMs, highlighting its effectiveness for CRS evaluation.

cs.HC↗

arXiv2025

Expectation Confirmation Preference Optimization for Multi-Turn Conversational Recommendation Agent

Recent advancements in Large Language Models (LLMs) have significantly propelled the development of Conversational Recommendation Agents (CRAs). However, these agents often generate short-sighted responses that fail to sustain user guidance and meet expectations. Although preference optimization has proven effective in aligning LLMs with user expectations, it remains costly and performs poorly in multi-turn dialogue. To address this challenge, we introduce a novel multi-turn preference optimization (MTPO) paradigm ECPO, which leverages Expectation Confirmation Theory to explicitly model the evolution of user satisfaction throughout multi-turn dialogues, uncovering the underlying causes of dissatisfaction. These causes can be utilized to support targeted optimization of unsatisfactory responses, thereby achieving turn-level preference optimization. ECPO ingeniously eliminates the significant sampling overhead of existing MTPO methods while ensuring the optimization process drives meaningful improvements. To support ECPO, we introduce an LLM-based user simulator, AILO, to simulate user feedback and perform expectation confirmation during conversational recommendations. Experimental results show that ECPO significantly enhances CRA's interaction capabilities, delivering notable improvements in both efficiency and effectiveness over existing MTPO methods.

cs.CL↗

arXiv2025

Explainable Recommendation with Simulated Human Feedback

Recent advancements in explainable recommendation have greatly bolstered user experience by elucidating the decision-making rationale. However, the existing methods actually fail to provide effective feedback signals for potentially better or worse generated explanations due to their reliance on traditional supervised learning paradigms in sparse interaction data. To address these issues, we propose a novel human-like feedback-driven optimization framework. This framework employs a dynamic interactive optimization mechanism for achieving human-centered explainable requirements without incurring high labor costs. Specifically, we propose to utilize large language models (LLMs) as human simulators to predict human-like feedback for guiding the learning process. To enable the LLMs to deeply understand the task essence and meet user's diverse personalized requirements, we introduce a human-induced customized reward scoring method, which helps stimulate the language understanding and logical reasoning capabilities of LLMs. Furthermore, considering the potential conflicts between different perspectives of explanation quality, we introduce a principled Pareto optimization that transforms the multi-perspective quality enhancement task into a multi-objective optimization problem for improving explanation performance. At last, to achieve efficient model training, we design an off-policy optimization pipeline. By incorporating a replay buffer and addressing the data distribution biases, we can effectively improve data utilization and enhance model generality. Extensive experiments on four datasets demonstrate the superiority of our approach.

cs.IR↗

arXiv2025

Improving Retrospective Language Agents via Joint Policy Gradient Optimization

In recent research advancements within the community, large language models (LLMs) have sparked great interest in creating autonomous agents. However, current prompt-based agents often heavily rely on large-scale LLMs. Meanwhile, although fine-tuning methods significantly enhance the capabilities of smaller LLMs, the fine-tuned agents often lack the potential for self-reflection and self-improvement. To address these challenges, we introduce a novel agent framework named RetroAct, which is a framework that jointly optimizes both task-planning and self-reflective evolution capabilities in language agents. Specifically, we develop a two-stage joint optimization process that integrates imitation learning and reinforcement learning, and design an off-policy joint policy gradient optimization algorithm with imitation learning regularization to enhance the data efficiency and training stability in agent tasks. RetroAct significantly improves the performance of open-source models, reduces dependency on closed-source LLMs, and enables fine-tuned agents to learn and evolve continuously. We conduct extensive experiments across various testing environments, demonstrating RetroAct has substantial improvements in task performance and decision-making processes.

cs.CL↗

arXiv2025

Enhancing Recommendation Explanations through User-Centric Refinement

Generating natural language explanations for recommendations has become increasingly important in recommender systems. Traditional approaches typically treat user reviews as ground truth for explanations and focus on improving review prediction accuracy by designing various model architectures. However, due to limitations in data scale and model capability, these explanations often fail to meet key user-centric aspects such as factuality, personalization, and sentiment coherence, significantly reducing their overall helpfulness to users. In this paper, we propose a novel paradigm that refines initial explanations generated by existing explainable recommender models during the inference stage to enhance their quality in multiple aspects. Specifically, we introduce a multi-agent collaborative refinement framework based on large language models. To ensure alignment between the refinement process and user demands, we employ a plan-then-refine pattern to perform targeted modifications. To enable continuous improvements, we design a hierarchical reflection mechanism that provides feedback on the refinement process from both strategic and content perspectives. Extensive experiments on three datasets demonstrate the effectiveness of our framework.

cs.IR↗

arXiv2024

Large Language Model-based Human-Agent Collaboration for Complex Task Solving

In recent developments within the research community, the integration of Large Language Models (LLMs) in creating fully autonomous agents has garnered significant interest. Despite this, LLM-based agents frequently demonstrate notable shortcomings in adjusting to dynamic environments and fully grasping human needs. In this work, we introduce the problem of LLM-based human-agent collaboration for complex task-solving, exploring their synergistic potential. In addition, we propose a Reinforcement Learning-based Human-Agent Collaboration method, ReHAC. This approach includes a policy model designed to determine the most opportune stages for human intervention within the task-solving process. We construct a human-agent collaboration dataset to train this policy model in an offline reinforcement learning environment. Our validation tests confirm the model's effectiveness. The results demonstrate that the synergistic efforts of humans and LLM-based agents significantly improve performance in complex tasks, primarily through well-planned, limited human intervention. Datasets and code are available at: https://github.com/XueyangFeng/ReHAC.

cs.CL↗

arXiv2023

A Survey on Large Language Model based Autonomous Agents

Autonomous agents have long been a prominent research focus in both academic and industry communities. Previous research in this field often focuses on training agents with limited knowledge within isolated environments, which diverges significantly from human learning processes, and thus makes the agents hard to achieve human-like decisions. Recently, through the acquisition of vast amounts of web knowledge, large language models (LLMs) have demonstrated remarkable potential in achieving human-level intelligence. This has sparked an upsurge in studies investigating LLM-based autonomous agents. In this paper, we present a comprehensive survey of these studies, delivering a systematic review of the field of LLM-based autonomous agents from a holistic perspective. More specifically, we first discuss the construction of LLM-based autonomous agents, for which we propose a unified framework that encompasses a majority of the previous work. Then, we present a comprehensive overview of the diverse applications of LLM-based autonomous agents in the fields of social science, natural science, and engineering. Finally, we delve into the evaluation strategies commonly used for LLM-based autonomous agents. Based on the previous studies, we also present several challenges and future directions in this field. To keep track of this field and continuously update our survey, we maintain a repository of relevant references at https://github.com/Paitesanshi/LLM-Agent-Survey.

cs.AI↗

arXiv2023

Fairness-aware Cross-Domain Recommendation

Cross-Domain Recommendation (CDR) is an effective way to alleviate the cold-start problem. However, previous work severely ignores fairness and bias when learning the mapping function, which is used to obtain the representations for fresh users in the target domain. To study this problem, in this paper, we propose a Fairness-aware Cross-Domain Recommendation model, called FairCDR. Our method achieves user-oriented group fairness by learning the fairness-aware mapping function. Since the overlapping data are quite limited and distributionally biased, FairCDR leverages abundant non-overlapping users and interactions to help alleviate these problems. Considering that each individual has different influence on model fairness, we propose a new reweighing method based on Influence Function (IF) to reduce unfairness while maintaining recommendation accuracy. Extensive experiments are conducted to demonstrate the effectiveness of our model.

cs.IR↗

arXiv2017

DeepMetabolism: A Deep Learning System to Predict Phenotype from Genome Sequencing

Life science is entering a new era of petabyte-level sequencing data. Converting such big data to biological insights represents a huge challenge for computational analysis. To this end, we developed DeepMetabolism, a biology-guided deep learning system to predict cell phenotypes from transcriptomics data. By integrating unsupervised pre-training with supervised training, DeepMetabolism is able to predict phenotypes with high accuracy (PCC>0.92), high speed (<30 min for >100 GB data using a single GPU), and high robustness (tolerate up to 75% noise). We envision DeepMetabolism to bridge the gap between genotype and phenotype and to serve as a springboard for applications in synthetic biology and precision medicine.

q-bio.GN↗

arXiv2017

Data-Driven Prediction of CRISPR-Based Transcription Regulation for Programmable Control of Metabolic Flux

Multiplex and multi-directional control of metabolic pathways is crucial for metabolic engineering to improve product yield of fuels, chemicals, and pharmaceuticals. To achieve this goal, artificial transcriptional regulators such as CRISPR-based transcription regulators have been developed to specifically activate or repress genes of interest. Here, we found that by deploying guide RNAs to target on DNA sites at different locations of genetic cassettes, we could use just one synthetic CRISPR-based transcriptional regulator to simultaneously activate and repress gene expressions. By using the pairwise datasets of guide RNAs and gene expressions, we developed a data-driven predictive model to rationally design this system for fine-tuning expression of target genes. We demonstrated that this system could achieve programmable control of metabolic fluxes when using yeast to produce versatile chemicals. We anticipate that this master CRISPR-based transcription regulator will be a valuable addition to the synthetic biology toolkit for metabolic engineering, speeding up the design-build-test cycle in industrial biomanufacturing as well as generating new biological insights on the fates of eukaryotic cells.

q-bio.BM↗