arXiv subjects

Kaigui Bian

Publications and source records attributed to Kaigui Bian.

At least 19 recordsLinked to original sources

arXiv2025

Multi-Docker-Eval: A `Shovel of the Gold Rush' Benchmark on Automatic Environment Building for Software Engineering

Automated environment configuration is a critical bottleneck in scaling software engineering (SWE) automation. To provide a reliable evaluation standard for this task, we present Multi-Docker-Eval benchmark. It includes 40 real-world repositories spanning 9 programming languages and measures both success in achieving executable states and efficiency under realistic constraints. Our extensive evaluation of state-of-the-art LLMs and agent frameworks reveals key insights: (1) the overall success rate of current models is low (F2P at most 37.7%), with environment construction being the primary bottleneck; (2) model size and reasoning length are not decisive factors, and open-source models like DeepSeek-V3.1 and Kimi-K2 are competitive in both efficiency and effectiveness; (3) agent framework and programming language also have significantly influence on success rate. These findings provide actionable guidelines for building scalable, fully automated SWE pipelines.

cs.SE↗

arXiv2025

Representation Quantization for Collaborative Filtering Augmentation

As the core algorithm in recommendation systems, collaborative filtering (CF) algorithms inevitably face the problem of data sparsity. Since CF captures similar users and items for recommendations, it is effective to augment the lacking user-user and item-item homogeneous linkages. However, existing methods are typically limited to connecting through overlapping interacted neighbors or through similar attributes and contents. These approaches are constrained by coarse-grained, sparse attributes and fail to effectively extract behavioral characteristics jointly from interaction sequences and attributes. To address these challenges, we propose a novel two-stage collaborative recommendation algorithm, DQRec: Decomposition-based Quantized Variational AutoEncoder (DQ-VAE) for Recommendation. DQRec augments features and homogeneous linkages by extracting the behavior characteristics jointly from interaction sequences and attributes, namely patterns, such as user multi-aspect interests. Inspired by vector quantization (VQ) technology, we propose a new VQ algorithm, DQ-VAE, which decomposes the pre-trained representation embeddings into distinct dimensions, and quantize them to generates semantic IDs. We utilize the generated semantic IDs as the extracted patterns mentioned above. By integrating these semantic ID patterns into the recommendation process through feature and linkage augmentation, the system enriches both latent and explicit user and item features, identifies pattern-similar neighbors, and thereby improves the efficiency of information diffusion. Experimental comparisons with baselines across multiple datasets demonstrate the superior performance of the proposed DQRec method.

cs.IR↗

arXiv2025

Towards Comprehensible Recommendation with Large Language Model Fine-tuning

Recommender systems have become increasingly ubiquitous in daily life. While traditional recommendation approaches primarily rely on ID-based representations or item-side content features, they often fall short in capturing the underlying semantics aligned with user preferences (e.g., recommendation reasons for items), leading to a semantic-collaborative gap. Recently emerged LLM-based feature extraction approaches also face a key challenge: how to ensure that LLMs possess recommendation-aligned reasoning capabilities and can generate accurate, personalized reasons to mitigate the semantic-collaborative gap. To address these issues, we propose a novel Content Understanding from a Collaborative Perspective framework (CURec), which generates collaborative-aligned content features for more comprehensive recommendations. CURec first adapts the LLM with recommendation objectives through post-training, equipping it with instruction-following and chain-of-thought reasoning capabilities. Next, we use the LLM-generated features to train the recommender model. Then, we treat the user-item prediction scores produced by the trained recommender model as reward signals to guide RL-based refinement of the LLM, thereby improving the quality of generated reasons and pattern. The corrected reasons are directly integrated into the recommender model to enhance comprehensibility and recommendation performance. Extensive experiments on public benchmarks demonstrate the superiority of CURec over existing methods.

cs.IR↗

arXiv2025

KnowMap: Efficient Knowledge-Driven Task Adaptation for LLMs

While Large Language Models (LLMs) possess significant capabilities in open-world agent tasks, they also face challenges in rapidly adapting to new, specialized tasks due to their reliance on static pre-trained knowledge. Traditional methods such as fine-tuning are often costly, data-intensive, and may lead to "catastrophic forgetting." Therefore, we present KnowMap, a novel approach that dynamically constructs a knowledge base from environmental and experiential data. KnowMap fine-tunes a small knowledge-embedding model to equip a larger LLM with valuable task-specific knowledge. Our experiments on the ScienceWorld benchmark demonstrate 17.71% improvement for the performance of gpt-4-turbo model. KnowMap not only provides an efficient and effective means for LLM task-adapting, but also highlights how integrating environmental and experiential knowledge can enhance LLMs' reasoning capabilities.

cs.CL↗

arXiv2025

PLD: A Choice-Theoretic List-Wise Knowledge Distillation

Knowledge distillation is a model compression technique in which a compact "student" network is trained to replicate the predictive behavior of a larger "teacher" network. In logit-based knowledge distillation, it has become the de facto approach to augment cross-entropy with a distillation term. Typically, this term is either a KL divergence that matches marginal probabilities or a correlation-based loss that captures intra- and inter-class relationships. In every case, it acts as an additional term to cross-entropy. This term has its own weight, which must be carefully tuned. In this paper, we adopt a choice-theoretic perspective and recast knowledge distillation under the Plackett-Luce model by interpreting teacher logits as "worth" scores. We introduce "Plackett-Luce Distillation (PLD)", a weighted list-wise ranking loss. In PLD, the teacher model transfers knowledge of its full ranking of classes, weighting each ranked choice by its own confidence. PLD directly optimizes a single "teacher-optimal" ranking. The true label is placed first, followed by the remaining classes in descending teacher confidence. This process yields a convex and translation-invariant surrogate that subsumes weighted cross-entropy. Empirically, across CIFAR-100, ImageNet-1K, and MS-COCO, PLD achieves consistent gains across diverse architectures and distillation objectives, including divergence-based, correlation-based, and feature-based methods, in both homogeneous and heterogeneous teacher-student pairs.

cs.LG↗

arXiv2024

Prompt Tuning for Item Cold-start Recommendation

The item cold-start problem is crucial for online recommender systems, as the success of the cold-start phase determines whether items can transition into popular ones. Prompt learning, a powerful technique used in natural language processing (NLP) to address zero- or few-shot problems, has been adapted for recommender systems to tackle similar challenges. However, existing methods typically rely on content-based properties or text descriptions for prompting, which we argue may be suboptimal for cold-start recommendations due to 1) semantic gaps with recommender tasks, 2) model bias caused by warm-up items contribute most of the positive feedback to the model, which is the core of the cold-start problem that hinders the recommender quality on cold-start items. We propose to leverage high-value positive feedback, termed pinnacle feedback as prompt information, to simultaneously resolve the above two problems. We experimentally prove that compared to the content description proposed in existing works, the positive feedback is more suitable to serve as prompt information by bridging the semantic gaps. Besides, we propose item-wise personalized prompt networks to encode pinnaclce feedback to relieve the model bias by the positive feedback dominance problem. Extensive experiments on four real-world datasets demonstrate the superiority of our model over state-of-the-art methods. Moreover, PROMO has been successfully deployed on a popular short-video sharing platform, a billion-user scale commercial short-video application, achieving remarkable performance gains across various commercial metrics within cold-start scenarios

cs.IR↗

arXiv2024

Online Item Cold-Start Recommendation with Popularity-Aware Meta-Learning

With the rise of e-commerce and short videos, online recommender systems that can capture users' interests and update new items in real-time play an increasingly important role. In both online and offline recommendation systems, the cold-start problem caused by interaction sparsity has been impacting the effectiveness of recommendations for cold-start items. Many cold-start scheme based on fine-tuning or knowledge transferring shows excellent performance on offline recommendation. Yet, these schemes are infeasible for online recommendation on streaming data pipelines due to different training method, computational overhead and time constraints. Inspired by the above questions, we propose a model-agnostic recommendation algorithm called Popularity-Aware Meta-learning (PAM), to address the item cold-start problem under streaming data settings. PAM divides the incoming data into different meta-learning tasks by predefined item popularity thresholds. The model can distinguish and reweight behavior-related and content-related features in each task based on their different roles in different popularity levels, thus adapting to recommendations for cold-start samples. These task-fixing design significantly reduces additional computation and storage costs compared to offline methods. Furthermore, PAM also introduced data augmentation and an additional self-supervised loss specifically designed for low-popularity tasks, leveraging insights from high-popularity samples. This approach effectively mitigates the issue of inadequate supervision due to the scarcity of cold-start samples. Experimental results across multiple public datasets demonstrate the superiority of our approach over other baseline methods in addressing cold-start challenges in online streaming data scenarios.

cs.IR↗

arXiv2024

PeriGuru: A Peripheral Robotic Mobile App Operation Assistant based on GUI Image Understanding and Prompting with LLM

Smartphones have significantly enhanced our daily learning, communication, and entertainment, becoming an essential component of modern life. However, certain populations, including the elderly and individuals with disabilities, encounter challenges in utilizing smartphones, thus necessitating mobile app operation assistants, a.k.a. mobile app agent. With considerations for privacy, permissions, and cross-platform compatibility issues, we endeavor to devise and develop PeriGuru in this work, a peripheral robotic mobile app operation assistant based on GUI image understanding and prompting with Large Language Model (LLM). PeriGuru leverages a suite of computer vision techniques to analyze GUI screenshot images and employs LLM to inform action decisions, which are then executed by robotic arms. PeriGuru achieves a success rate of 81.94% on the test task set, which surpasses by more than double the method without PeriGuru's GUI image interpreting and prompting design. Our code is available on https://github.com/Z2sJ4t/PeriGuru.

cs.RO↗

arXiv2024

Meta-Backscatter: A New ISAC Paradigm for Battery-Free Internet of Things

The meta-material sensor has been regarded as a next-generation sensing technology for the battery-free Internet of Things (IoT) due to its battery-free characteristic and improved sensing performance. The meta-material sensors function as backscatter tags that change their reflection coefficients with the conditions of sensing targets such as temperature and gas concentration, allowing transceivers to perform sensing by analyzing the reflected signals from the sensors. Simultaneously, the sensors also function as environmental scatterers, creating additional signal paths to enhance communication performance. Therefore, the meta-material sensor potentially provides a new paradigm of Integrated Sensing and Communication (ISAC) for the battery-free IoT system. In this article, we first propose a Meta-Backscatter system that utilizes meta-material sensors to achieve diverse sensing functionalities and improved communication performance. We begin with the introduction of the metamaterial sensor and further elaborate on the Meta-Backscatter system. Subsequently, we present optimization strategies for meta-material sensors, transmitters, and receivers to strike a balance between sensing and communication. Furthermore, this article provides a case study of the system and examines the feasibility and trade-off through the simulation results. Finally, potential extensions of the system and their related research challenges are addressed.

eess.SY↗

arXiv2023

Towards Ubiquitous Positioning by Leveraging Reconfigurable Intelligent Surface

The received signal strength (RSS) based technique is widely utilized for ubiquitous positioning due to its advantage of simple implementability. However, its accuracy is limited because the RSS values of adjacent locations can be very difficult to distinguish. Against this background, we propose the novel RSS-based positioning scheme enabled by reconfigurable intelligent surface (RIS). By modifying the reflection coefficient of the RIS, the propagation channels are programmed in such a way that the differences between the RSS values of adjacent locations can be enlarged to improve the positioning accuracy. New challenge lies in the selection of suitable reflection coefficients for high-accuracy positioning. To tackle this challenge, we formulate the RIS-aided positioning problem and design an iterative algorithm to solve the problem. The effectiveness of the proposed positioning scheme is validated through simulations.

eess.SP↗

arXiv2023

MetaRadar: Indoor Localization by Reconfigurable Metamaterials

Indoor localization has drawn much attention owing to its potential for supporting location based services. Among various indoor localization techniques, the received signal strength (RSS) based technique is widely researched. However, in conventional RSS based systems where the radio environment is unconfigurable, adjacent locations may have similar RSS values, which limits the localization precision. In this paper, we present MetaRadar, which explores reconfigurable radio reflection with a surface/plane made of metamaterial units for multi-user localization. By changing the reflectivity of metamaterial, MetaRadar modifies the radio channels at different locations, and improves localization accuracy by making RSS values at adjacent locations have significant differences. However, in MetaRadar, it is challenging to build radio maps for all the radio environments generated by metamaterial units and select suitable maps from all the possible maps to realize a high accuracy localization. To tackle this challenge, we propose a compressive construction technique which can predict all the possible radio maps, and propose a configuration optimization algorithm to select favorable metamaterial reflectivities and the corresponding radio maps. The experimental results show a significant improvement from a decimeter-level localization error in the traditional RSS-based systems to a centimeter-level one in MetaRadar.

eess.SP↗

arXiv2023

MetaLocalization: Reconfigurable Intelligent Surface Aided Multi-user Wireless Indoor Localization

The received signal strength (RSS) based technique is extensively utilized for localization in the indoor environments. Since the RSS values of neighboring locations may be similar, the localization accuracy of the RSS based technique is limited. To tackle this problem, in this paper, we propose to utilize reconfigurable intelligent surface (RIS) for the RSS based multi-user localization. As the RIS is able to customize the radio channels by adjusting the phase shifts of the signals reflected at the surface, the localization accuracy in the RIS aided scheme can be improved by choosing the proper phase shifts with significant differences of RSS values among adjacent locations. However, it is challenging to select the optimal phase shifts because the decision function for location estimation and the phase shifts are coupled. To tackle this challenge, we formulate the optimization problem for the RIS-aided localization, derive the optimal decision function, and design the phase shift optimization (PSO) algorithm to solve the formulated problem efficiently. Analysis of the proposed RIS aided technique is provided, and the effectiveness is validated through simulation.

eess.SP↗

arXiv2023

MetaRadar: Multi-target Detection for Reconfigurable Intelligent Surface Aided Radar Systems

As a widely used localization and sensing technique, radars will play an important role in future wireless networks. However, the wireless channels between the radar and the targets are passively adopted by traditional radars, which limits the performance of target detection. To address this issue, we propose to use the reconfigurable intelligent surface (RIS) to improve the detection accuracy of radar systems due to its capability to customize channel conditions by adjusting its phase shifts, which is referred to as MetaRadar. In such a system, it is challenging to jointly optimize both radar waveforms and RIS phase shifts in order to improve the multi-target detection performance. To tackle this challenge, we design a waveform and phase shift optimization (WPSO) algorithm to effectively solve the multi-target detection problem, and also analyze the performance of the proposed MetaRadar scheme theoretically. Simulation results show that the detection performance of the MetaRadar scheme is significantly better than that of the traditional radar schemes.

eess.SP↗

arXiv2022

Federated Neural Architecture Search

To preserve user privacy while enabling mobile intelligence, techniques have been proposed to train deep neural networks on decentralized data. However, training over decentralized data makes the design of neural architecture quite difficult as it already was. Such difficulty is further amplified when designing and deploying different neural architectures for heterogeneous mobile platforms. In this work, we propose an automatic neural architecture search into the decentralized training, as a new DNN training paradigm called Federated Neural Architecture Search, namely federated NAS. To deal with the primary challenge of limited on-client computational and communication resources, we present FedNAS, a highly optimized framework for efficient federated NAS. FedNAS fully exploits the key opportunity of insufficient model candidate re-training during the architecture search process, and incorporates three key optimizations: parallel candidates training on partial clients, early dropping candidates with inferior performance, and dynamic round numbers. Tested on large-scale datasets and typical CNN architectures, FedNAS achieves comparable model accuracy as state-of-the-art NAS algorithm that trains models with centralized data, and also reduces the client cost by up to two orders of magnitude compared to a straightforward design of federated NAS.

cs.LG↗

arXiv2022

AMEIR: Automatic Behavior Modeling, Interaction Exploration and MLP Investigation in the Recommender System

Recently, deep learning models have been widely spread in the industrial recommender systems and boosted the recommendation quality. Though having achieved remarkable success, the design of task-aware recommender systems usually requires manual feature engineering and architecture engineering from domain experts. To relieve those human efforts, we explore the potential of neural architecture search (NAS) and introduce AMEIR for Automatic behavior Modeling, interaction Exploration and multi-layer perceptron (MLP) Investigation in the Recommender system. The core contributions of AMEIR are the three-stage search space and the tailored three-step searching pipeline. Specifically, AMEIR divides the complete recommendation models into three stages of behavior modeling, interaction exploration, MLP aggregation, and introduces a novel search space containing three tailored subspaces that cover most of the existing methods and thus allow for searching better models. To find the ideal architecture efficiently and effectively, AMEIR realizes the one-shot random search in recommendation progressively on the three stages and assembles the search results as the final outcome. Further analysis reveals that AMEIR's search space could cover most of the representative recommendation models, which demonstrates the universality of our design. The extensive experiments over various scenarios reveal that AMEIR outperforms competitive baselines of elaborate manual design and leading algorithmic complex NAS methods with lower model complexity and comparable time cost, indicating efficacy, efficiency and robustness of the proposed method.

cs.LG↗

arXiv2022

DAIS: Automatic Channel Pruning via Differentiable Annealing Indicator Search

The convolutional neural network has achieved great success in fulfilling computer vision tasks despite large computation overhead against efficient deployment. Structured (channel) pruning is usually applied to reduce the model redundancy while preserving the network structure, such that the pruned network can be easily deployed in practice. However, existing structured pruning methods require hand-crafted rules which may lead to tremendous pruning space. In this paper, we introduce Differentiable Annealing Indicator Search (DAIS) that leverages the strength of neural architecture search in the channel pruning and automatically searches for the effective pruned model with given constraints on computation overhead. Specifically, DAIS relaxes the binarized channel indicators to be continuous and then jointly learns both indicators and model parameters via bi-level optimization. To bridge the non-negligible discrepancy between the continuous model and the target binarized model, DAIS proposes an annealing-based procedure to steer the indicator convergence towards binarized states. Moreover, DAIS designs various regularizations based on a priori structural knowledge to control the pruning sparsity and to improve model performance. Experimental results show that DAIS outperforms state-of-the-art pruning methods on CIFAR-10, CIFAR-100, and ImageNet.

cs.CV↗

arXiv2022

Towards Ubiquitous Sensing and Localization With Reconfigurable Intelligent Surfaces

In future cellular systems, wireless localization and sensing functions will be built-in for specific applications, e.g., navigation, transportation, and healthcare, and to support flexible and seamless connectivity. Driven by this trend, the need rises for fine-resolution sensing solutions and cm-level localization accuracy, while the accuracy of current wireless systems is limited by the quality of the propagation environment. Recently, with the development of new materials, reconfigurable intelligent surfaces (RISs) provide an opportunity to reshape and control the electromagnetic characteristics of the environment, which can be utilized to improve the performance of wireless sensing and localization. In this tutorial, we will first review the background and motivation to utilize wireless signals for sensing and localization. Next, we introduce how to incorporate RIS into applications of sensing and localization, including key challenges and enabling techniques, and then some case studies will be presented. Finally, future research directions will also be discussed.

cs.IT↗

arXiv2021

Meta-material Sensors based Internet of Things for 6G Communications

In the coming 6G communications, the internet of things (IoT) serves as a key enabler to collect environmental information and is expected to achieve ubiquitous deployment. However, it is challenging for traditional IoT sensors to meet this expectation because of their requirements of power supplies and frequent maintenance, which are due to their power-demanding sense and transmit modules. To address this challenge, we propose a meta-IoT sensing system, where the IoT sensors are based on specially designed meta-materials. The meta-IoT sensors achieve simultaneous sensing and transmission by physical reflection and require no power supplies. In order to design a meta-IoT sensing system with optimal sensing accuracy, we jointly consider the sensing and transmission of meta-IoT sensors and propose efficient algorithms to optimize the meta-IoT structure and the sensing function at the receiver. As an example, we apply the meta-IoT system to sensing environmental temperature and humidity levels. Simulation results show that by using the proposed algorithm, the sensing accuracy can be largely increased.

eess.SP↗