arXiv subjects

Qian Dong

Publications and source records attributed to Qian Dong.

At least 19 recordsLinked to original sources

arXiv2026

Separating Capability from Permission: A Governance Framework for Agentic AI Autonomy Levels

As AI systems increasingly exhibit agentic behavior, discussions of autonomy often conflate what systems are technically capable of doing with what they should be permitted to do in practice. This paper introduces a governance framework that explicitly separates Allowed Autonomy Levels (AAL), which define the degree of autonomy an AI agent is authorized to exercise given risk, oversight, and accountability considerations, from Autonomous Capability Levels (ACL), which characterize an agent's inherent technical abilities. We present a structured set of autonomy levels spanning reactive execution, decision support, supervised action, goal-directed autonomy, and delegated operational authority, and describe how control, reversibility, and accountability change as autonomy increases. To operationalize this framework, we propose a risk-aware decision process for assigning allowed autonomy, analyze how risk and accountability evolve across autonomy levels, and demonstrate its application through a deployed enterprise data engineering agent, illustrating how a system assessed at a high capability level can be deliberately constrained to a lower allowed autonomy based on risk, reversibility, and organizational readiness. By distinguishing authorization from capability, this work provides practical guidance for the design, deployment, and governance of Agentic AI systems.

cs.AI↗

arXiv2026

An Asynchronous Two-Speed Kalman Filter for Real-Time UUV Cooperative Navigation Under Acoustic Delays

In Global Navigation Satellite System (GNSS)-denied underwater environments, individual unmanned underwater vehicles (UUVs) suffer from unbounded dead-reckoning drift, making collaborative navigation (CN) crucial for accurate state estimation. However, the severe communication delay inherent in underwater acoustic channels poses serious challenges to real-time state estimation. Traditional filters, such as Extended Kalman Filters (EKFs) or Unscented Kalman Filters (UKFs), usually block the main control loop while waiting for delayed data, or effectively discard Out-of-Sequence Measurements (OOSMs), resulting in serious drift. To address this, we propose an Asynchronous Two-Speed Kalman Filter (TSKF) enhanced by a novel projection mechanism, which we term Variational History Distillation (VHD). The proposed architecture decouples the estimation process into two parallel threads: a fast-rate thread that utilizes Gaussian Process (GP) compensated dead reckoning to guarantee high-frequency real-time control, and a slow-rate thread dedicated to processing asynchronously delayed collaborative information. By introducing a Finite-Length Circular State Buffer (FLCSB), the algorithm applies delayed measurements to their corresponding historical states, and utilizes a VHD-based projection to fast-forward the correction to the current time without computationally heavy recalculations. Simulation results demonstrate that the proposed TSKF maintains a trajectory error comparable to computationally intensive batch-optimization methods under severe delays (up to 30\,s). Executing in sub-millisecond time, it significantly outperforms standard EKF/UKF. The results demonstrate an effective control, communication, and computing (3C) co-design that significantly enhances the resilience of autonomous marine automation systems.

cs.RO↗

arXiv2026

Communication Outage-Resistant UUV State Estimation: A Variational History Distillation Approach

The reliable operation of Unmanned Underwater Vehicle (UUV) clusters is highly dependent on continuous acoustic communication. However, this communication method is highly susceptible to intermittent interruptions. When communication outages occur, standard state estimators such as the Unscented Kalman Filter (UKF) will be forced to make open-loop predictions. If the environment contains unmodeled dynamic factors, such as unknown ocean currents, this estimation error will grow rapidly, which may eventually lead to mission failure. To address this critical issue, this paper proposes a Variational History Distillation (VHD) approach. VHD regards trajectory prediction as an approximate Bayesian reasoning process, which links a standard motion model based on physics with a pattern extracted directly from the past trajectory of the UUV. This is achieved by synthesizing ``virtual measurements'' distilled from historical trajectories. Recognizing that the reliability of extrapolated historical trends degrades over extended prediction horizons, an adaptive confidence mechanism is introduced. This mechanism allows the filter to gradually reduce the trust of virtual measurements as the communication outage time is extended. Extensive Monte Carlo simulations in a high-fidelity environment demonstrate that the proposed method achieves a 91% reduction in prediction Root Mean Square Error (RMSE), reducing the error from approximately 170 m to 15 m during a 40-second communication outage. These results demonstrate that VHD can maintain robust state estimation performance even under complete communication loss.

cs.RO↗

arXiv2026

IndexCache: Accelerating Sparse Attention via Cross-Layer Index Reuse

Long-context agentic workflows have emerged as a defining use case for large language models, making attention efficiency critical for both inference speed and serving cost. Sparse attention addresses this challenge effectively, and DeepSeek Sparse Attention (DSA) is a representative production-grade solution: a lightweight lightning indexer selects the top-k most relevant tokens per query, reducing core attention from $O(L^2)$ to $O(Lk)$. However, the indexer itself retains $O(L^2)$ complexity and must run independently at every layer, despite the fact that the resulting top-k selections are highly similar across consecutive layers. We present IndexCache, which exploits this cross-layer redundancy by partitioning layers into a small set of Full layers that run their own indexers and a majority of Shared layers that simply reuse the nearest Full layer's top-k indices. We propose two complementary approaches to determine and optimize this configuration. Training-free IndexCache applies a greedy search algorithm that selects which layers to retain indexers by directly minimizing language modeling loss on a calibration set, requiring no weight updates. Training-aware IndexCache introduces a multi-layer distillation loss that trains each retained indexer against the averaged attention distributions of all layers it serves, enabling even simple interleaved patterns to match full-indexer accuracy. Experimental results on a 30B DSA model show that IndexCache can remove 75% of indexer computations with negligible quality degradation, achieving up to 1.82$\times$ prefill speedup and 1.48$\times$ decode speedup compared to standard DSA. These positive results are further confirmed by our preliminary experiments on the production-scale GLM-5 model (Figure 1).

cs.CL↗

arXiv2026

GLM-5: from Vibe Coding to Agentic Engineering

We present GLM-5, a next-generation foundation model designed to transition the paradigm of vibe coding to agentic engineering. Building upon the agentic, reasoning, and coding (ARC) capabilities of its predecessor, GLM-5 adopts DSA to significantly reduce training and inference costs while maintaining long-context fidelity. To advance model alignment and autonomy, we implement a new asynchronous reinforcement learning infrastructure that drastically improves post-training efficiency by decoupling generation from training. Furthermore, we propose novel asynchronous agent RL algorithms that further improve RL quality, enabling the model to learn from complex, long-horizon interactions more effectively. Through these innovations, GLM-5 achieves state-of-the-art performance on major open benchmarks. Most critically, GLM-5 demonstrates unprecedented capability in real-world coding tasks, surpassing previous baselines in handling end-to-end software engineering challenges. Code, models, and more information are available at https://github.com/zai-org/GLM-5.

cs.LG↗

arXiv2026

LegalOne: A Family of Foundation Models for Reliable Legal Reasoning

While Large Language Models (LLMs) have demonstrated impressive general capabilities, their direct application in the legal domain is often hindered by a lack of precise domain knowledge and complexity of performing rigorous multi-step judicial reasoning. To address this gap, we present LegalOne, a family of foundational models specifically tailored for the Chinese legal domain. LegalOne is developed through a comprehensive three-phase pipeline designed to master legal reasoning. First, during mid-training phase, we propose Plasticity-Adjusted Sampling (PAS) to address the challenge of domain adaptation. This perplexity-based scheduler strikes a balance between the acquisition of new knowledge and the retention of original capabilities, effectively establishing a robust legal foundation. Second, during supervised fine-tuning, we employ Legal Agentic CoT Distillation (LEAD) to distill explicit reasoning from raw legal texts. Unlike naive distillation, LEAD utilizes an agentic workflow to convert complex judicial processes into structured reasoning trajectories, thereby enforcing factual grounding and logical rigor. Finally, we implement a Curriculum Reinforcement Learning (RL) strategy. Through a progressive reinforcement process spanning memorization, understanding, and reasoning, LegalOne evolves from simple pattern matching to autonomous and reliable legal reasoning. Experimental results demonstrate that LegalOne achieves state-of-the-art performance across a wide range of legal tasks, surpassing general-purpose LLMs with vastly larger parameter counts through enhanced knowledge density and efficiency. We publicly release the LegalOne weights and the LegalKit evaluation framework to advance the field of Legal AI, paving the way for deploying trustworthy and interpretable foundation models in high-stakes judicial applications.

cs.CL↗

arXiv2025

LangPrecip: Language-Aware Multimodal Precipitation Nowcasting

Short-term precipitation nowcasting is an inherently uncertain and under-constrained spatiotemporal forecasting problem, especially for rapidly evolving and extreme weather events. Existing generative approaches rely primarily on visual conditioning, leaving future motion weakly constrained and ambiguous. We propose a language-aware multimodal nowcasting framework(LangPrecip) that treats meteorological text as a semantic motion constraint on precipitation evolution. By formulating nowcasting as a semantically constrained trajectory generation problem under the Rectified Flow paradigm, our method enables efficient and physically consistent integration of textual and radar information in latent space.We further introduce LangPrecip-160k, a large-scale multimodal dataset with 160k paired radar sequences and motion descriptions. Experiments on Swedish and MRMS datasets show consistent improvements over state-of-the-art methods, achieving over 60 \% and 19\% gains in heavy-rainfall CSI at an 80-minute lead time.

cs.LG↗

arXiv2025

Discovery of Repeating Transitions in 16 Changing-look Active Galactic Nuclei

The repeating changing-look active galactic nuclei (RCL AGNs) exhibit multiple appearances and disappearances of broad emission lines (BELs), whose underlying mechanism remains a puzzle. Expanding the sample of RCL AGNs is valuable for constraining the transition timescale and probing the accretion physics driving CL behaviors. This study aims to identify RCL AGNs using the multi-epoch spectroscopic data of confirmed CL AGNs from the Sloan Digital Sky Survey, Large Sky Area Multi-Object Fiber Spectroscopic Telescope, and Dark Energy Spectroscopic Instrument, supplemented with mid-infrared (MIR) light curves. Through selection criteria and visual inspection, we identify 22 RCL AGNs among 299 CL AGNs, corresponding to an occurrence rate of about 7\%, indicating that repeated transitions are not extremely rare in CL AGNs. Among the 22 RCL AGNs, 16 are newly identified, which significantly expands the known RCL AGN sample. Based on the spectra and densely sampled MIR light curves, we derive MIR variability timescales for 18 RCL AGNs, and find no significant correlation between the timescale and the black hole mass.

astro-ph.GA↗

arXiv2025

AdapSCA-PSO: An Adaptive Localization Algorithm with AI-Based Hybrid SCA-PSO for IoT WSNs

The accurate localization of sensor nodes is a fundamental requirement for the practical application of the Internet of Things (IoT). To enable robust localization across diverse environments, this paper proposes a hybrid meta-heuristic localization algorithm. Specifically, the algorithm integrates the Sine Cosine Algorithm (SCA), which is effective in global search, with Particle Swarm Optimization (PSO), which excels at local search. An adaptive switching module is introduced to dynamically select between the two algorithms. Furthermore, the initialization, fitness evaluation, and parameter settings of the algorithm have been specifically redesigned and optimized to address the characteristics of the node localization problem. Simulation results across varying numbers of sensor nodes demonstrate that, compared to standalone PSO and the unoptimized SCAPSO algorithm, the proposed method significantly reduces the number of required iterations and achieves an average localization error reduction of 84.97%.

cs.NI↗

arXiv2025

SelfRACG: Enabling LLMs to Self-Express and Retrieve for Code Generation

Existing retrieval-augmented code generation (RACG) methods typically use an external retrieval module to fetch semantically similar code snippets used for generating subsequent fragments. However, even for consecutive code fragments, the content often diverges due to logical progression, resulting in a content gap. This gap undermines the performance of current RACG methods, as \textit{external} retrieval modules based on content matching fail to infer the specific information need of LLMs to generate the next code fragment. Therefore, we propose \textbf{SelfRACG}, a novel paradigm that enables large language models (LLMs) to \textbf{Self}-express their information needs to enhance \textbf{RACG}. Specifically, SelfRACG includes an information need expression module and a two-stage information need-guided training strategy, which encourages LLMs to express their information need. Extensive experiments demonstrate that SelfRACG can retrieve external knowledge that better aligns with the LLM's own information needs, resulting in superior generation performance compared to vanilla RACG.

cs.IR↗

arXiv2025

Dynamic and Parametric Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) has become a foundational paradigm for equipping large language models (LLMs) with external knowledge, playing a critical role in information retrieval and knowledge-intensive applications. However, conventional RAG systems typically adopt a static retrieve-then-generate pipeline and rely on in-context knowledge injection, which can be suboptimal for complex tasks that require multihop reasoning, adaptive information access, and deeper integration of external knowledge. Motivated by these limitations, the research community has moved beyond static retrieval and in-context knowledge injection. Among the emerging directions, this tutorial delves into two rapidly growing and complementary research areas on RAG: Dynamic RAG and Parametric RAG. Dynamic RAG adaptively determines when and what to retrieve during the LLM's generation process, enabling real-time adaptation to the LLM's evolving information needs. Parametric RAG rethinks how retrieved knowledge should be injected into LLMs, transitioning from input-level to parameter-level knowledge injection for enhanced efficiency and effectiveness. This tutorial offers a comprehensive overview of recent advances in these emerging research areas. It also shares theoretical foundations and practical insights to support and inspire further research in RAG.

cs.CL↗

arXiv2025

An Improved Grey Wolf Optimizer Inspired by Advanced Cooperative Predation for UAV Shortest Path Planning

With the widespread application of Unmanned Aerial Vehicles (UAVs) in domains like military reconnaissance, emergency rescue, and logistics delivery, efficiently planning the shortest flight path has become a critical challenge. Traditional heuristic-based methods often suffer from the inability to escape from local optima, which limits their effectiveness in finding the shortest path. To address these issues, a novel Improved Grey Wolf Optimizer (IGWO) is presented in this study. The proposed IGWO incorporates an Advanced Cooperative Predation (ACP) and a Lens Opposition-based Learning Strategy (LOBL) in order to improve the optimization capability of the method. Simulation results show that IGWO ranks first in optimization performance on benchmark functions F1-F5, F7, and F9-F12, outperforming all other compared algorithms. Subsequently, IGWO is applied to UAV shortest path planning in various obstacle-laden environments. Simulation results show that the paths planned by IGWO are, on average, shorter than those planned by GWO, PSO, and WOA by 1.70m, 1.68m, and 2.00m, respectively, across four different maps.

cs.RO↗

arXiv2025

RAIL: An Accurate and Fast Angle-inferred Localization Algorithm for UAV-WSN Systems

Location information is a fundamental requirement for unmanned aerial vehicles (UAVs) and other wireless sensor networks (WSNs). However, accurately and efficiently localizing sensor nodes with diverse functionalities remains a significant challenge, particularly in a hardware-constrained environment. To address this issue and enhance the applicability of artificial intelligence (AI), this paper proposes a localization algorithm that does not require additional hardware. Specifically, the angle between a node and the anchor nodes is estimated based on the received signal strength indication (RSSI). A subsequent localization strategy leverages the inferred angular relationships in conjunction with a bounding box. Experimental evaluations in three scenarios with varying number of nodes demonstrate that the proposed method achieves substantial improvements in localization accuracy, reducing the average error by 72.4% compared to the Min-Max and RSSI-based DV-Hop algorithms, respectively.

cs.NI↗

arXiv2025

Topology Partitioning-based Self-Organized Localization in Indoor WSNs with Unknown Obstacles

Accurate indoor node localization is critical for practical Wireless Sensor Network (WSN) applications, as Global Positioning System (GPS) fails to provide reliable Line-of-Sight (LoS) conditions in most indoor environments. Real-world localization scenarios often involve unknown obstacles with unpredictable shapes, sizes, quantities, and layouts. These obstacles introduce significant deviations in measured distances between sensor nodes when communication links traverse them, severely compromising localization accuracy. To address this challenge, this paper proposes a robust range-based localization method that strategically identifies and severs obstructed communication paths, leveraging network topology to mitigate obstacle-induced errors. Across diverse obstacle configurations and node densities, the algorithm successfully severed 87% of obstacle-affected paths on average. Under the assumption that Received Signal Strength Indicator (RSSI) provides accurate distance measurements under LoS conditions, the achieved localization accuracy exceeds 99.99%.

cs.NI↗

arXiv2025

Qilin: A Multimodal Information Retrieval Dataset with APP-level User Sessions

User-generated content (UGC) communities, especially those featuring multimodal content, improve user experiences by integrating visual and textual information into results (or items). The challenge of improving user experiences in complex systems with search and recommendation (S\&R) services has drawn significant attention from both academia and industry these years. However, the lack of high-quality datasets has limited the research progress on multimodal S\&R. To address the growing need for developing better S\&R services, we present a novel multimodal information retrieval dataset in this paper, namely Qilin. The dataset is collected from Xiaohongshu, a popular social platform with over 300 million monthly active users and an average search penetration rate of over 70\%. In contrast to existing datasets, \textsf{Qilin} offers a comprehensive collection of user sessions with heterogeneous results like image-text notes, video notes, commercial notes, and direct answers, facilitating the development of advanced multimodal neural retrieval models across diverse task settings. To better model user satisfaction and support the analysis of heterogeneous user behaviors, we also collect extensive APP-level contextual signals and genuine user feedback. Notably, Qilin contains user-favored answers and their referred results for search requests triggering the Deep Query Answering (DQA) module. This allows not only the training \& evaluation of a Retrieval-augmented Generation (RAG) pipeline, but also the exploration of how such a module would affect users' search behavior. Through comprehensive analysis and experiments, we provide interesting findings and insights for further improving S\&R systems. We hope that \textsf{Qilin} will significantly contribute to the advancement of multimodal content platforms with S\&R services in the future.

cs.IR↗

arXiv2024

LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods

The rapid advancement of Large Language Models (LLMs) has driven their expanding application across various fields. One of the most promising applications is their role as evaluators based on natural language responses, referred to as ''LLMs-as-judges''. This framework has attracted growing attention from both academia and industry due to their excellent effectiveness, ability to generalize across tasks, and interpretability in the form of natural language. This paper presents a comprehensive survey of the LLMs-as-judges paradigm from five key perspectives: Functionality, Methodology, Applications, Meta-evaluation, and Limitations. We begin by providing a systematic definition of LLMs-as-Judges and introduce their functionality (Why use LLM judges?). Then we address methodology to construct an evaluation system with LLMs (How to use LLM judges?). Additionally, we investigate the potential domains for their application (Where to use LLM judges?) and discuss methods for evaluating them in various contexts (How to evaluate LLM judges?). Finally, we provide a detailed analysis of the limitations of LLM judges and discuss potential future directions. Through a structured and comprehensive analysis, we aim aims to provide insights on the development and application of LLMs-as-judges in both research and practice. We will continue to maintain the relevant resource list at https://github.com/CSHaitao/Awesome-LLMs-as-Judges.

cs.CL↗

arXiv2024

CalibraEval: Calibrating Prediction Distribution to Mitigate Selection Bias in LLMs-as-Judges

The use of large language models (LLMs) as automated evaluation tools to assess the quality of generated natural language, known as LLMs-as-Judges, has demonstrated promising capabilities and is rapidly gaining widespread attention. However, when applied to pairwise comparisons of candidate responses, LLM-based evaluators often exhibit selection bias. Specifically, their judgments may become inconsistent when the option positions or ID tokens are swapped, compromising the effectiveness and fairness of the evaluation result. To address this challenge, we introduce CalibraEval, a novel label-free method for mitigating selection bias during inference. Specifically, CalibraEval reformulates debiasing as an optimization task aimed at adjusting observed prediction distributions to align with unbiased prediction distributions. To solve this optimization problem, we propose a non-parametric order-preserving algorithm (NOA). This algorithm leverages the partial order relationships between model prediction distributions, thereby eliminating the need for explicit labels and precise mathematical function modeling.Empirical evaluations of LLMs in multiple representative benchmarks demonstrate that CalibraEval effectively mitigates selection bias and improves performance compared to existing debiasing methods. This work marks a step toward building more robust and unbiased automated evaluation frameworks, paving the way for improved reliability in AI-driven assessments

cs.CL↗

arXiv2024

A Novel Improved Beluga Whale Optimization Algorithm for Solving Localization Problem in Swarm Robotic Systems

In Swarm Robotic Systems (SRSs), only a few robots are equipped with Global Positioning System (GPS) devices, known as anchors. A challenge lies in inferring the positions of other unknown robots based on the positions of anchors. Existing solutions estimate their positions using distance measurements between unknown robots and anchors. Based on existing solutions, this study proposes a novel meta-heuristic algorithm - Improved Beluga Whale Optimization Algorithm (IBWO) to address the localization problem of SRSs, focusing on enhancing the accuracy of localization results. Simulation results demonstrate the effectiveness of this study. Specifically, we test the localization accuracy of robots under different proportions of anchors, different communication radius of robots, and different total number of robots. Compared to the traditional multilateration method and four other localization methods based on meta-heuristic algorithms, the localization accuracy of this method is consistently superior.

cs.NI↗