arXiv subjects

Jie Tang

Publications and source records attributed to Jie Tang.

At least 127 records · Page 7Linked to original sources

arXiv2025

Variable Block-Correlation Modeling and Optimization for Secrecy Analysis in Fluid Antenna Systems

Fluid antenna systems (FAS) are emerging as a transformative enabler for sixth-generation (6G) wireless communications, providing unprecedented spatial diversity through dynamic reconfiguration of antenna ports. However, the inherent spatial correlation among ports poses significant challenges for accurate analysis. Conventional models such as Jakes are analytically intractable, while oversimplified constant-correlation models fail to capture the true behavior. In this work, we address these challenges by applying the variable block-correlation model (VBCM) -- originally proposed by Ramírez-Espinosa \textit{et al.} in 2024 -- to FAS security analysis, and by developing comprehensive optimization methods to enhance analytical accuracy. We derive new closed-form expressions for average secrecy capacity (ASC) and secrecy outage probability (SOP), demonstrating that the VBCM framework achieves simulation-aligned accuracy, with relative errors consistently below $5\%$ (compared to $10$--$15\%$ for constant-correlation models). To maximize ASC, we further design two algorithms: a grid search (GS) method and a gradient descent (GD) method. Numerical results reveal that the VBCM-based approach not only provides reliable insights into FAS security performance, but also yields substantial gains -- ASC improvements exceeding $120\%$ in high-threat scenarios and $18$--$19\%$ performance enhancements for compact antenna configurations. These findings underscore the practical value of integrating VBCM into FAS security analysis and optimization, establishing it as a powerful tool for advancing 6G communication systems.

eess.SP↗

arXiv2025

TDRM: Smooth Reward Models with Temporal Difference for LLM RL and Inference

Reward models are central to both reinforcement learning (RL) with language models and inference-time verification. However, existing reward models often lack temporal consistency, leading to ineffective policy updates and unstable RL training. We introduce TDRM, a method for learning smoother and more reliable reward models by minimizing temporal differences (TD) for training-time reinforcement learning and inference-time verification. Experiments show that TD-trained process reward models (PRMs) improve performance across Best-of-N (up to 6.6%) and tree-search (up to 23.7%) settings. When combined with Reinforcement Learning with Verifiable Rewards (RLVR), TD-trained PRMs lead to more data-efficient RL -- achieving comparable performance with just 2.5k data to what baseline methods require 50.1k data to attain -- and yield higher-quality language model policies in 8 model variants (5 series), e.g., Qwen2.5-(0.5B, 1,5B), GLM4-9B-0414, GLM-Z1-9B-0414, Qwen2.5-Math-(1.5B, 7B), and DeepSeek-R1-Distill-Qwen-(1.5B, 7B). We release all code at https://github.com/THUDM/TDRM.

cs.LG↗

arXiv2025

SIRI: Scaling Iterative Reinforcement Learning with Interleaved Compression

We introduce SIRI, Scaling Iterative Reinforcement Learning with Interleaved Compression, a simple yet effective RL approach for Large Reasoning Models (LRMs) that enables more efficient and accurate reasoning. Existing studies have observed repetitive thinking patterns in LRMs, and attempts to reduce them often come at the cost of performance. In this paper, we show that this trade-off can be overcome through a training regime that iteratively alternates between compressing and expanding the reasoning budget, by dynamically adjusting the maximum rollout length during training. The compression phase cuts the rollout length, forcing the model to make precise and valuable decisions within a limited context, which effectively reduces redundant tokens and increases reasoning density. The expansion phase then relaxes the length limit, providing space for the model to explore and plan in long-horizon settings. Remarkably, we find that after each compression-expansion cycle, the model's performance improves even as its output length decreases, steadily pushing it closer to the Pareto frontier in the performance-efficiency trade-off. Training on DeepSeek-R1-Distill-Qwen-1.5B, SIRI-low improves performance on AIME24 by 43.2% while reducing token usage by 46.9% after three iterations, and SIRI-high achieves the highest accuracy compared to all other methods (Figure 1). Our findings shed light on the potential of periodically oscillating the LRM's output truncation length during training to dynamically balance exploration and efficiency in reasoning, converging towards an optimal "sweet spot" between the two. Our models are publicly available.

cs.LG↗

arXiv2025

Conf-Profile: A Confidence-Driven Reasoning Paradigm for Label-Free User Profiling

User profiling, as a core technique for user understanding, aims to infer structural attributes from user information. Large Language Models (LLMs) provide a promising avenue for user profiling, yet the progress is hindered by the lack of comprehensive benchmarks. To bridge this gap, we propose ProfileBench, an industrial benchmark derived from a real-world video platform, encompassing heterogeneous user data and a well-structured profiling taxonomy. However, the profiling task remains challenging due to the difficulty of collecting large-scale ground-truth labels, and the heterogeneous and noisy user information can compromise the reliability of LLMs. To approach label-free and reliable user profiling, we propose a Confidence-driven Profile reasoning framework Conf-Profile, featuring a two-stage paradigm. We first synthesize high-quality labels by leveraging advanced LLMs with confidence hints, followed by confidence-weighted voting for accuracy improvement and confidence calibration for a balanced distribution. The multiple profile results, rationales, and confidence scores are aggregated and distilled into a lightweight LLM. We further enhance the reasoning ability via confidence-guided unsupervised reinforcement learning, which exploits confidence for difficulty filtering, quasi-ground truth voting, and reward weighting. Experimental results demonstrate that Conf-Profile delivers substantial performance through the two-stage training, improving F1 by 13.97 on Qwen3-8B.

cs.AI↗

arXiv2025

UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning

The development of autonomous agents for graphical user interfaces (GUIs) presents major challenges in artificial intelligence. While recent advances in native agent models have shown promise by unifying perception, reasoning, action, and memory through end-to-end learning, open problems remain in data scalability, multi-turn reinforcement learning (RL), the limitations of GUI-only operation, and environment stability. In this technical report, we present UI-TARS-2, a native GUI-centered agent model that addresses these challenges through a systematic training methodology: a data flywheel for scalable data generation, a stabilized multi-turn RL framework, a hybrid GUI environment that integrates file systems and terminals, and a unified sandbox platform for large-scale rollouts. Empirical evaluation demonstrates that UI-TARS-2 achieves significant improvements over its predecessor UI-TARS-1.5. On GUI benchmarks, it reaches 88.2 on Online-Mind2Web, 47.5 on OSWorld, 50.6 on WindowsAgentArena, and 73.3 on AndroidWorld, outperforming strong baselines such as Claude and OpenAI agents. In game environments, it attains a mean normalized score of 59.8 across a 15-game suite-roughly 60% of human-level performance-and remains competitive with frontier proprietary models (e.g., OpenAI o3) on LMGame-Bench. Additionally, the model can generalize to long-horizon information-seeking tasks and software engineering benchmarks, highlighting its robustness across diverse agent tasks. Detailed analyses of training dynamics further provide insights into achieving stability and efficiency in large-scale agent RL. These results underscore UI-TARS-2's potential to advance the state of GUI agents and exhibit strong generalization to real-world interactive scenarios.

cs.AI↗

arXiv2025

Rotatable Antenna Enabled Wireless Communication and Sensing: Opportunities and Challenges

Non-fixed flexible antenna architectures, such as fluid antenna system (FAS), movable antenna (MA), and pinching antenna, have garnered significant interest in recent years. Among them, rotatable antenna (RA) is an emerging technology that offers significant potential to enhance wireless communication and sensing performance by flexibly adjusting the boresight of directional antennas. Specifically, RA can flexibly reconfigure its boresight direction via mechanical or electronic means, thereby improving communication channel conditions and/or enhancing sensing resolution and range. In this article, we first provide an overview of RA, covering its hardware architectures and radiation pattern characterization. We then illustrate how RA improves communication performance through interference mitigation, spatial multiplexing, and flexible beamforming, as well as sensing capabilities in terms of coverage, resolution, and multi-target/dimensional sensing. Furthermore, we highlight representative applications of RA and discuss key design challenges in RA systems, including rotational scanning scheduling, channel estimation/sensing, boresight optimization, and RA configuration. Finally, both experimental and simulation results are provided to validate the performance gains achieved by RA for both communication and sensing. Leveraging its unique capabilities in flexible antenna/array rotation to adapt to various communication/sensing requirements and channel conditions, RA is poised to become a key enabler of future intelligent, resilient, and agile wireless networks.

cs.IT↗

arXiv2025

VPO: Aligning Text-to-Video Generation Models with Prompt Optimization

Video generation models have achieved remarkable progress in text-to-video tasks. These models are typically trained on text-video pairs with highly detailed and carefully crafted descriptions, while real-world user inputs during inference are often concise, vague, or poorly structured. This gap makes prompt optimization crucial for generating high-quality videos. Current methods often rely on large language models (LLMs) to refine prompts through in-context learning, but suffer from several limitations: they may distort user intent, omit critical details, or introduce safety risks. Moreover, they optimize prompts without considering the impact on the final video quality, which can lead to suboptimal results. To address these issues, we introduce VPO, a principled framework that optimizes prompts based on three core principles: harmlessness, accuracy, and helpfulness. The generated prompts faithfully preserve user intents and, more importantly, enhance the safety and quality of generated videos. To achieve this, VPO employs a two-stage optimization approach. First, we construct and refine a supervised fine-tuning (SFT) dataset based on principles of safety and alignment. Second, we introduce both text-level and video-level feedback to further optimize the SFT model with preference learning. Our extensive experiments demonstrate that VPO significantly improves safety, alignment, and video quality compared to baseline methods. Moreover, VPO shows strong generalization across video generation models. Furthermore, we demonstrate that VPO could outperform and be combined with RLHF methods on video generation models, underscoring the effectiveness of VPO in aligning video generation models. Our code and data are publicly available at https://github.com/thu-coai/VPO.

cs.CV↗

arXiv2025

SoAy: A Solution-based LLM API-using Methodology for Academic Information Seeking

Applying large language models (LLMs) for academic API usage shows promise in reducing researchers' academic information seeking efforts. However, current LLM API-using methods struggle with complex API coupling commonly encountered in academic queries. To address this, we introduce SoAy, a solution-based LLM API-using methodology for academic information seeking. It uses code with a solution as the reasoning method, where a solution is a pre-constructed API calling sequence. The addition of the solution reduces the difficulty for the model to understand the complex relationships between APIs. Code improves the efficiency of reasoning. To evaluate SoAy, we introduce SoAyBench, an evaluation benchmark accompanied by SoAyEval, built upon a cloned environment of APIs from AMiner. Experimental results demonstrate a 34.58-75.99\% performance improvement compared to state-of-the-art LLM API-based baselines. All datasets, codes, tuned models, and deployed online services are publicly accessible at https://github.com/RUCKBReasoning/SoAy.

cs.CL↗

arXiv2025

LVBench: An Extreme Long Video Understanding Benchmark

Recent progress in multimodal large language models has markedly enhanced the understanding of short videos (typically under one minute), and several evaluation datasets have emerged accordingly. However, these advancements fall short of meeting the demands of real-world applications such as embodied intelligence for long-term decision-making, in-depth movie reviews and discussions, and live sports commentary, all of which require comprehension of long videos spanning several hours. To address this gap, we introduce LVBench, a benchmark specifically designed for long video understanding. Our dataset comprises publicly sourced videos and encompasses a diverse set of tasks aimed at long video comprehension and information extraction. LVBench is designed to challenge multimodal models to demonstrate long-term memory and extended comprehension capabilities. Our extensive evaluations reveal that current multimodal models still underperform on these demanding long video understanding tasks. Through LVBench, we aim to spur the development of more advanced models capable of tackling the complexities of long video comprehension. Our data and code are publicly available at: https://lvbench.github.io.

cs.CV↗

arXiv2025

GuARD: Effective Anomaly Detection through a Text-Rich and Graph-Informed Language Model

Anomaly detection on text-rich graphs is widely prevalent in real life, such as detecting incorrectly assigned academic papers to authors and detecting bots in social networks. The remarkable capabilities of large language models (LLMs) pave a new revenue by utilizing rich-text information for effective anomaly detection. However, simply introducing rich texts into LLMs can obscure essential detection cues and introduce high fine-tuning costs. Moreover, LLMs often overlook the intrinsic structural bias of graphs which is vital for distinguishing normal from abnormal node patterns. To this end, this paper introduces GuARD, a text-rich and graph-informed language model that combines key structural features from graph-based methods with fine-grained semantic attributes extracted via small language models for effective anomaly detection on text-rich graphs. GuARD is optimized with the progressive multi-modal multi-turn instruction tuning framework in the task-guided instruction tuning regime tailed to incorporate both rich-text and structural modalities. Extensive experiments on four datasets reveal that GuARD outperforms graph-based and LLM-based anomaly detection methods, while offering up to 5$\times$ times speedup in training and 5$\times$ times speedup in inference over vanilla long-context LLMs on the large-scale WhoIsWho dataset.

cs.CL↗

arXiv2025

High-gain optical parametric amplification with a continuous-wave pump using a domain-engineered thin-film lithium niobate waveguide

While thin film lithium niobate (TFLN) is known for efficient signal generation, on-chip signal amplification remains challenging from fully integrated optical communication circuits. Here we demonstrate the continuous-wave-pump optical parametric amplification (OPA) using an x-cut domain-engineered TFLN waveguide, with high gain over the telecom band up to 13.9 dB, and test it for high signal-to-noise ratio signal amplification using a commercial optical communication module pair. Fabricated in wafer scale using common process as devices including modulators, this OPA device marks an important step in TFLN photonic integration.

physics.optics↗

arXiv2025

LPS-GNN : Deploying Graph Neural Networks on Graphs with 100-Billion Edges

Graph Neural Networks (GNNs) have emerged as powerful tools for various graph mining tasks, yet existing scalable solutions often struggle to balance execution efficiency with prediction accuracy. These difficulties stem from iterative message-passing techniques, which place significant computational demands and require extensive GPU memory, particularly when dealing with the neighbor explosion issue inherent in large-scale graphs. This paper introduces a scalable, low-cost, flexible, and efficient GNN framework called LPS-GNN, which can perform representation learning on 100 billion graphs with a single GPU in 10 hours and shows a 13.8% improvement in User Acquisition scenarios. We examine existing graph partitioning methods and design a superior graph partition algorithm named LPMetis. In particular, LPMetis outperforms current state-of-the-art (SOTA) approaches on various evaluation metrics. In addition, our paper proposes a subgraph augmentation strategy to enhance the model's predictive performance. It exhibits excellent compatibility, allowing the entire framework to accommodate various GNN algorithms. Successfully deployed on the Tencent platform, LPS-GNN has been tested on public and real-world datasets, achieving performance lifts of 8. 24% to 13. 89% over SOTA models in online applications.

cs.LG↗

arXiv2025

UavNetSim-v1: A Python-based Simulation Platform for UAV Communication Networks

In unmanned aerial vehicle (UAV) networks, communication protocols and algorithms are essential for cooperation and collaboration between UAVs. Simulation provides a cost-effective solution for prototyping, debugging, and analyzing protocols and algorithms, avoiding the prohibitive expenses of field experiments. In this paper, we present ``UavNetSim-v1'', an open-source Python-based simulation platform designed for rapid development, testing, and evaluating the protocols and algorithms in UAV networks. ``UavNetSim-v1'' provides most of the functionalities developers may need, including routing/medium access control (MAC) protocols, topology control algorithms and mobility/energy models, while maintaining ease of use. Furthermore, the platform supports comprehensive performance evaluation and features an interactive visualization interface for in-depth algorithm analysis. In short, ``UavNetSim-v1'' lends itself to both rapid prototyping and educational purposes, and can serve as a lightweight yet powerful alternative to mature network simulators for UAV communication research.

cs.NI↗

arXiv2025

Efficient Channel Estimation for Rotatable Antenna-Enabled Wireless Communication

Non-fixed flexible antenna architectures, such as fluid antenna system (FAS), movable antenna (MA), and pinching antenna, have garnered significant interest in recent years. Among them, rotatable antenna (RA) is a promising antenna architecture that exploits additional spatial degrees of freedom (DoFs) to enhance the communication performance. To fully obtain the performance gain provided by RAs, accurate channel state information (CSI) is essential for adjusting the orientation/boresight of each antenna. In this letter, we propose an efficient channel estimation scheme for RA communication systems, where the base station (BS) can sequentially and adaptively adjust the orientations of RAs to enrich the environmental observations from diverse angular perspectives, thereby enhancing the channel estimation accuracy. The proposed scheme includes two main procedures that are conducted alternately during each channel training period. Specifically, the first procedure is to estimate the CSI with given RAs' orientations, involving the angle-of-arrivals (AoAs) information and path gains. Then, based on the estimated CSI, the second procedure adjusts the RAs' orientations to maximize the effective channel gain. Simulation results demonstrate that the proposed channel estimation method outperforms other benchmark schemes.

cs.IT↗

arXiv2025

SWE-Dev: Building Software Engineering Agents with Training and Inference Scaling

Large language models (LLMs) have advanced rapidly from conversational problem solving to addressing real-world tasks involving tool use, such as software engineering (SWE). Recent LLM-powered toolkits, such as OpenAI Codex and Cursor, have offered end-to-end automation of the software development process. However, building effective SWE agents remains challenging due to the lack of high-quality training data and effective test cases. To address this issue, we present SWE-Dev, an SWE agent built upon open-source LLMs. First, we develop a robust pipeline to synthesize test cases for patch evaluation. Second, we scale up agent trajectories to construct the training data for building SWE-Dev. Experiments on the SWE-bench-Verified benchmark show that the SWE-Dev models can achieve top performance among all open SWE agents. Specifically, the success rates of the SWE-Dev 7B and 32B parameter models reach 23.4% and 36.6%, respectively, outperforming state-of-the-art open-source models. All code, models, and datasets are publicly available at https://github.com/THUDM/SWE-Dev.

cs.AI↗

arXiv2025

CoT-based Synthesizer: Enhancing LLM Performance through Answer Synthesis

Current inference scaling methods, such as Self-consistency and Best-of-N, have proven effective in improving the accuracy of LLMs on complex reasoning tasks. However, these methods rely heavily on the quality of candidate responses and are unable to produce correct answers when all candidates are incorrect. In this paper, we propose a novel inference scaling strategy, CoT-based Synthesizer, which leverages CoT reasoning to synthesize superior answers by analyzing complementary information from multiple candidate responses, even when all candidate responses are flawed. To enable a lightweight and cost-effective implementation, we introduce an automated data generation pipeline that creates diverse training data. This allows smaller LLMs trained on this data to improve the inference accuracy of larger models, including API-based LLMs. Experimental results across four benchmark datasets with seven policy models demonstrate that our method significantly enhances performance, with gains of 11.8% for Llama3-8B and 10.3% for GPT-4o on the MATH dataset. The corresponding training data and code are publicly available on https://github.com/RUCKBReasoning/CoT-based-Synthesizer.

cs.CL↗

arXiv2025

T1: Advancing Language Model Reasoning through Reinforcement Learning and Inference Scaling

Large language models (LLMs) have demonstrated remarkable capabilities in complex reasoning tasks. However, existing approaches mainly rely on imitation learning and struggle to achieve effective test-time scaling. While reinforcement learning (RL) holds promise for enabling self-exploration, recent attempts yield modest improvements in complex reasoning. In this paper, we present T1 to scale RL by encouraging exploration and understand inference scaling. We first initialize the LLM using synthesized chain-of-thought data that integrates trial-and-error and self-verification. To scale RL training, we promote increased sampling diversity through oversampling. We demonstrate that T1 with open LLMs as its base exhibits inference scaling behavior and achieves superior performance on challenging math reasoning benchmarks. More importantly, we present a simple strategy to examine inference scaling, where increased inference budgets directly lead to T1's better performance without any additional verification.

cs.LG↗

arXiv2025

TreeRL: LLM Reinforcement Learning with On-Policy Tree Search

Reinforcement learning (RL) with tree search has demonstrated superior performance in traditional reasoning tasks. Compared to conventional independent chain sampling strategies with outcome supervision, tree search enables better exploration of the reasoning space and provides dense, on-policy process rewards during RL training but remains under-explored in On-Policy LLM RL. We propose TreeRL, a reinforcement learning framework that directly incorporates on-policy tree search for RL training. Our approach includes intermediate supervision and eliminates the need for a separate reward model training. Existing approaches typically train a separate process reward model, which can suffer from distribution mismatch and reward hacking. We also introduce a cost-effective tree search approach that achieves higher search efficiency under the same generation token budget by strategically branching from high-uncertainty intermediate steps rather than using random branching. Experiments on challenging math and code reasoning benchmarks demonstrate that TreeRL achieves superior performance compared to traditional ChainRL, highlighting the potential of tree search for LLM. TreeRL is open-sourced at https://github.com/THUDM/TreeRL.

cs.LG↗