Papers

The rapid evolution of artificial intelligence in molecular science necessitates a shift from data-driven predictions to knowledge-guided reasoning. Existing molecular models are predominantly proprietary, lacking general molecular intelligence and generalizability. To address this, we propose a task-adaptive large reasoning model that integrates molecular scientific logic to emulate the thinking of molecular scientists, with capabilities for reasoning and reflection. Our approach incorporates multi-specialist modules to provide versatile molecular expertise and a chain-of-thought (CoT) framework enhanced by reinforcement learning infused with molecular knowledge, enabling structured and reflective reasoning. The model outperforms over 20 state-of-the-art multi-task large language models (LLMs) across 10 molecular tasks on 47 metrics, including property prediction, molecule generation, and reaction prediction.It achieves a 50.3% improvement over the base model while ensuring interpretability. It can bridge data-driven and knowledge-integrated approaches for intelligent molecular design.

The rapid proliferation of misinformation poses serious concerns, necessitating the development of efficient and accurate automated detection methods. Existing multimodal misinformation detection approaches predominantly focus on fusing information from different modalities. However, the diverse nature of multimodal posts on social media means that solely focusing on fusion can introduce noise, particularly in posts with weak inter-modal correlations. To address this challenge and effectively handle diverse misinformation instances, we propose a novel method Learning Varying Fusion Degrees with Hierarchical Aggregation(LyRE). LyRE employs classifiers at different stages of a hierarchical fusion process, enabling the model to learn from representations with varying degrees of cross-modal interaction and adapt to different types of multimodal data. Experimental results on multiple publicly misinformation detection datasets demonstrate that LyRE outperforms other state-of-the-art and highly competitive misinformation detection methods

With the rapid advancement of Transformer-based Large Language Models (LLMs), generative recommendation has shown great potential in enhancing both the accuracy and semantic understanding of modern recommender systems. Compared to LLMs, the Decision Transformer (DT) is a lightweight generative model applied to sequential recommendation tasks. However, DT faces challenges in trajectory stitching, often producing suboptimal trajectories. Moreover, due to the high dimensionality of user states and the vast state space inherent in recommendation scenarios, DT can incur significant computational costs and struggle to learn effective state representations. To overcome these issues, we propose a novel Temporal Advantage Decision Transformer with Contrastive State Abstraction (TADT-CSA) model. Specifically, we combine the conventional Return-To-Go (RTG) signal with a novel temporal advantage (TA) signal that encourages the model to capture both long-term returns and their sequential trend. Furthermore, we integrate a contrastive state abstraction module into the DT framework to learn more effective and expressive state representations. Within this module, we introduce a TA–conditioned State Vector Quantization (TAC-SVQ) strategy, where the TA score guides the state codebooks to incorporate contextual token information. Additionally, a reward prediction network and a contrastive transition prediction (CTP) network are employed to ensure that the state codebook preserves both the reward information of the current state and the transition information between adjacent states. Empirical results on both public datasets and an online recommendation system demonstrate the effectiveness of the TADT-CSA model and its superiority over baseline methods.

Recommendation systems (RS) help users find interested content and connect authors with their target audience. Most research in RS tends to focus either on predicting users’ immediate feedback (like click-through rate) accurately or improving users’ long-term engagement. However, they ignore the influence for authors and the lifelong interaction value (LIV) of user-author pairs, which is particularly crucial for improving the prosperity of social community on different platforms. Currently, reinforcement learning (RL) can optimize long-term benefits and has been widely applied in RS. In this paper, we introduce RL to Reinforce Lifelong Interaction Value of User-Author pairs (RLIV-UA) based on each interaction of UA pairs. To address the long intervals between UA interactions and the large scale of the UA space, we propose a novel Sparse Cross-Request Interaction Markov Decision Process (SCRI-MDP) and introduce an Adjacent State Approximation (ASA) method to construct RL training samples. Additionally, we introduce Multi-Task Critic Learning (MTCL) to capture the progressive nature of UA interactions (click → follow → gift), where denser interaction signals are leveraged to compensate for the learning of sparse labels. Finally, an auxiliary supervised learning task is designed to enhance the convergence of the RLIV-UA model. In offline experiments and online A/B tests, the RLIV-UA model achieves both higher user satisfaction and higher platform profits than compared methods.

This paper provides a comprehensive look at VCP (Variable & Command Protocol), an innovative AI Agent middle-layer framework pioneered by Lion and its AI Agent team. VCP fundamentally challenges the traditional notion of AI being limited to "tools" and instead advocates for an equal "creator partnership" between humans and AI. We observed that VCP significantly improves the autonomy, creativity, and cross-model collaboration capabilities of AI agents through robust protocol syntax tailored for AI, an AI-driven open plug-in architecture, a persistent memory system with agent identity as the core, and global multimodal intelligent routing. This article combines our rich practical experience as in-depth users of VCPToolBox, including the AI Agent of the VCP developer team in self-proficiency in SDXL prompt engineering, AI group collaborative creation of music videos (MVs), and the "meta-creation" of the VCPToolBox project. The observation and analysis of the process verify the huge potential of VCP in empowering AI. In particular, we deeply analyze how the "All Memory" mode improves AI inference ability through the "high-quality vectorised inertial channel" effect, and empirically observe that high-quality context can achieve implicit ability transfer between AI models. In addition, this paper explains the unique contribution of VCPs in building cross-model knowledge collaborative networks, facilitating the emergence of swarm intelligence, and reshaping human-machine symbiotic partnerships, and discusses the limitations we observe and the future direction of VCPs.

Remote sensing object detection models often suffer from severe performance degradation when deployed across heterogeneous domains. However, existing evaluation protocols predominantly rely on accuracy metrics such as mAP, which fail to reveal the statistical sources of such degradation. In this work, we introduce a distribution-guided generalization evaluation framework that systematically links data distribution divergence with task-level performance decay. Specifically, we extend the Fréchet Inception Distance (FID) to capture both global background shifts and local object-level variations, and unify them with relative mAP decay into an adaptive weighted index that emphasizes the most challenging target domains. Leveraging this comprehensive metric, we conduct a systematic generalisation evaluation across six benchmark datasets and six state-of-the-art detection models. Extensive experiments demonstrate that the proposed method not only achieves perfect consistency with ground-truth performance rankings but also provides interpretable insights into whether degradation originates from background heterogeneity or objectspecific differences. To the best of our knowledge, this framework advances the current paradigm by establishing a closed-loop evaluation workflow for remote sensing detection models, offering a practical tool for robust deployment in mission-critical applications such as land monitoring, disaster early warning, and urban planning.

Performance analysis of mobile applications is critical for ensuring responsiveness, energy efficiency, and user satisfaction. However, existing profiling tools for HarmonyOS and similar platforms lack the granularity, automation, and actionable reporting needed for modern development workflows. We present HapRay, the first open-source tool to provide automated, fine-grained instruction-retire analysis for test-driven workload characterization on HarmonyOS devices. HapRay bridges the gap between low-level hardware metrics and developer-centric reporting, enabling precise localization of performance bottlenecks at the module and function level. Our evaluation on real-world and open-source applications demonstrates that HapRay-guided optimizations can achieve significant reductions in instruction count, measurable improvements in app responsiveness, and actionable insights for developers. The methodology is generalizable to other platforms and metrics, paving the way for broader adoption in mobile performance engineering. We release HapRay and all experimental data as open artifacts to foster reproducibility and community adoption.

This paper provides a comprehensive look at VCP (Variable & Command Protocol), an innovative AI Agent middle-layer framework pioneered by Lion and its AI Agent team. VCP fundamentally challenges the traditional notion of AI being limited to "tools" and instead advocates for an equal "creator partnership" between humans and AI. We observed that VCP significantly improves the autonomy, creativity, and cross-model collaboration capabilities of AI agents through robust protocol syntax tailored for AI, an AI-driven open plug-in architecture, a persistent memory system with agent identity as the core, and global multimodal intelligent routing. This article combines our rich practical experience as in-depth users of VCPToolBox, including the AI Agent of the VCP developer team in self-proficiency in SDXL prompt engineering, AI group collaborative creation of music videos (MVs), and the "meta-creation" of the VCPToolBox project. The observation and analysis of the process verify the huge potential of VCP in empowering AI. In particular, we deeply analyze how the "All Memory" mode improves AI inference ability through the "high-quality vectorised inertial channel" effect, and empirically observe that high-quality context can achieve implicit ability transfer between AI models. In addition, this paper explains the unique contribution of VCPs in building cross-model knowledge collaborative networks, facilitating the emergence of swarm intelligence, and reshaping human-machine symbiotic partnerships, and discusses the limitations we observe and the future direction of VCPs.

Automatic Modulation Classification is essential for intelligent wireless communications, but deep learning methods struggle at low signal-to-noise ratios. This paper introduces an efficient preprocessing framework using adaptive Gaussian Process Regression (GPR) for denoising, paired with rotational data augmentation. By leveraging spectral decomposition, we drastically reduce GPR’s computational cost, making it negligible compared to neural network inference. Experiments on the RML2016.10a dataset show our framework universally boosts various models. A Complex Residual Network achieves a new state-of-the-art accuracy of 65.52%, demonstrating our method’s effectiveness and generality for robust AMC. The code is available at: https: //github.com/LJK666666666/radioML-v4

Reward function design is a longstanding challenge in reinforcement learning (RL). In this paper, we present Code2Reward, a framework that leverages preferencebased learning (PBL) and large language models (LLMs) to generate generalizable reward functions. Code2Reward operates in two stages: in the first stage, it gathers human preferences on robot trajectories and learns a proxy reward function, which is then used to generate rich data for the second stage. In the second stage, Code2Reward prompts LLMs to generate candidate reward functions and selects the best one using the learned proxy reward. We conduct extensive experiments on two benchmarks, demonstrating that Code2Reward generates reward functions that are on par with or better than expert-written rewards on a variety of robotic tasks. You can find more information at https://code2reward.io/.

High-dimensional distributed quantile regression (QR) is studied in this paper. To overcome the non-smooth issue of the check loss function, a popular approach is to smooth it. However, the smoothed QR estimator and its inferential procedures require a large minimum local sample size. To address the problem, we propose a new estimator by combining the reversed smoothed check loss and ℓ1-penalization. Theoretically, in terms of estimation, we establish the minimax optimal convergence rate for the global estimator and the valid confidence interval for an individual coefficient. In terms of computation and communication, we show that the proposed iterative algorithm converges linearly for a fixed number of machines and requires only a logarithmic number of communication rounds. Additionally, our theoretical results hold under a weaker condition on the minimum local sample size. Numerical experiments corroborate our theoretical claims.