Zhenyu Huang

Shahrokh Akhlaghi, Ning Zhou, Zhenyu Huang

Accurate estimation of the dynamic states of a synchronous machine (e.g., rotor s angle and speed) is essential in monitoring and controlling transient stability of a power system. It is well known that the covariance matrixes of process noise (Q) and measurement noise (R) have a significant impact on the Kalman filter s performance in estimating dynamic states. The conventional ad-hoc approaches for estimating the covariance matrixes are not adequate in achieving the best filtering performance. To address this problem, this paper proposes an adaptive filtering approach to adaptively estimate Q and R based on innovation and residual to improve the dynamic state estimation accuracy of the extended Kalman filter (EKF). It is shown through the simulation on the two-area model that the proposed estimation method is more robust against the initial errors in Q and R than the conventional method in estimating the dynamic states of a synchronous machine.

Tencent Hunyuan Team, Ao Liu, Botong Zhou et al. · tencent-ai

As Large Language Models (LLMs) rapidly advance, we introduce Hunyuan-TurboS, a novel large hybrid Transformer-Mamba Mixture of Experts (MoE) model. It synergistically combines Mamba's long-sequence processing efficiency with Transformer's superior contextual understanding. Hunyuan-TurboS features an adaptive long-short chain-of-thought (CoT) mechanism, dynamically switching between rapid responses for simple queries and deep "thinking" modes for complex problems, optimizing computational resources. Architecturally, this 56B activated (560B total) parameter model employs 128 layers (Mamba2, Attention, FFN) with an innovative AMF/MF block pattern. Faster Mamba2 ensures linear complexity, Grouped-Query Attention minimizes KV cache, and FFNs use an MoE structure. Pre-trained on 16T high-quality tokens, it supports a 256K context length and is the first industry-deployed large-scale Mamba model. Our comprehensive post-training strategy enhances capabilities via Supervised Fine-Tuning (3M instructions), a novel Adaptive Long-short CoT Fusion method, Multi-round Deliberation Learning for iterative improvement, and a two-stage Large-scale Reinforcement Learning process targeting STEM and general instruction-following. Evaluations show strong performance: overall top 7 rank on LMSYS Chatbot Arena with a score of 1356, outperforming leading models like Gemini-2.0-Flash-001 (1352) and o4-mini-2025-04-16 (1345). TurboS also achieves an average of 77.9% across 23 automated benchmarks. Hunyuan-TurboS balances high performance and efficiency, offering substantial capabilities at lower inference costs than many reasoning models, establishing a new paradigm for efficient large-scale pre-trained models.

Jianming Lian, Shaobu Wang, Ruisheng Diao et al.

As power systems become more and more interconnected, the inter-area oscillations has become a serious factor limiting large power transfer among different areas. Underdamped (Undamped) inter-area oscillations may cause system breakup and even lead to large-scale blackout. Traditional damping controllers include Power System Stabilizer (PSS) and Flexible AC Transmission System (FACTS) controller, which adds additional damping to the inter-area oscillation modes by affecting the real power in an indirect manner. However, the effectiveness of these controllers is restricted to the neighborhood of a prescribed set of operating conditions. In this paper, decentralized robust controllers are developed to improve the damping ratios of the inter-area oscillation modes by directly affecting the real power through the turbine governing system. The proposed control strategy requires only local signals and is robust to the variations in operation condition and system topology. The effectiveness of the proposed robust controllers is illustrated by detailed case studies on two different test systems.

Ziqiao Shang, Lingyue Ge, Yang Chen et al.

Systematic evaluation of Multimodal Large Language Models (MLLMs) is crucial for advancing Artificial General Intelligence (AGI). However, existing benchmarks remain insufficient for rigorously assessing their reasoning capabilities under multi-criteria constraints. To bridge this gap, we introduce MapTab, a multimodal benchmark specifically designed to evaluate holistic multi-criteria reasoning in MLLMs via route planning tasks. MapTab requires MLLMs to perceive and ground visual cues from map images alongside route attributes (e.g., Time, Price) from structured tabular data. The benchmark encompasses two scenarios: Metromap, covering metro networks in 160 cities across 52 countries, and Travelmap, depicting 168 representative tourist attractions from 19 countries. In total, MapTab comprises 328 images, 196,800 route planning queries, and 3,936 QA queries, all incorporating 4 key criteria: Time, Price, Comfort, and Reliability. Extensive evaluations across 15 representative MLLMs reveal that current models face substantial challenges in multi-criteria multimodal reasoning. Notably, under conditions of limited visual perception, multimodal collaboration often underperforms compared to unimodal approaches. We believe MapTab provides a challenging and realistic testbed to advance the systematic evaluation of MLLMs. Our code is available at https://github.com/Ziqiao-Shang/MapTab.

Liang Jiang, Zhenyu Huang, Jia Liu et al.

Most domain adaptation methods for machine reading comprehension (MRC) use a pre-trained question-answer (QA) construction model to generate pseudo QA pairs for MRC transfer. Such a process will inevitably introduce mismatched pairs (i.e., noisy correspondence) due to i) the unavailable QA pairs in target documents, and ii) the domain shift during applying the QA construction model to the target domain. Undoubtedly, the noisy correspondence will degenerate the performance of MRC, which however is neglected by existing works. To solve such an untouched problem, we propose to construct QA pairs by additionally using the dialogue related to the documents, as well as a new domain adaptation method for MRC. Specifically, we propose Robust Domain Adaptation for Machine Reading Comprehension (RMRC) method which consists of an answer extractor (AE), a question selector (QS), and an MRC model. Specifically, RMRC filters out the irrelevant answers by estimating the correlation to the document via the AE, and extracts the questions by fusing the candidate questions in multiple rounds of dialogue chats via the QS. With the extracted QA pairs, MRC is fine-tuned and provides the feedback to optimize the QS through a novel reinforced self-training method. Thanks to the optimization of the QS, our method will greatly alleviate the noisy correspondence problem caused by the domain shift. To the best of our knowledge, this could be the first study to reveal the influence of noisy correspondence in domain adaptation MRC models and show a feasible way to achieve robustness to mismatched pairs. Extensive experiments on three datasets demonstrate the effectiveness of our method.

Yijie Lin, Jie Zhang, Zhenyu Huang et al.

Existing video-language studies mainly focus on learning short video clips, leaving long-term temporal dependencies rarely explored due to over-high computational cost of modeling long videos. To address this issue, one feasible solution is learning the correspondence between video clips and captions, which however inevitably encounters the multi-granularity noisy correspondence (MNC) problem. To be specific, MNC refers to the clip-caption misalignment (coarse-grained) and frame-word misalignment (fine-grained), hindering temporal learning and video understanding. In this paper, we propose NOise Robust Temporal Optimal traNsport (Norton) that addresses MNC in a unified optimal transport (OT) framework. In brief, Norton employs video-paragraph and clip-caption contrastive losses to capture long-term dependencies based on OT. To address coarse-grained misalignment in video-paragraph contrast, Norton filters out the irrelevant clips and captions through an alignable prompt bucket and realigns asynchronous clip-caption pairs based on transport distance. To address the fine-grained misalignment, Norton incorporates a soft-maximum operator to identify crucial words and key frames. Additionally, Norton exploits the potential faulty negative samples in clip-caption contrast by rectifying the alignment target with OT assignment to ensure precise temporal modeling. Extensive experiments on video retrieval, videoQA, and action segmentation verify the effectiveness of our method. Code is available at https://lin-yijie.github.io/projects/Norton.

Ling Team, Anqi Shen, Baihui Li et al.

We present Ring-1T, the first open-source, state-of-the-art thinking model with a trillion-scale parameter. It features 1 trillion total parameters and activates approximately 50 billion per token. Training such models at a trillion-parameter scale introduces unprecedented challenges, including train-inference misalignment, inefficiencies in rollout processing, and bottlenecks in the RL system. To address these, we pioneer three interconnected innovations: (1) IcePop stabilizes RL training via token-level discrepancy masking and clipping, resolving instability from training-inference mismatches; (2) C3PO++ improves resource utilization for long rollouts under a token budget by dynamically partitioning them, thereby obtaining high time efficiency; and (3) ASystem, a high-performance RL framework designed to overcome the systemic bottlenecks that impede trillion-parameter model training. Ring-1T delivers breakthrough results across critical benchmarks: 93.4 on AIME-2025, 86.72 on HMMT-2025, 2088 on CodeForces, and 55.94 on ARC-AGI-1. Notably, it attains a silver medal-level result on the IMO-2025, underscoring its exceptional reasoning capabilities. By releasing the complete 1T parameter MoE model to the community, we provide the research community with direct access to cutting-edge reasoning capabilities. This contribution marks a significant milestone in democratizing large-scale reasoning intelligence and establishes a new baseline for open-source model performance.

Ling Team, Caizhi Tang, Chilin Fu et al.

This technical report presents Ring-Lite-Distill, a lightweight reasoning model derived from our open-source Mixture-of-Experts (MoE) Large Language Models (LLMs) Ling-Lite. This study demonstrates that through meticulous high-quality data curation and ingenious training paradigms, the compact MoE model Ling-Lite can be further trained to achieve exceptional reasoning capabilities, while maintaining its parameter-efficient architecture with only 2.75 billion activated parameters, establishing an efficient lightweight reasoning architecture. In particular, in constructing this model, we have not merely focused on enhancing advanced reasoning capabilities, exemplified by high-difficulty mathematical problem solving, but rather aimed to develop a reasoning model with more comprehensive competency coverage. Our approach ensures coverage across reasoning tasks of varying difficulty levels while preserving generic capabilities, such as instruction following, tool use, and knowledge retention. We show that, Ring-Lite-Distill's reasoning ability reaches a level comparable to DeepSeek-R1-Distill-Qwen-7B, while its general capabilities significantly surpass those of DeepSeek-R1-Distill-Qwen-7B. The models are accessible at https://huggingface.co/inclusionAI

Xuechen Mu, Zhenyu Huang, Kewei Li et al.

Recent advancements in feature representation and dimension reduction have highlighted their crucial role in enhancing the efficacy of predictive modeling. This work introduces TemporalPaD, a novel end-to-end deep learning framework designed for temporal pattern datasets. TemporalPaD integrates reinforcement learning (RL) with neural networks to achieve concurrent feature representation and feature reduction. The framework consists of three cooperative modules: a Policy Module, a Representation Module, and a Classification Module, structured based on the Actor-Critic (AC) framework. The Policy Module, responsible for dimensionality reduction through RL, functions as the actor, while the Representation Module for feature extraction and the Classification Module collectively serve as the critic. We comprehensively evaluate TemporalPaD using 29 UCI datasets, a well-known benchmark for validating feature reduction algorithms, through 10 independent tests and 10-fold cross-validation. Additionally, given that TemporalPaD is specifically designed for time series data, we apply it to a real-world DNA classification problem involving enhancer category and enhancer strength. The results demonstrate that TemporalPaD is an efficient and effective framework for achieving feature reduction, applicable to both structured data and sequence datasets. The source code of the proposed TemporalPaD is freely available as supplementary material to this article and at http://www.healthinformaticslab.org/supp/.

Qiuhua Huang, Renke Huang, Weituo Hao et al.

Power system emergency control is generally regarded as the last safety net for grid security and resiliency. Existing emergency control schemes are usually designed off-line based on either the conceived "worst" case scenario or a few typical operation scenarios. These schemes are facing significant adaptiveness and robustness issues as increasing uncertainties and variations occur in modern electrical grids. To address these challenges, for the first time, this paper developed novel adaptive emergency control schemes using deep reinforcement learning (DRL), by leveraging the high-dimensional feature extraction and non-linear generalization capabilities of DRL for complex power systems. Furthermore, an open-source platform named RLGC has been designed for the first time to assist the development and benchmarking of DRL algorithms for power system control. Details of the platform and DRL-based emergency control schemes for generator dynamic braking and under-voltage load shedding are presented. Extensive case studies performed in both two-area four-machine system and IEEE 39-Bus system have demonstrated the excellent performance and robustness of the proposed schemes.

Ling Team, Bin Hu, Cai Chen et al.

We present Ring-lite, a Mixture-of-Experts (MoE)-based large language model optimized via reinforcement learning (RL) to achieve efficient and robust reasoning capabilities. Built upon the publicly available Ling-lite model, a 16.8 billion parameter model with 2.75 billion activated parameters, our approach matches the performance of state-of-the-art (SOTA) small-scale reasoning models on challenging benchmarks (e.g., AIME, LiveCodeBench, GPQA-Diamond) while activating only one-third of the parameters required by comparable models. To accomplish this, we introduce a joint training pipeline integrating distillation with RL, revealing undocumented challenges in MoE RL training. First, we identify optimization instability during RL training, and we propose Constrained Contextual Computation Policy Optimization(C3PO), a novel approach that enhances training stability and improves computational throughput via algorithm-system co-design methodology. Second, we empirically demonstrate that selecting distillation checkpoints based on entropy loss for RL training, rather than validation metrics, yields superior performance-efficiency trade-offs in subsequent RL training. Finally, we develop a two-stage training paradigm to harmonize multi-domain data integration, addressing domain conflicts that arise in training with mixed dataset. We will release the model, dataset, and code.

Mingyang Zhu, Laigang Guo, Zhenyu Huang et al.

In this paper, we develop a characteristic set (CS)-based method for deriving full-rank equivalence conditions of symbolic matrices over the binary field. Such full-rank conditions are of fundamental importance for many linear coding problems in communication and information theory. Building on the developed CS-based method, we present an algorithm called Binary Characteristic Set for Full Rank (BCSFR), which efficiently derives the full-rank equivalence conditions as the zeros of a series of characteristic sets. In other words, the BCSFR algorithm can characterize all feasible linear coding schemes for certain linear coding problems (e.g., linear network coding and distributed storage coding), where full-rank constraints are imposed on several symbolic matrices to guarantee decodability or other properties of the codes. The derived equivalence conditions can be used to simplify the optimization of coding schemes, since the intractable full-rank constraints in the optimization problem are explicitly characterized by simple triangular-form equality constraints.

Emilie Purvine, Eduardo Cotilla-Sanchez, Mahantesh Halappanavar et al.

Dynamic model reduction in power systems is necessary for improving computational efficiency. Traditional model reduction using linearized models or online analysis is not adequate to capture dynamic behaviors of the power system, especially with the new mix of intermittent generation and intelligent consumption making the power system more dynamic and non-linear. Real-time dynamic model reduction has emerged to fill this important need. This paper explores using clustering techniques to analyze real-time phasor measurements to identify groups of generators with similar behavior, as well as a representative generator from each group for dynamic model reduction. Two clustering techniques -- graph clustering and k-means -- are considered. These techniques are compared with a previously developed dynamic model reduction approach using Singular Value Decomposition. Two sample power grid data sets are used to test these different model reduction techniques. Based on the algorithms' relative performance, recommendations are provided for practical use.

Yu-Hong Yeung, Alex Pothen, Mahantesh Halappanavar et al.

We present AMPS, an augmented matrix approach to update the solution to a linear system of equations when the matrix is modified by a few elements within a principal submatrix. This problem arises in the dynamic security analysis of a power grid, where operators need to perform N - k contingency analysis, i.e., determine the state of the system when exactly k links from N fail. Our algorithms augment the matrix to account for the changes in it, and then compute the solution to the augmented system without refactoring the modified matrix. We provide two algorithms, a direct method, and a hybrid direct-iterative method for solving the augmented system. We also exploit the sparsity of the matrices and vectors to accelerate the overall computation. We analyze the time complexity of both algorithms, and show that it is bounded by the number of nonzeros in a subset of the columns of the Cholesky factor that are selected by the nonzeros in the sparse right-hand-side vector. Our algorithms are compared on three power grids with PARDISO, a parallel direct solver, and CHOLMOD, a direct solver with the ability to modify the Cholesky factors of the matrix. We show that our augmented algorithms outperform PARDISO (by two orders of magnitude), and CHOLMOD (by a factor of up to 5). Further, our algorithms scale better than CHOLMOD as the number of elements updated increases. The solutions are computed with high accuracy. Our algorithms are capable of computing N - k contingency analysis on a 778 thousand bus grid, updating a solution with k = 20 elements in 16 milliseconds on an Intel Xeon processor.