arXiv subjects

Chunyu Zhao

Publications and source records attributed to Chunyu Zhao.

11 recordsLinked to original sources

arXiv2026

The MiniMax-M2 Series: Mini Activations Unleashing Max Real-World Intelligence

We introduce the MiniMax-M2 series, a family of Mixture-of-Experts language models built around the principle that mini activations can unleash maximum real-world intelligence. The flagship M2 contains 229.9B total parameters with only 9.8B activated per token. Designed end-to-end for agentic deployment, the M2 series rests on three components: (i) agent-driven data pipelines producing large-scale, verifiable trajectories across agentic coding and agentic cowork, each grounded in an executable workspace and an artifact-aligned reward; (ii) Forge, a scalable agent-native RL system that adapts to long-horizon agent trajectories, paired with windowed-FIFO scheduling, prefix-tree merging, inference optimization, and a clean training-inference-agent decoupling that supports both white-box and black-box agents; (iii) the latest M2.7 checkpoint takes an early step toward self-evolution -- autonomously debugging training runs and modifying its own scaffold. Across M2 through M2.7, this combination translates a mini-activation footprint into frontier-tier performance on agentic coding, deep search, office-task, and reasoning benchmarks.

cs.AI↗

arXiv2026

Controlled epitaxy of room-temperature quantum emitters in gallium nitride

The ability to generate quantum light at room temperature on a mature semiconductor platform opens up new possibilities for quantum technologies. Heteroepitaxial growth of gallium nitride on silicon substrates offers the opportunity to leverage existing expertise and wafer-scale manufacturing to integrate bright quantum emitters in this material within cavities, diodes, and photonic circuits. Until now, it has only been possible to grow GaN QEs at uncontrolled depths on sapphire substrates, which is disadvantageous for potential device architectures. Here, we report a method to produce GaN QEs by metal-organic vapor phase epitaxy at a controlled depth in the crystal through the application of silane treatment and subsequent growth of 3D islands. We demonstrate this process on highly technologically relevant silicon substrates, producing room-temperature QEs with a high Debye Waller factor and strongly anti-bunched emission.

cond-mat.mtrl-sci↗

arXiv2025

Physical Computing at the Data Processing Inequality Limit

Wave-physics-based intelligent sensing has driven multidisciplinary applications from smart industries to decision-making systems. Traditional sensing paradigms transform physical waveforms into human-understandable intermediate representations through preprocessing. Such transformations inherently cause information loss owing to data processing inequality (DPI). Here, we established a theoretical framework for physical computing at the DPI upper limit. Physical computing avoids information loss during preprocessing by directly extracting information from physical waveforms, achieving the theoretical maximum of accessible information as determined by the DPI. Furthermore, physical computing comprehensively utilizes multiple dimensions of physical waveforms, thereby enhancing the upper limit of information capture capability. Electromagnetic sensing experiments have demonstrated that physical computing can achieve 100% sensing accuracy, substantially outperforming traditional sensing paradigms. The proposed theoretical framework of physical computing offers a promising path towards enhancing the information-capture capability of next-generation intelligent sensing systems.

physics.app-ph↗

arXiv2025

Efficient Cloud-Edge-Device Query Execution Based on Collaborative Scan Operator

In cloud-edge-device (CED) collaborative query (CQ) processing, by leveraging CED collaboration, the advantages of both cloud computing and edge resources can be fully integrated. However, it is difficult to implement collaborative operators that can flexibly switch between the cloud and the edge during query execution. Thus, in this paper, we aim to improve the query performance when the edge resources reach a bottleneck. To achieve seamless switching of query execution between the cloud and edge, we propose a CQ processing method by establishing a CED collaborative framework based on the collaborative scan operator, so that query execution can be transferred to the cloud at any time when the edge resources are saturated. Extensive experiments show that, under sufficient network download bandwidth, the CED collaborative scan operator can effectively alleviate the performance degradation of scan operators caused by high I/O load and CPU wait time at the edge. It also achieves balanced resource scheduling between the cloud and edge.

cs.DB↗

arXiv2025

SalM$^{2}$: An Extremely Lightweight Saliency Mamba Model for Real-Time Cognitive Awareness of Driver Attention

Driver attention recognition in driving scenarios is a popular direction in traffic scene perception technology. It aims to understand human driver attention to focus on specific targets/objects in the driving scene. However, traffic scenes contain not only a large amount of visual information but also semantic information related to driving tasks. Existing methods lack attention to the actual semantic information present in driving scenes. Additionally, the traffic scene is a complex and dynamic process that requires constant attention to objects related to the current driving task. Existing models, influenced by their foundational frameworks, tend to have large parameter counts and complex structures. Therefore, this paper proposes a real-time saliency Mamba network based on the latest Mamba framework. As shown in Figure 1, our model uses very few parameters (0.08M, only 0.09~11.16% of other models), while maintaining SOTA performance or achieving over 98% of the SOTA model's performance.

cs.CV↗

arXiv2024

CONCERTO: Complex Query Execution Mechanism-Aware Learned Cost Estimation

With the growing demand for massive data analysis, many DBMSs have adopted complex underlying query execution mechanisms, including vectorized operators, parallel execution, and dynamic pipeline modifications. However, there remains a lack of targeted Query Performance Prediction (QPP) methods for these complex execution mechanisms and their interactions, as most existing approaches focus on traditional tree-shaped query plans and static serial executors. To address this challenge, this paper proposes CONCERTO, a Complex query executiON meChanism-awaE leaRned cosT estimatiOn method. CONCERTO first establishes independent resource cost models for each physical operator. It then constructs a Directed Acyclic Graph (DAG) consisting of a dataflow tree backbone and resource competition relationships among concurrent operators. After calibrating the cost impact of parallel operator execution using Graph Attention Networks (GATs) with additional attention mechanisms, CONCERTO extracts and aggregates cost vector trees through Temporal Convolutional Networks (TCNs), ultimately achieving effective query performance prediction. Experimental results demonstrate that CONCERTO achieves higher prediction accuracy than existing methods.

cs.DB↗

arXiv2024

AIM 2024 Challenge on Video Saliency Prediction: Methods and Results

This paper reviews the Challenge on Video Saliency Prediction at AIM 2024. The goal of the participants was to develop a method for predicting accurate saliency maps for the provided set of video sequences. Saliency maps are widely exploited in various applications, including video compression, quality assessment, visual perception studies, the advertising industry, etc. For this competition, a previously unused large-scale audio-visual mouse saliency (AViMoS) dataset of 1500 videos with more than 70 observers per video was collected using crowdsourced mouse tracking. The dataset collection methodology has been validated using conventional eye-tracking data and has shown high consistency. Over 30 teams registered in the challenge, and there are 7 teams that submitted the results in the final phase. The final phase solutions were tested and ranked by commonly used quality metrics on a private test subset. The results of this evaluation and the descriptions of the solutions are presented in this report. All data, including the private test subset, is made publicly available on the challenge homepage - https://challenges.videoprocessing.ai/challenges/video-saliency-prediction.html.

cs.CV↗

arXiv2024

Hexagonal network of photocurrent enhancement in few-layer graphene/InGaN quantum dot junctions

Strain in two-dimensional (2D) materials has attracted particular attention owing to the remarkable modification of electronic and optical properties. However, emergent electromechanical phenomena and hidden mechanisms, such as strain-superlattice-induced topological states or flexoelectricity under strain gradient, remain under debate. Here, using scanning photocurrent microscopy, we observe significant photocurrent enhancement in hybrid vertical junction devices made of strained few-layer graphene and InGaN quantum dots. Optoelectronic response and photoluminescence measurements demonstrate a possible mechanism closely tied to the flexoelectric effect in few-layer graphene, where the strain can induce a lateral built-in electric field and assist the separation of electron-hole pairs. Photocurrent mapping reveals an unprecedentedly ordered hexagonal network, suggesting the potential to create a superlattice by strain engineering. Our work provides insights into optoelectronic phenomena in the presence of strain and paves the way for practical applications associated with strained 2D materials.

cond-mat.mes-hall↗

arXiv2021

Auto-KWS 2021 Challenge: Task, Datasets, and Baselines

Auto-KWS 2021 challenge calls for automated machine learning (AutoML) solutions to automate the process of applying machine learning to a customized keyword spotting task. Compared with other keyword spotting tasks, Auto-KWS challenge has the following three characteristics: 1) The challenge focuses on the problem of customized keyword spotting, where the target device can only be awakened by an enrolled speaker with his specified keyword. The speaker can use any language and accent to define his keyword. 2) All dataset of the challenge is recorded in realistic environment. It is to simulate different user scenarios. 3) Auto-KWS is a "code competition", where participants need to submit AutoML solutions, then the platform automatically runs the enrollment and prediction steps with the submitted code.This challenge aims at promoting the development of a more personalized and flexible keyword spotting system. Two baseline systems are provided to all participants as references.

cs.SD↗

arXiv2018

Interfacial bound exciton state in a hybrid structure of monolayer WS2 and InGaN quantum dots

Van der Waals heterostructures usually formed using atomic thin transition metal dichalcogenides (TMDCs) with a direct bandgap in the near-infrared to the visible range are promising candidates for low-dimensional optoelectronic applications. The interlayer interaction or coupling between two-dimensional (2D) layer and the substrate or between adjacent 2D layers plays an important role in modifying properties of the individual 2D material or device performances through Coulomb interaction or forming interlayer excitons. Here, we report the realization of quasi zero-dimensional (0D) photon emission of WS2 in a coupled hybrid structure of monolayer WS2 and InGaN quantum dots (QDs). An interfacial bound exciton, i.e., coupling between excitons in WS2 and the electrons in QDs has been identified. The emission of this interfacial bound exciton inherits the 0-D confinement of QDs as well as the spin-valley physics of excitons in monolayer WS2. The effective coupling between 2D materials and conventional semiconductors observed in this work provides an effective way to realize 0D emission of 2D materials and opens the potential of compact on-chip integration of valleytronics and conventional electronics/optoelectronics.

cond-mat.mes-hall↗

arXiv2014

PDBCirclePlot: A Novel Visualization Method for Protein Structures

Interactive molecular graphics applications facilitate analysis of three dimensional protein structures. Naturally, non-interactive 2-D snapshots of the protein structures do not convey the same level of geometric detail. Several 2-D visualization methods have been in use to summarize structural information, including contact maps and 2-D cartoon views. We present a new approach for 2-D visualization of protein structures where amino acid residues are displayed on a circle and spatially close residues are depicted by links. Furthermore, residue-specific properties, such as conservation, accessibility, temperature factor, can be displayed as plots on the same circular view.

q-bio.QM↗