arXiv subjects

Haoran Zhang

Publications and source records attributed to Haoran Zhang.

At least 91 records · Page 5Linked to original sources

arXiv2025

StarBench: A Turn-Based RPG Benchmark for Agentic Multimodal Decision-Making and Information Seeking

Human players do more than press buttons: they ground what they see on screen into precise keyboard-mouse actions and, when stuck, they seek information before trying again. We ask whether current vision-language models (VLMs) can do the same. Despite encouraging results under simplified control or tool scaffolds, human-like play in a real client - mapping raw screenshots to temporally coherent low-level actions while deciding when to ask for guidance - remains an open challenge. We introduce StarBench, a turn-based RPG benchmark derived from Honkai: Star Rail that targets these two human-like competencies: multimodal decision-making from pixels to actions and agentic information seeking. StarBench standardizes evaluation across eight combat tasks and two regimes with shared tasks and metrics: (i) direct control, where agents receive only screenshots and must emit low-level primitives (click and keypress) with no semantic hints; and (ii) tool-assisted control, where higher-level intents can be mapped to primitives by detectors and OCR outputs provide optional textualized observations to ease UI grounding. To mirror human practice, StarBench also includes an ask-or-act diagnostic that measures whether and when agents choose to request brief guidance before proceeding, and how that choice affects subsequent performance. We report reference baselines for contemporary VLMs and a human reference. Results expose sizable gaps in perception-to-control fidelity in the direct regime, while showing that judicious information seeking correlates with improved success, establishing StarBench as a reproducible yardstick for agentic information seeking and multimodal decision-making in real-client play.

cs.AI↗

arXiv2025

Antarctic Infrared Binocular Telescope. I. System Overview, Laboratory Testing, and On-Sky Performance Evaluation

Infrared time-domain surveys remain significantly underdeveloped compared with their optical counterparts. We have developed the Antarctic Infrared Binocular Telescope (AIRBT) to study the dynamic infrared sky at Dome A, Antarctica, taking advantage of the superb infrared observational conditions at this site. AIRBT consists of two identical 15 cm f/3 optical tube assemblies and two cost-effective indium gallium arsenide (InGaAs) cameras equipped with J and H filters, respectively. The cameras have 640 x 512 pixels with a size of 15 micrometers, providing a scale of 6.9 arcseconds per pixel and a field of view of 1.22 x 0.97 square degrees. We characterize the performance of the InGaAs cameras, including bias, readout noise, dark current, nonlinearity, and photon transfer curve. Our analysis highlights the distinct behaviors of InGaAs cameras compared with charge-coupled devices (CCDs). The bias and readout noise show temperature dependence, and the noise measured from the photon transfer curves has additional components that increase with exposure time. On-sky tests were conducted in October 2022 including system calibration, limiting depth, and photometric precision. For a single 3-second exposure, we achieved 5-sigma limiting magnitudes of 11.2 mag (Vega system) in J band and 9.7 mag in H band. The best photometric precision reached 20 millimagnitudes at the bright end, which could be further improved to sub-percent levels through image stacking. AIRBT was installed at Dome A in January 2023, and scientific observations began as soon as darkness set in.

astro-ph.IM↗

arXiv2025

SAGE-Music: Low-Latency Symbolic Music Generation via Attribute-Specialized Key-Value Head Sharing

Low-latency symbolic music generation is essential for real-time improvisation and human-AI co-creation. Existing transformer-based models, however, face a trade-off between inference speed and musical quality. Traditional acceleration techniques such as embedding pooling significantly degrade quality, while recently proposed Byte Pair Encoding (BPE) methods - though effective on single-track piano data - suffer large performance drops in multi-track settings, as revealed by our analysis. We propose Attribute-Specialized Key-Value Head Sharing (AS-KVHS), adapted to music's structured symbolic representation, achieving about 30% inference speedup with only a negligible (about 0.4%) quality drop in objective evaluations and slight improvements in subjective listening tests. Our main contributions are (1) the first systematic study of BPE's generalizability in multi-track symbolic music, and (2) the introduction of AS-KVHS for low-latency symbolic music generation. Beyond these, we also release SAGE-Music, an open-source benchmark that matches or surpasses state-of-the-art models in generation quality.

cs.SD↗

arXiv2025

The Role of Computing Resources in Publishing Foundation Model Research

Cutting-edge research in Artificial Intelligence (AI) requires considerable resources, including Graphics Processing Units (GPUs), data, and human resources. In this paper, we evaluate of the relationship between these resources and the scientific advancement of foundation models (FM). We reviewed 6517 FM papers published between 2022 to 2024, and surveyed 229 first-authors to the impact of computing resources on scientific output. We find that increased computing is correlated with national funding allocations and citations, but our findings don't observe the strong correlations with research environment (academic or industrial), domain, or study methodology. We advise that individuals and institutions focus on creating shared and affordable computing opportunities to lower the entry barrier for under-resourced researchers. These steps can help expand participation in FM research, foster diversity of ideas and contributors, and sustain innovation and progress in AI. The data will be available at: https://mit-calc.csail.mit.edu/

cs.CY↗

arXiv2025

Probing the cosmological 21~cm global signal from the Antarctic ice sheet

The redshifted 21 cm line, arising from neutral hydrogen, offers a unique probe into the intergalactic medium and the first stars and galaxies formed in the early universe. However, detecting this signal is a challenging task because of artificial radio-frequency interference (RFI) and systematic errors such as ground effects. The interior of the Antarctic continent provides an excellent location to make such observations, with minimal RFI and relatively stable foreground signals. Moreover, a flat plateau in central Antarctica, with an ice cap over 2000 m deep, will show less ground reflection of radio waves, reducing the signal complexity in the area around the probing antenna. It may be advantageous to perform cosmological 21 cm experiments in Antarctica, and a 21 cm Antarctic global spectrum experiment can potentially be deployed on the Antarctic ice cap. We have performed preliminary instrumental design, system calibration, and implementation of such an instrument optimized for extreme cold and capable of long-term autonomous operation. This system shows the ability to effectively detect the 21~cm signal, confirming Antarctica as an excellent observational site for radio cosmology.

astro-ph.IM↗

arXiv2025

VCoT-Grasp: Grasp Foundation Models with Visual Chain-of-Thought Reasoning for Language-driven Grasp Generation

Robotic grasping is one of the most fundamental tasks in robotic manipulation, and grasp detection/generation has long been the subject of extensive research. Recently, language-driven grasp generation has emerged as a promising direction due to its practical interaction capabilities. However, most existing approaches either lack sufficient reasoning and generalization capabilities or depend on complex modular pipelines. Moreover, current grasp foundation models tend to overemphasize dialog and object semantics, resulting in inferior performance and restriction to single-object grasping. To maintain strong reasoning ability and generalization in cluttered environments, we propose VCoT-Grasp, an end-to-end grasp foundation model that incorporates visual chain-of-thought reasoning to enhance visual understanding for grasp generation. VCoT-Grasp adopts a multi-turn processing paradigm that dynamically focuses on visual inputs while providing interpretable reasoning traces. For training, we refine and introduce a large-scale dataset, VCoT-GraspSet, comprising 167K synthetic images with over 1.36M grasps, as well as 400+ real-world images with more than 1.2K grasps, annotated with intermediate bounding boxes. Extensive experiments on both VCoT-GraspSet and real robot demonstrate that our method significantly improves grasp success rates and generalizes effectively to unseen objects, backgrounds, and distractors. More details can be found at https://zhanghr2001.github.io/VCoT-Grasp.github.io.

cs.RO↗

arXiv2025

Data-Driven Long-Term Asset Allocation with Tsallis Entropy Regularization

This paper addresses the problem of dynamic asset allocation under uncertainty, which can be formulated as a linear quadratic (LQ) control problem with multiplicative noise. To handle exploration exploitation trade offs and induce sparse control actions, we introduce Tsallis entropy as a regularization term. We develop an entropy regularized policy iteration scheme and provide theoretical guarantees for its convergence. For cases where system dynamics are unknown, we further propose a fully data driven algorithm that estimates Q functions using an instrumental variable least squares approach, allowing efficient and stable policy updates. Our framework connects entropy-regularized stochastic control with model free reinforcement learning, offering new tools for intelligent decision making in finance and automation.

math.OC↗

arXiv2025

Towards an AI Musician: Synthesizing Sheet Music Problems for Musical Reasoning

Enhancing the ability of Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) to interpret sheet music is a crucial step toward building AI musicians. However, current research lacks both evaluation benchmarks and training data for sheet music reasoning. Inspired by mathematics, where simple operations yield infinite verifiable problems, we introduce a novel approach that treats core music theory rules, such as those governing beats and intervals, as programmatic functions to systematically synthesize a vast and diverse corpus of sheet music reasoning problems. This approach allows us to introduce a data synthesis framework that generates verifiable sheet music questions in both textual and visual modalities, leading to the Synthetic Sheet Music Reasoning Benchmark (SSMR-Bench) and a complementary training set. Evaluation results on SSMR-Bench highlight the key role reasoning plays in interpreting sheet music, while also pointing out the ongoing challenges in understanding sheet music in a visual format. By leveraging synthetic data for RLVR, all models show significant improvements on the SSMR-Bench. Additionally, they also demonstrate considerable advancements on previously established human-crafted benchmarks, such as MusicTheoryBench and the music subset of MMMU. Finally, our results show that the enhanced reasoning ability can also facilitate music composition.

cs.CL↗

arXiv2025

Nontrivial topology in one- and two-dimensional asymmetric systems with chiral boundary states

Symmetry plays an important role in the topological band theory. In contrary, study on the topological properties of the asymmetric systems is rather limited, especially in higher-dimensional systems. In this work, we explore a new theory to study the topology in various one-dimensional (1D) and two-dimensional (2D) asymmetric systems with chiral boundary states. Starting from the simple SSHm model, we show the chiral topology of its edge states by redefining sublattices. Meanwhile, based on its Rice-Mele-like effective Hamiltonian, a new topological invariant $\bar{Z}$ can be defined and the bulk-edge correspondence is established. With this clear physical picture, our theory can be extended to the more complex asymmetric ladder models, or even the 2D asymmetric systems. In the 2D BBH3 model, new chiral corner states with redefined lattices are found based on our method. These corner states are independent of any spatial symmetry and exhibit the characteristics of topological bound states in the continuum (TBICs). Moreover, the topological invariant can be calculated by introducing $\bar{Z}$ into 2D. At last, we propose an acoustic experiment of the BBH3 model where chiral corner states are numerically observed. Our work exhibits a new approach to study the topological properties of asymmetric systems. By redefining sublattices, we find that the models with entirely different structures might share the same topological origins.

cond-mat.mes-hall↗

arXiv2025

A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning

Robotic real-world reinforcement learning (RL) with vision-language-action (VLA) models is bottlenecked by sparse, handcrafted rewards and inefficient exploration. We introduce VLAC, a general process reward model built upon InternVL and trained on large scale heterogeneous datasets. Given pairwise observations and a language goal, it outputs dense progress delta and done signal, eliminating task-specific reward engineering, and supports one-shot in-context transfer to unseen tasks and environments. VLAC is trained on vision-language datasets to strengthen perception, dialogic and reasoning capabilities, together with robot and human trajectories data that ground action generation and progress estimation, and additionally strengthened to reject irrelevant prompts as well as detect regression or stagnation by constructing large numbers of negative and semantically mismatched samples. With prompt control, a single VLAC model alternately generating reward and action tokens, unifying critic and policy. Deployed inside an asynchronous real-world RL loop, we layer a graded human-in-the-loop protocol (offline demonstration replay, return and explore, human guided explore) that accelerates exploration and stabilizes early learning. Across four distinct real-world manipulation tasks, VLAC lifts success rates from about 30\% to about 90\% within 200 real-world interaction episodes; incorporating human-in-the-loop interventions yields a further 50% improvement in sample efficiency and achieves up to 100% final success.

cs.RO↗

arXiv2025

4D-PreNet: A Unified Preprocessing Framework for 4D-STEM Data Analysis

Automated experimentation with real time data analysis in scanning transmission electron microscopy (STEM) often require end-to-end framework. The four-dimensional scanning transmission electron microscopy (4D-STEM) with high-throughput data acquisition has been constrained by the critical bottleneck results from data preprocessing. Pervasive noise, beam center drift, and elliptical distortions during high-throughput acquisition inevitably corrupt diffraction patterns, systematically biasing quantitative measurements. Yet, conventional correction algorithms are often material-specific and fail to provide a robust, generalizable solution. In this work, we present 4D-PreNet, an end-to-end deep-learning pipeline that integrates attention-enhanced U-Net and ResNet architectures to simultaneously perform denoising, center correction, and elliptical distortion calibration. The network is trained on large, simulated datasets encompassing a wide range of noise levels, drift magnitudes, and distortion types, enabling it to generalize effectively to experimental data acquired under varying conditions. Quantitative evaluations demonstrate that our pipeline reduces mean squared error by up to 50% during denoising and achieves sub-pixel center localization in the center detection task, with average errors below 0.04 pixels. The outputs are bench-marked against traditional algorithms, highlighting improvements in both noise suppression and restoration of diffraction patterns, thereby facilitating high-throughput, reliable 4D-STEM real-time analysis for automated characterization.

cs.CV↗

arXiv2025

Evidence of scaling advantage on an NP-Complete problem with enhanced quantum solvers

Achieving quantum advantage remains a key milestone in the noisy intermediate-scale quantum era. Without rigorous complexity proofs, scaling advantage-where quantum resource requirements grow more slowly than their classical counterparts-serves as the primary indicator. However, direct applications of quantum optimization algorithms to classically intractable problems have yet to demonstrate this advantage. To address this challenge, we develop enhanced quantum solvers for the NP-complete one-in-three Boolean satisfiability problem. We propose a restricting space reduction algorithm (RSRA) that achieves optimal search space dimensionality, thereby reducing both qubits and time complexity for various quantum solvers. Extensive numerical investigations on problem instances with up to 65 variables demonstrate that our enhanced quantum approximate optimization algorithm (QAOA) and quantum adiabatic algorithm (QAA)-based solvers outperform state-of-the-art classical solvers, with the QAA-based solver providing a lower bound for our method while exhibiting scaling advantage. Furthermore, we experimentally implement our enhanced solvers on a superconducting quantum processor with 13 qubits, confirming the predicted performance improvements. Collectively, our results provide empirical evidence of quantum speedup for an NP-complete problem.

quant-ph↗

arXiv2025

Dual-disentangle Framework for Diversified Sequential Recommendation

Sequential recommendation predicts user preferences over time and has achieved remarkable success. However, the growing length of user interaction sequences and the complex entanglement of evolving user interests and intentions introduce significant challenges to diversity. To address these, we propose a model-agnostic Dual-disentangle framework for Diversified Sequential Recommendation (DDSRec). The framework refines user interest and intention modeling by adopting disentangling perspectives in interaction modeling and representation learning, thereby balancing accuracy and diversity in sequential recommendations. Extensive experiments on multiple public datasets demonstrate the effectiveness and superiority of DDSRec in terms of accuracy and diversity for sequential recommendations.

cs.IR↗

arXiv2025

SAM: A Stability-Aware Cache Manager for Multi-Tenant Embedded Databases

The co-location of multiple database instances on resource constrained edge nodes creates significant cache contention, where traditional schemes are inefficient and unstable under dynamic workloads. To address this, we present SAM(a Stability-Aware Manager), an autonomic cache manager that establishes decision stability as a first-class design principle. It achieves this through its core control policy, AURA(Autonomic Utility-balancing Resource Allocator), which resolves the classic exploitation-exploration dilemma by synthesizing two orthogonal factors: the H-factor, representing proven historical efficiency (exploitation), and the V-factor, for estimated marginal gain (exploration). Through this practical synthesis and adaptive control, SAM achieves sustained high performance with strategic stability and robustness in volatile conditions. Extensive experiments against 14 diverse baselines demonstrate SAM's superiority. It achieves top-tier throughput while being uniquely resilient to complex workload shifts and adversarial workloads like cache pollution. Furthermore, its decision latency is highly scalable, remaining nearly constant as the system grows to 120 databases. Crucially, SAM achieves superior decision stability -- maintaining consistent optimization directions despite noise, avoiding performance oscillations while ensuring predictable Quality of Service. These results prove that a principled, stability-aware design is essential for sustained high performance in real-world, large-scale systems.

cs.DB↗

arXiv2025

Player-Centric Multimodal Prompt Generation for Large Language Model Based Identity-Aware Basketball Video Captioning

Existing sports video captioning methods often focus on the action yet overlook player identities, limiting their applicability. Although some methods integrate extra information to generate identity-aware descriptions, the player identities are sometimes incorrect because the extra information is independent of the video content. This paper proposes a player-centric multimodal prompt generation network for identity-aware sports video captioning (LLM-IAVC), which focuses on recognizing player identities from a visual perspective. Specifically, an identity-related information extraction module (IRIEM) is designed to extract player-related multimodal embeddings. IRIEM includes a player identification network (PIN) for extracting visual features and player names, and a bidirectional semantic interaction module (BSIM) to link player features with video content for mutual enhancement. Additionally, a visual context learning module (VCLM) is designed to capture the key video context information. Finally, by integrating the outputs of the above modules as the multimodal prompt for the large language model (LLM), it facilitates the generation of descriptions with player identities. To support this work, we construct a new benchmark called NBA-Identity, a large identity-aware basketball video captioning dataset with 9,726 videos covering 9 major event types. The experimental results on NBA-Identity and VC-NBA-2022 demonstrate that our proposed model achieves advanced performance. Code and dataset are publicly available at https://github.com/Zeyu1226-mt/LLM-IAVC.

cs.CV↗

arXiv2025

De novo design of alpha-helical peptide amphiphiles repairing fragmented collagen type I via supramolecular co-assembly

The hierarchical triple-helix structure of collagen type I, Col I, is essential for extracellular matrix support and integrity. However, current reconstruction strategies face challenges such as chain mismatch, preventing proper fibril formation. Here, we report a supramolecular co-assembly strategy using a de novo-designed alpha-helical peptide amphiphile (APA) of just seven amino acids. The APA features a hydrophobic palmitic acid tail, which stabilizes the helical structure and promotes co-assembly upon interaction with complementary molecular structures. This minimal design enables selective recognition of fragmented collagen (FC), restoring triple-helix conformation and guiding fibre formation. We applied this mechanism to engineer FC-rich nanofat (NF) into a mechanically reinforced biomaterial. Integration of APA-NF with coaxial 3D printing enabled spatial control of structure and function. In a porcine model, this platform enhanced in situ vascularized adipose tissue regeneration. Our results demonstrate that hierarchical reconstruction of collagen via peptide-guided supramolecular assembly offers a promising strategy for soft tissue repair.

physics.chem-ph↗

arXiv2025

4D-MISR: A unified model for low-dose super-resolution imaging via feature fusion

While electron microscopy offers crucial atomic-resolution insights into structure-property relationships, radiation damage severely limits its use on beam-sensitive materials like proteins and 2D materials. To overcome this challenge, we push beyond the electron dose limits of conventional electron microscopy by adapting principles from multi-image super-resolution (MISR) that have been widely used in remote sensing. Our method fuses multiple low-resolution, sub-pixel-shifted views and enhances the reconstruction with a convolutional neural network (CNN) that integrates features from synthetic, multi-angle observations. We developed a dual-path, attention-guided network for 4D-STEM that achieves atomic-scale super-resolution from ultra-low-dose data. This provides robust atomic-scale visualization across amorphous, semi-crystalline, and crystalline beam-sensitive specimens. Systematic evaluations on representative materials demonstrate comparable spatial resolution to conventional ptychography under ultra-low-dose conditions. Our work expands the capabilities of 4D-STEM, offering a new and generalizable method for the structural analysis of radiation-vulnerable materials.

cs.CV↗

arXiv2025

A Literature Review on Simulation in Conversational Recommender Systems

Conversational Recommender Systems (CRSs) have garnered attention as a novel approach to delivering personalized recommendations through multi-turn dialogues. This review developed a taxonomy framework to systematically categorize relevant publications into four groups: dataset construction, algorithm design, system evaluation, and empirical studies, providing a comprehensive analysis of simulation methods in CRSs research. Our analysis reveals that simulation methods play a key role in tackling CRSs' main challenges. For example, LLM-based simulation methods have been used to create conversational recommendation data, enhance CRSs algorithms, and evaluate CRSs. Despite several challenges, such as dataset bias, the limited output flexibility of LLM-based simulations, and the gap between text semantic space and behavioral semantics, persist due to the complexity in Human-Computer Interaction (HCI) of CRSs, simulation methods hold significant potential for advancing CRS research. This review offers a thorough summary of the current research landscape in this domain and identifies promising directions for future inquiry.

cs.HC↗