SearcharxivSearch

arXiv subjects

Yifang Wang

Publications and source records attributed to Yifang Wang.

At least 19 recordsLinked to original sources

HypoForge: A Self-Improving Multi-Agent Framework for Automated Hypothesis Generation and Testing via Scientific Skill Learning

Large language models (LLMs) have enabled AI scientist systems to automate scientific discovery, yet existing approaches most rely on static prompting or fixed workflows and fail to accumulate experience for continual improvement. We propose HypoForge, an experience-guided multi-agent framework that learns reusable scientific skills for automated hypothesis generation and hypothesis testing. HypoForge is built on the observation that these two stages involve different supervision signals. For hypothesis generation, where explicit feedback is unavailable, HypoForge adopts an adversarial generator--discriminator mechanism to improve reasoning through comparative critique. For hypothesis testing, where empirical feedback is available, HypoForge learns testing skills from execution outcomes and ground-truth results. By matching skill learning strategies with stage-specific supervision, HypoForge enables continual improvement without fine-tuning foundation models. Experiments on hypothesis generation and testing benchmarks show that HypoForge consistently outperforms existing AI scientist frameworks and skill-level variants. Further analysis demonstrates the effectiveness of the proposed stage-specific skill learning paradigms.

cs.MA

ArtAnno: Annotating Implicit Semantics in Artworks through LLM Agent-Driven Bidirectional Human-AI Augmentation

High-quality annotation of artworks is essential for computational art research, yet extracting implicit semantics remains challenging due to the reliance on culturally grounded meanings and deep contextual knowledge behind the images. Current AI-assisted annotation tools often lack assistance or rely on one-way workflows where experts have to perform extra manual calibrations to improve AI models, resulting in limited efficiency. To address this, we propose Bidirectional Human-AI Augmentation(BiHAA), a closed-loop framework in which skills and domain knowledge base evolve through real-time interaction and bidirectional HAI augmentation. Informed by a formative study with 20 artwork annotators from different backgrounds, we implement this framework in ArtAnno, an artwork annotation system driven by a multi-agent architecture. The system includes a Proactive Agentic Support Module, where AI augments humans through semantic mining and label suggestion, and an Interaction-Driven Evolution Module, where human expertise continuously enhances the AI through distilling annotation trajectories into reusable experience. Evaluation through a user study and two case studies demonstrates that our framework and system improve annotation efficiency, enable knowledge accumulation, and reduce the effort of information seeking and verification for annotators with limited domain expertise. We conclude by discussing broader implications and future directions.

cs.HC

MedFM-Robust: Benchmarking Robustness of Medical Foundation Models

Medical foundation models have achieved remarkable clinical performance, yet their robustness under real-world perturbations remains underexplored. We present a robustness benchmark comprising 40 perturbation types (12 base, 28 medical-specific) across eight imaging modalities, evaluating five VLMs (LLaVA-Med, MedGemma, MedGemma-1.5, Gemini-2.5-flash and GPT-4o-mini) on VQA, visual grounding, and captioning, alongside two segmentation models (MedSAM, SAM-Med2D) with five fine-tuning strategies. Our findings reveal: (1) Fine-tuning strategy dominates robustness, with LoRA exhibiting nearly double the degradation of full fine-tuning, while SAM-Med2D's Adapter offers favorable efficiency-robustness trade-off. (2) Medical-specific perturbations disproportionately damage segmentation, with 9 of 15 top corruptions being domain-specific. (3) LoRA-tuned visual grounding drops over 40 points, whereas zero-shot captioning remains stable (<7% drop). Zero-shot VQA shows model-dependent robustness--medical models drop under 20% while Gemini-2.5-flash drops 54%. General-purpose VLMs achieve higher VQA accuracy but fail on grounding; among medical VLMs, MedGemma demonstrates the best overall stability. These results provide deployment guidelines and underscore the necessity of domain-specific robustness evaluation for medical AI. Our code is available at: https://abnerai.github.io/MedFM-Robust.

cs.CV

Realistic Detector Geometry Modeling and Its Impact on Event Reconstruction in JUNO

JUNO is designed to determine the neutrino mass ordering with an energy resolution of 3% at 1 MeV. In the real detector, however, deformations of the central stainless-steel structure during installation lead to deviations of the photomultiplier tube (PMT) positions from their design values. Based on the limited survey data of the PMTs and the stainless-steel truss, we perform a correlation analysis of the measured points and propose a method to predict the positions of all PMTs. Using the resulting realistic geometry, we demonstrate that the detector deformation has a negligible effect on the energy reconstruction. In contrast, inaccuracies in the assumed geometry can introduce vertex biases of up to 40 mm. Incorporating the realistic geometry into the calibration-based PMT response model removes this bias and preserves the stability of the reconstruction algorithms.

physics.ins-det

Figures as Interfaces: Toward LLM-Native Artifacts for Scientific Discovery

Large language models (LLMs) are transforming scientific workflows, not only through their generative capabilities but also through their emerging ability to use tools, reason about data, and coordinate complex analytical tasks. Yet in most human-AI collaborations, the primary outputs, figures, are still treated as static visual summaries: once rendered, they are handled by both humans and multimodal LLMs as images to be re-interpreted from pixels or captions. The emergent capabilities of LLMs open an opportunity to fundamentally rethink this paradigm. In this paper, we introduce the concept of LLM-native figures: data-driven artifacts that are simultaneously human-legible and machine-addressable. Unlike traditional plots, each artifact embeds complete provenance: the data subset, analytical operations and code, and visualization specification used to generate it. As a result, an LLM can "see through" the figure--tracing selections back to their sources, generating code to extend analyses, and orchestrating new visualizations through natural-language instructions or direct manipulation. We implement this concept through a hybrid language-visual interface that integrates LLM agents with a bidirectional mapping between figures and underlying data. Using the science of science domain as a testbed, we demonstrate that LLM-native figures can accelerate discovery, improve reproducibility, and make reasoning transparent across agents and users. More broadly, this work establishes a general framework for embedding provenance, interactivity, and explainability into the artifacts of modern research, redefining the figure not as an end product, but as an interface for discovery. For more details, please refer to the demo video available at www.llm-native-figure.com.

cs.HC

SAIF: A Stability-Aware Inference Framework for Medical Image Segmentation with Segment Anything Model

Segment Anything Model (SAM) enable scalable medical image segmentation but suffer from inference-time instability when deployed as a frozen backbone. In practice, bounding-box prompts often contain localization errors, and fixed threshold binarization introduces additional decision uncertainty. These factors jointly cause high prediction variance, especially near object boundaries, degrading reliability. We propose the Stability-Aware Inference Framework (SAIF), a training-free and plug-and-play inference framework that improves robustness by explicitly modeling prompt and threshold uncertainty. SAIF constructs a joint uncertainty space via structured box perturbations and threshold variations, evaluates each hypothesis using decision stability and boundary consistency, and introduces a stability-consistency score to filter unstable candidates and perform stability-weighted fusion in probability space. Experiments on Synapse, CVC-ClinicDB, Kvasir-SEG, and CVC-300 demonstrate that SAIF consistently improves segmentation accuracy and robustness, achieving state-of-the-art performance without retraining or architectural modification. Our anonymous code is released at https://anonymous.4open.science/r/SAIF.

cs.CV

Semantic Class Distribution Learning for Debiasing Semi-Supervised Medical Image Segmentation

Medical image segmentation is critical for computer-aided diagnosis. However, dense pixel-level annotation is time-consuming and costly, and medical datasets often exhibit severe class imbalance. Such an imbalance causes minority structures to be overwhelmed by dominant classes in feature representations, hindering the learning of discriminative features and making reliable segmentation particularly challenging. To address this, we propose the Semantic Class Distribution Learning (SCDL) framework, a plug-and-play module that mitigates supervision and representation biases by learning structured class-conditional feature distributions. SCDL integrates Class Distribution Bidirectional Alignment (CDBA) to align embeddings with learnable class proxies and leverages Semantic Anchor Constraints (SAC) to guide proxies using labeled data. Experiments on the Synapse and AMOS datasets demonstrate that SCDL largely improves segmentation performance across both overall and class-level metrics, with particularly notable gains for several low-frequency organs. Our anonymous code is released at https://anonymous.4open.science/r/SCDL.

cs.CV

A novel perspective on crystal electromagnetic calorimeter design for the CEPC

Crystal electromagnetic calorimeters (ECALs) are essential for high-precision measurements of electrons and photons in particle physics experiments. However, the conventional design, in which long crystal bars point radially toward the interaction region and lack longitudinal segmentation, is incompatible with the three-dimensional shower imaging required by Particle Flow Approach (PFA). We propose a novel perspective on crystal ECAL design to address this limitation. The key innovation is a geometric reconfiguration in which crystal bars are oriented to face the interaction region and arranged orthogonally in adjacent longitudinal layers. This layout achieves fine spatial segmentation of energy deposits by correlating measurements of orthogonal crystal bars. An interleaved structure of regular and inverted trapezoidal modules is incorporated to maximize structural uniformity and detector hermeticity. This design is engineered to preserve the excellent intrinsic energy resolution of crystal ECALs while simultaneously providing the detailed three-dimensional shower imaging essential for PFA. Simulation results confirm the feasibility of achieving excellent energy resolution of $1.14\%/\sqrt{E} \oplus 0.44\%$. Consequently, the proposed design repositions crystal ECAL as a foundational component for PFA-oriented detector systems at facilities such as the Circular Electron Positron Collider (CEPC), offering a new technical pathway to advance the physics goals of future colliders.

physics.ins-det

Conceptual Design of a Novel Highly Granular Crystal Electromagnetic Calorimeter for Future Higgs Factories

Next-generation high-energy electron-positron colliders, operating as Higgs factories, require an unprecedented jet energy resolution for precision measurements of Higgs and Z/W bosons. To address this challenge, a conceptual design is presented for a novel high-granularity crystal electromagnetic calorimeter that combines the superior intrinsic energy resolution of a homogeneous calorimeter with the fine segmentation required for particle-flow reconstruction. The crystal electromagnetic calorimeter design is based on orthogonally arranged long scintillating crystal bars read out by silicon photomultipliers (SiPMs) at both ends. Key design specifications were established through comprehensive simulation studies. Critical technical considerations, including crystal choices, photosensors, electronics, mechanical support, and radiation damage, are discussed. A dedicated digitisation framework was developed to realistically model effects from the crystal, SiPMs, and readout electronics. The performance of a single calorimeter module was evaluated using simulated electron showers. Simulation results for a single module demonstrate an excellent electromagnetic energy resolution of $1.12\%/\sqrt{E(\mathrm{GeV})}\oplus0.22\%$ and an energy linearity within $\pm0.5\%$ for electrons from 3 GeV to 100 GeV. The performance significantly exceeds the design requirement of $\leq 3\%/\sqrt{E(\mathrm{GeV})}\oplus1\%$. The results establish the feasibility of the proposed high-granularity crystal calorimeter concept and point to a promising pathway toward the precision calorimetry required for future high-energy electron-positron collider experiments.

physics.ins-det

EMINDS: Understanding User Behavior Progression for Mental Health Exploration on Social Media

Mental health is an urgent societal issue, and social scientists are increasingly turning to online mental health communities (OMHCs) to analyze user behavior data for early intervention. However, existing sequence mining techniques fall short of the urgent need to explore the behavior progression of different groups (e.g., recovery or deterioration groups) and track the potential long-term impact of behaviors on mental health status. To address this issue, we introduce EMINDS, a visual analytics system built on a novel automatic mining pipeline that extracts distinct behavior stages and assesses the potential impact of frequent stage patterns on mental health status over time. The system includes a set of interactive visualizations that summarize the meaning of each behavior stage and the evolution of different stage patterns. We feature a pattern-centric Sankey diagram to reveal contextual information about the impact of stage patterns on mental health, helping experts understand the specific changes in sequences before and after a stage pattern. We evaluated the effectiveness and usability of EMINDS through two case studies and expert interviews, which examined the potential stage patterns impacting long-term mental health by analyzing user behaviors on Reddit.

cs.HC

Interaction-Augmented Instruction: Modeling the Synergy of Prompts and Interactions in Human-GenAI Collaboration

Text prompt is the most common way for human-generative AI (GenAI) communication. Though convenient, it is challenging to convey fine-grained and referential intent. One promising solution is to combine text prompts with precise GUI interactions, like brushing and clicking. However, there lacks a formal model to capture synergistic designs between prompts and interactions, hindering their comparison and innovation. To fill this gap, via an iterative and deductive process, we develop the Interaction-Augmented Instruction (IAI) model, a compact entity-relation graph formalizing how the combination of interactions and text prompts enhances human-GenAI communication. With the model, we distill twelve recurring and composable atomic interaction paradigms from prior tools, verifying our model's capability to facilitate systematic design characterization and comparison. Four usage scenarios further demonstrate the model's utility in applying, refining, and innovating these paradigms. These results illustrate the IAI model's descriptive, discriminative, and generative power for shaping future GenAI systems.

cs.HC

First experimental proof of PET imaging based on multi-anode MCP-PMTs with Cherenkov radiator-integrated window

Improving the coincidence time resolution (CTR) of time-of-flight positron emission tomography (TOF-PET) systems to achieve a higher signal-to-noise ratio (SNR) gain or even direct positron emission imaging (dPEI) is of paramount importance for many advanced new clinical applications of PET imaging. This places higher demands on the timing performance of all aspects of PET systems. One effective approach is to use microchannel plate photomultiplier tubes (MCP-PMTs) for prompt Cherenkov photon detection. In this study, we developed a dual-module Cherenkov PET imaging experimental platform, utilising our proprietary 8 * 8-anode Cherenkov radiator-integrated window MCP-PMTs in combination with custom-designed multi-channel electronics, and designed a specific calibration and correction method for the platform. Using this platform, a CTR of 103 ps FWHM was achieved. We overcame the limitations of single-anode detectors in previous experiments, significantly enhanced imaging efficiency and achieved module-level Cherenkov PET imaging for the first time. Imaging experiments involving radioactive sources and phantoms of various shapes and types were conducted, which preliminarily validated the feasibility and advancement of this imaging method. In addition, the effects of normalisation correction and the interaction probability between the gamma rays and the MCP on the images and experimental results were analysed and verified.

physics.ins-det

dN/dx Reconstruction with Deep Learning for High-Granularity TPCs

Particle identification (PID) is essential for future particle physics experiments such as the Circular Electron-Positron Collider and the Future Circular Collider. A high-granularity Time Projection Chamber (TPC) not only provides precise tracking but also enables dN/dx measurements for PID. The dN/dx method estimates the number of primary ionization electrons, offering significant improvements in PID performance. However, accurate reconstruction remains a major challenge for this approach. In this paper, we introduce a deep learning model, the Graph Point Transformer (GraphPT), for dN/dx reconstruction. In our approach, TPC data are represented as point clouds. The network backbone adopts a U-Net architecture built upon graph neural networks, incorporating an attention mechanism for node aggregation specifically optimized for point cloud processing. The proposed GraphPT model surpasses the traditional truncated mean method in PID performance. In particular, the $K/\pi$ separation power improves by approximately 10% to 20% in the momentum interval from 5 to 20 GeV/c.

hep-ex

Funding the Frontier: Visualizing the Broad Impact of Science and Science Funding

Understanding the broad impact of science and science funding is critical to ensuring that science investments and policies align with societal needs. Existing research links science funding to the output of scientific publications but largely leaves out the downstream uses of science and the myriad ways in which investing in science may impact human society. As funders seek to allocate scarce funding resources across a complex research landscape, there is an urgent need for informative and transparent tools that allow for comprehensive assessments and visualization of the impact of funding. Here we present Funding the Frontier (FtF), a visual analysis system for researchers, funders, policymakers, university leaders, and the broad public to analyze multidimensional impacts of funding and make informed decisions regarding research investments and opportunities. The system is built on a massive data collection that connects 7M research grants to 140M scientific publications, 160M patents, 10.9M policy documents, 800K clinical trials, and 5.8M newsfeeds, with 1.8B citation linkages among these entities, systematically linking science funding to its downstream impacts. As such, Funding the Frontier is distinguished by its multifaceted impact analysis framework. The system incorporates diverse impact metrics and predictive models that forecast future investment opportunities into an array of coordinated views, allowing for easy exploration of funding and its outcomes. We evaluate the effectiveness and usability of the system using case studies and expert interviews. Feedback suggests that our system not only fulfills the primary analysis needs of its target users, but the rich datasets of the complex science ecosystem and the proposed analysis framework also open new avenues for both visualization and the science of science research.

cs.HC

Towards a detection of reactor $\overlineν^{}_e \to \overlineν^{}_μ$ and $\overlineν^{}_e \to \overlineν^{}_τ$ oscillations with possible CP violation

We propose an unprecedented detection of reactor $\overlineν^{}_e \to \overlineν^{}_μ$ and $\overlineν^{}_e \to \overlineν^{}_τ$ oscillations by using elastic antineutrino-electron scattering processes $\overlineν^{}_α+ e^- \to \overlineν^{}_α+ e^-$ (for $α= e, μ, τ$), among which the $\overlineν^{}_e$ events can be singled out by accurately measuring the $\overlineν^{}_e$ flux via the inverse beta decay $\overlineν^{}_e + p \to e^+ + n$. A proof-of-concept study shows that such measurements will not only be able to test the conservation of probability for reactor antineutrino oscillations, but also offer a new possibility to probe leptonic CP violation at the one-loop level.

hep-ph

Photon reconstruction using the Hough transform in imaging calorimeters

Photon reconstruction in calorimeters represents a crucial challenge in particle physics experiments, especially in high-density environments where shower overlapping probabilities become significant. We present an energy-core-based photon reconstruction method. It is achieved through extending the application of the Hough transform to exploit the energy-core structure of photon showers. The method, validated through simulations of the CEPC crystal electromagnetic calorimeter, achieves a reconstruction efficiency of nearly 100% for photons with energies exceeding 2 GeV and a separation efficiency approaching 100% for two 5 GeV photons, when the distance between them reaches the granularity limit of the calorimeter. This energy-core-based photon reconstruction method, integrated with an energy splitting technique, enhances the performance of photon measurement and provides a promising tool for imaging calorimeters, particularly those requiring high precision in photon detection in complex event topologies with high multiplicity.

physics.ins-det

Fluorescence time profile measurement of LAB based liquid scintillator in response to medium relativistic ion particles

Liquid scintillator is widely used in particle physics experiments due to its high light yield, good timing resolution, scalability and low cost. Certain liquid scintillators exhibit pulse shape discrimination capabilities because of difference in fluorescence timing properties induced by different particles. Its fluoresence timing properties have been measured mostly for radioactive decay sources at MeV energies. We present a novel measurement of fluorescence time properties of LAB based liquid scintillator in response to high-energy ions of hydrogen (Z = 1), helium (Z = 2) and Krypton at around 200-300 MeV/u for the first time. We compared the results to those from radioactive sources and observed a distinct $dE/dX$ dependence, regardless of the particle type. These findings are essential for physics searches such as the diffuse supernova neutrino background in large liquid scintillator detectors like JUNO, and are also critical towards understanding the underlying scintillation timing mechanism.

physics.ins-det

Robust Multimodal Learning for Ophthalmic Disease Grading via Disentangled Representation

This paper discusses how ophthalmologists often rely on multimodal data to improve diagnostic accuracy. However, complete multimodal data is rare in real-world applications due to a lack of medical equipment and concerns about data privacy. Traditional deep learning methods typically address these issues by learning representations in latent space. However, the paper highlights two key limitations of these approaches: (i) Task-irrelevant redundant information (e.g., numerous slices) in complex modalities leads to significant redundancy in latent space representations. (ii) Overlapping multimodal representations make it difficult to extract unique features for each modality. To overcome these challenges, the authors propose the Essence-Point and Disentangle Representation Learning (EDRL) strategy, which integrates a self-distillation mechanism into an end-to-end framework to enhance feature selection and disentanglement for more robust multimodal learning. Specifically, the Essence-Point Representation Learning module selects discriminative features that improve disease grading performance. The Disentangled Representation Learning module separates multimodal data into modality-common and modality-unique representations, reducing feature entanglement and enhancing both robustness and interpretability in ophthalmic disease diagnosis. Experiments on multimodal ophthalmology datasets show that the proposed EDRL strategy significantly outperforms current state-of-the-art methods.

cs.CV