arXiv subjects

Bo Li

Publications and source records attributed to Bo Li.

At least 271 records · Page 15Linked to original sources

arXiv2025

Cooperative Energy Scheduling of Multi-Microgrids Based on Risk-Sensitive Reinforcement Learning

With the rapid development of distributed renewable energy, multi-microgrids play an increasingly important role in improving the flexibility and reliability of energy supply. Reinforcement learning has shown great potential in coordination strategies due to its model-free nature. Current methods lack explicit quantification of the relationship between individual and joint risk values, resulting in obscured credit assignment. Moreover, they often depend on explicit communication, which becomes inefficient as system complexity grows. To address these challenges, this paper proposes a risk-sensitive reinforcement learning framework with shared memory (RRL-SM) for multi-microgrid scheduling. Specifically, a risk-sensitive value factorization scheme is proposed to quantify the relationship between individual and joint risk values by leveraging distributional modeling and attention-based representations, thereby aligning local decisions with global risk objectives. An implicit shared-memory coordination mechanism is implemented through a global memory space to enhance the overall efficiency of decentralized decision-making. Collectively, the integrated approach delivers more reliable cooperative scheduling under renewable energy uncertainty. Simulation results show that RRL-SM reduces load-shedding risk by 84.5%, demonstrating a favorable balance between reliability and economic performance.

eess.SY↗

arXiv2025

Quantifying Functional Criticality of Lifelines Through Mobility-Derived Population-Facility Dependence for Human-Centered Resilience

Lifeline infrastructure underpins the continuity of daily life, yet conventional criticality assessments remain largely asset-centric, inferring importance from physical capacity or network topology rather than actual behavioral reliance. This disconnect frequently obscures the true societal cost of disruption, particularly in underserved communities where residents lack service alternatives. This study bridges the gap between engineering risk analysis and human mobility analysis by introducing functional criticality, a human-centered metric that quantifies the behavioral indispensability of specific facilities based on large-scale visitation patterns. Leveraging 1.02 million anonymized mobility records for Harris County, Texas, we operationalize lifeline criticality as a function of visitation intensity, catchment breadth, and origin-specific substitutability. Results reveal that dependence is highly concentrated: a small subset of super-critical facilities (2.8% of grocery stores and 14.8% of hospitals) supports a disproportionate share of routine access. By coupling these behavioral scores with probabilistic flood hazard models for 2020 and 2060, we demonstrate a pronounced risk-multiplier effect. While physical flood depths increase only moderately under future climate scenarios, the population-weighted vulnerability of access networks surges by 67.6%. This sharp divergence establishes that future hazards will disproportionately intersect with the specific nodes communities rely on most. The proposed framework advances resilience assessment by embedding human behavior directly into the definition of infrastructure importance, providing a scalable foundation for equitable, adaptive, and human-centered resilience planning.

cs.CY↗

arXiv2025

LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training

We present LLaVA-OneVision-1.5, a novel family of Large Multimodal Models (LMMs) that achieve state-of-the-art performance with significantly reduced computational and financial costs. Different from the existing works, LLaVA-OneVision-1.5 provides an open, efficient, and reproducible framework for building high-quality vision-language models entirely from scratch. The LLaVA-OneVision-1.5 release comprises three primary components: (1) Large-Scale Curated Datasets: We construct an 85M concept-balanced pretraining dataset LLaVA-OneVision-1.5-Mid-Traning and a meticulously curated 22M instruction dataset LLaVA-OneVision-1.5-Instruct. (2) Efficient Training Framework: We develop a complete end-to-end efficient training framework leveraging an offline parallel data packing strategy to facilitate the training of LLaVA-OneVision-1.5 within a $16,000 budget. (3) State-of-the-art Performance: Experimental results demonstrate that LLaVA-OneVision-1.5 yields exceptionally competitive performance across a broad range of downstream tasks. Specifically, LLaVA-OneVision-1.5-8B outperforms Qwen2.5-VL-7B on 18 of 27 benchmarks, and LLaVA-OneVision-1.5-4B surpasses Qwen2.5-VL-3B on all 27 benchmarks. (4) RL-based Post-training: We unlock the model's latent potential through a lightweight RL stage, effectively eliciting robust chain-of-thought reasoning to significantly boost performance on complex multimodal reasoning tasks.

cs.CV↗

arXiv2025

A Physics-Informed Neural Network Framework for Simulating Creep Buckling in Growing Viscoelastic Biological Tissues

Modeling viscoelastic behavior is crucial in engineering and biomechanics, where materials undergo time-dependent deformations, including stress relaxation, creep buckling and biological tissue development. Traditional numerical methods, like the finite element method, often require explicit meshing, artificial perturbations or embedding customised programs to capture these phenomena, adding computational complexity. In this study, we develop an energy-based physics-informed neural network (PINN) framework using an incremental approach to model viscoelastic creep, stress relaxation, buckling, and growth-induced morphogenesis. Physics consistency is ensured by training neural networks to minimize the systems potential energy functional, implicitly satisfying equilibrium and constitutive laws. We demonstrate that this framework can naturally capture creep buckling without pre-imposed imperfections, leveraging inherent training dynamics to trigger instabilities. Furthermore, we extend our framework to biological tissue growth and morphogenesis, predicting both uniform expansion and differential growth-induced buckling in cylindrical structures. Results show that the energy-based PINN effectively predicts viscoelastic instabilities, post-buckling evolution and tissue morphological evolution, offering a promising alternative to traditional methods. This study demonstrates that PINN can be a flexible robust tool for modeling complex, time-dependent material behavior, opening possible applications in structural engineering, soft materials, and tissue development.

cs.CE↗

arXiv2025

Image2Net: Datasets, Benchmark and Hybrid Framework to Convert Analog Circuit Diagrams into Netlists

Large Language Model (LLM) exhibits great potential in designing of analog integrated circuits (IC) because of its excellence in abstraction and generalization for knowledge. However, further development of LLM-based analog ICs heavily relies on textual description of analog ICs, while existing analog ICs are mostly illustrated in image-based circuit diagrams rather than text-based netlists. Converting circuit diagrams to netlists help LLMs to enrich the knowledge of analog IC. Nevertheless, previously proposed conversion frameworks face challenges in further application because of limited support of image styles and circuit elements. Up to now, it still remains a challenging task to effectively convert complex circuit diagrams into netlists. To this end, this paper constructs and opensources a new dataset with rich styles of circuit diagrams as well as balanced distribution of simple and complex analog ICs. And a hybrid framework, named Image2Net, is proposed for practical conversion from circuit diagrams to netlists. The netlist edit distance (NED) is also introduced to precisely assess the difference between the converted netlists and ground truth. Based on our benchmark, Image2Net achieves 80.77% successful rate, which is 34.62%-45.19% higher than previous works. Specifically, the proposed work shows 0.116 averaged NED, which is 62.1%-69.6% lower than state-of-the-arts. Our datasets and benchmark are available at https://github.com/LAD021/ci2n_datasets.

cs.AR↗

arXiv2025

VulnLLM-R: Specialized Reasoning LLM with Agent Scaffold for Vulnerability Detection

We propose VulnLLM-R, the~\emph{first specialized reasoning LLM} for vulnerability detection. Our key insight is that LLMs can reason about program states and analyze the potential vulnerabilities, rather than simple pattern matching. This can improve the model's generalizability and prevent learning shortcuts. However, SOTA reasoning LLMs are typically ultra-large, closed-source, or have limited performance in vulnerability detection. To address this, we propose a novel training recipe with specialized data selection, reasoning data generation, reasoning data filtering and correction, and testing-phase optimization. Using our proposed methodology, we train a reasoning model with seven billion parameters. Through extensive experiments on SOTA datasets across Python, C/C++, and Java, we show that VulnLLM-R has superior effectiveness and efficiency than SOTA static analysis tools and both open-source and commercial large reasoning models. We further conduct a detailed ablation study to validate the key designs in our training recipe. Finally, we construct an agent scaffold around our model and show that it outperforms CodeQL and AFL++ in real-world projects. Our agent further discovers a set of zero-day vulnerabilities in actively maintained repositories. This work represents a pioneering effort to enable real-world, project-level vulnerability detection using AI agents powered by specialized reasoning models. The code is available at~\href{https://github.com/ucsb-mlsec/VulnLLM-R}{github}.

cs.CR↗

arXiv2025

OpenMMReasoner: Pushing the Frontiers for Multimodal Reasoning with an Open and General Recipe

Recent advancements in large reasoning models have fueled growing interest in extending such capabilities to multimodal domains. However, despite notable progress in visual reasoning, the lack of transparent and reproducible data curation and training strategies remains a major barrier to scalable research. In this work, we introduce OpenMMReasoner, a fully transparent two-stage recipe for multimodal reasoning spanning supervised fine-tuning (SFT) and reinforcement learning (RL). In the SFT stage, we construct an 874K-sample cold-start dataset with rigorous step-by-step validation, providing a strong foundation for reasoning capabilities. The subsequent RL stage leverages a 74K-sample dataset across diverse domains to further sharpen and stabilize these abilities, resulting in a more robust and efficient learning process. Extensive evaluations demonstrate that our training recipe not only surpasses strong baselines but also highlights the critical role of data quality and training design in shaping multimodal reasoning performance. Notably, our method achieves a 11.6% improvement over the Qwen2.5-VL-7B-Instruct baseline across nine multimodal reasoning benchmarks, establishing a solid empirical foundation for future large-scale multimodal reasoning research. We open-sourced all our codes, pipeline, and data at https://github.com/EvolvingLMMs-Lab/OpenMMReasoner.

cs.AI↗

arXiv2025

Rethinking the Learning Paradigm for Facial Expression Recognition

Due to the subjective crowdsourcing annotations and the inherent inter-class similarity of facial expressions, the real-world Facial Expression Recognition (FER) datasets usually exhibit ambiguous annotation. To simplify the learning paradigm, most previous methods convert ambiguous annotation results into precise one-hot annotations and train FER models in an end-to-end supervised manner. In this paper, we rethink the existing training paradigm and propose that it is better to use weakly supervised strategies to train FER models with original ambiguous annotation.

cs.CV↗

arXiv2025

A Definition of AGI

The lack of a concrete definition for Artificial General Intelligence (AGI) obscures the gap between today's specialized AI and human-level cognition. This paper introduces a quantifiable framework to address this, defining AGI as matching the cognitive versatility and proficiency of a well-educated adult. To operationalize this, we ground our methodology in Cattell-Horn-Carroll theory, the most empirically validated model of human cognition. The framework dissects general intelligence into ten core cognitive domains-including reasoning, memory, and perception-and adapts established human psychometric batteries to evaluate AI systems. Application of this framework reveals a highly "jagged" cognitive profile in contemporary models. While proficient in knowledge-intensive domains, current AI systems have critical deficits in foundational cognitive machinery, particularly long-term memory storage. The resulting AGI scores (e.g., GPT-4 at 27%, GPT-5 at 57%) concretely quantify both rapid progress and the substantial gap remaining before AGI.

cs.AI↗

arXiv2025

Topological Pseudospin Hall Effect and Multi-frequency Corner Modes in Kagome-based Lattices

Topological phases and modes, including pseudospin-Hall-selective edge transport and corner states, provide robust control of wave propagation and modal confinement in classical wave platforms. Under a tight-binding framework, we theoretically investigate two lattice designs derived from the kagome lattice. These extended kagome lattices support a series of localized modes, including pseudospin-Hall-like topological edge states and corner modes in different bandgaps and frequencies, which were not only achieved under lower lattice symmetries than Wu-Hu lattices, but also enable more degrees of freedom in topological and localized modes. By introducing two types of extended kagome lattices with different topological properties, multiple interesting phenomena, including newly emerged multiple groups of corner states, parametric tunable pseudospin Hall effect, and type-II corner states without long-range interactions, are found in theoretical models, which are possible and viable to achieve in artificial classical systems such as photonic, acoustic, or electrical circuits.

cond-mat.mes-hall↗

arXiv2025

Mitigating Indirect Prompt Injection via Instruction-Following Intent Analysis

Indirect prompt injection attacks (IPIAs), where large language models (LLMs) follow malicious instructions hidden in input data, pose a critical threat to LLM-powered agents. In this paper, we present IntentGuard, a general defense framework based on instruction-following intent analysis. The key insight of IntentGuard is that the decisive factor in IPIAs is not the presence of malicious text, but whether the LLM intends to follow instructions from untrusted data. Building on this insight, IntentGuard leverages an instruction-following intent analyzer (IIA) to identify which parts of the input prompt the model recognizes as actionable instructions, and then flag or neutralize any overlaps with untrusted data segments. To instantiate the framework, we develop an IIA that uses three "thinking intervention" strategies to elicit a structured list of intended instructions from reasoning-enabled LLMs. These techniques include start-of-thinking prefilling, end-of-thinking refinement, and adversarial in-context demonstration. We evaluate IntentGuard on two agentic benchmarks (AgentDojo and Mind2Web) using two reasoning-enabled LLMs (Qwen-3-32B and gpt-oss-20B). Results demonstrate that IntentGuard achieves (1) no utility degradation in all but one setting and (2) strong robustness against adaptive prompt injection attacks (e.g., reducing attack success rates from 100% to 8.5% in a Mind2Web scenario).

cs.CR↗

arXiv2025

VQRAE: Representation Quantization Autoencoders for Multimodal Understanding, Generation and Reconstruction

Unifying multimodal understanding, generation and reconstruction representation in a single tokenizer remains a key challenge in building unified models. Previous research predominantly attempts to address this in a dual encoder paradigm, e.g., utilizing the separate encoders for understanding and generation respectively or balancing semantic representations and low-level features with contrastive loss. In this paper, we propose VQRAE, a Vector Quantization version of Representation AutoEncoders, which pioneers the first exploration in unified representation to produce Continuous semantic features for image understanding and Discrete tokens for visual generation within a unified tokenizer. Specifically, we build upon pretrained vision foundation models with a symmetric ViT decoder and adopt a two-stage training strategy: first, it freezes the encoder and learns a high-dimensional semantic VQ codebook with pixel reconstruction objective; then jointly optimizes the encoder with self-distillation constraints. This design enables negligible semantic information for maintaining the ability of multimodal understanding, discrete tokens that are compatible for generation and fine-grained reconstruction. Besides, we identify the intriguing property in quantizing semantic encoders that rely on high-dimensional codebook in contrast to the previous common practice of low-dimensional codebook in image reconstruction. The semantic VQ codebook can achieve a 100% utilization ratio at a dimension of 1536. VQRAE presents competitive performance on several benchmarks of visual understanding, generation and reconstruction with promising scaling property in the autoregressive paradigm for its discrete merits.

cs.CV↗

arXiv2025

Visual Generation Tuning

Large Vision Language Models (VLMs) effectively bridge the modality gap through extensive pretraining, acquiring sophisticated visual representations aligned with language. However, it remains underexplored whether these representations, optimized for multimodal understanding tasks, harbor an inherent potential for visual generation. In this paper, we propose VGT, Visual Generation Tuning, a novel paradigm designed to stimulate the underlying capabilities of visual generation within any vision language models. By performing efficient visual generation tuning on well-pretrained VLMs, we significantly mitigate the alignment costs and accelerate the convergence of autoregressive modeling in the continuous space (20x speedup). Specifically, we dismiss the entangled pixel-level VAEs designed for diffusion transformers and formulate VGT-AE through aligning the semantic encoders from pretrained VLMs with the latent representations of pixel decoders. In image reconstruction tasks, we achieve 26.67 PSNR and 0.50 rFID at a 28x compression ratio, outperforming specialized VAEs; in visual generation tasks, we achieve state-of-the-art outcomes among autoregressive models, 0.77 on GenEval and 78.73 on DPG-Bench. Furthermore, our proposed VGT showcases significant scaling promise and is versatile for endowing any VLMs trained for multimodal understanding with the capabilities of visual generation, which paves the new avenue to explore next-generation unified multimodal foundation models. Models and codes are available at https://github.com/hustvl/VGT.

cs.CV↗

arXiv2025

ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoning

Autonomous agents powered by foundation models have seen widespread adoption across various real-world applications. However, they remain highly vulnerable to malicious instructions and attacks, which can result in severe consequences such as privacy breaches and financial losses. More critically, existing guardrails for LLMs are not applicable due to the complex and dynamic nature of agents. To tackle these challenges, we propose ShieldAgent, the first guardrail agent designed to enforce explicit safety policy compliance for the action trajectory of other protected agents through logical reasoning. Specifically, ShieldAgent first constructs a safety policy model by extracting verifiable rules from policy documents and structuring them into a set of action-based probabilistic rule circuits. Given the action trajectory of the protected agent, ShieldAgent retrieves relevant rule circuits and generates a shielding plan, leveraging its comprehensive tool library and executable code for formal verification. In addition, given the lack of guardrail benchmarks for agents, we introduce ShieldAgent-Bench, a dataset with 3K safety-related pairs of agent instructions and action trajectories, collected via SOTA attacks across 6 web environments and 7 risk categories. Experiments show that ShieldAgent achieves SOTA on ShieldAgent-Bench and three existing benchmarks, outperforming prior methods by 11.3% on average with a high recall of 90.1%. Additionally, ShieldAgent reduces API queries by 64.7% and inference time by 58.2%, demonstrating its high precision and efficiency in safeguarding agents.

cs.LG↗

arXiv2025

CDR-Agent: Intelligent Selection and Execution of Clinical Decision Rules Using Large Language Model Agents

Clinical decision-making is inherently complex and fast-paced, particularly in emergency departments (EDs) where critical, rapid and high-stakes decisions are made. Clinical Decision Rules (CDRs) are standardized evidence-based tools that combine signs, symptoms, and clinical variables into decision trees to make consistent and accurate diagnoses. CDR usage is often hindered by the clinician's cognitive load, limiting their ability to quickly recall and apply the appropriate rules. We introduce CDR-Agent, a novel LLM-based system designed to enhance ED decision-making by autonomously identifying and applying the most appropriate CDRs based on unstructured clinical notes. To validate CDR-Agent, we curated two novel ED datasets: synthetic and CDR-Bench, although CDR-Agent is applicable to non ED clinics. CDR-Agent achieves a 56.3\% (synthetic) and 8.7\% (CDR-Bench) accuracy gain relative to the standalone LLM baseline in CDR selection. Moreover, CDR-Agent significantly reduces computational overhead. Using these datasets, we demonstrated that CDR-Agent not only selects relevant CDRs efficiently, but makes cautious yet effective imaging decisions by minimizing unnecessary interventions while successfully identifying most positively diagnosed cases, outperforming traditional LLM prompting approaches. Code for our work can be found at: https://github.com/zhenxianglance/medagent-cdr-agent

cs.LG↗

arXiv2025

An Equality Set Projection Approach for TSO-DSO Coordination Dispatch

Coordinated optimization dispatch (COD) of transmission system operator (TSO) and distribution system operator (DSO) can effectively ensure system security and efficiency under high-penetration distributed energy resource (DER) integration. Researches of large-scale COD problem can be categorized into iterative approaches that allow DSO to dispatch independently, and non-iterative methods based on projections of feasible regions (FR). However, the iterative methods suffer from low computational convergence and efficiency, while non-iterative methods struggle to solve equivalent projections with high-dimensional FR. To address these issues, this paper proposes a TSO-DSO coordinated dispatch approach based on an accelerated non-iterative Equality Set Projection (ESP) algorithm. First, ESP algorithm is employed to overcome the bottleneck of high-dimensional FR construction. Second, an regularization-based accelerated method is proposed to reduce computational burden when degeneracy occurs. Accelerated ESP algorithm constructs projection of FR via adjacent facet searching. Therefore, it is less sensitive to the increase of vertices and could efficiently construct the projection of high-dimensional FR. Case studies on a polyhedron dataset, IEEE 33-Bus System and T118D10 TSO-DSO system demonstrate the effectiveness and computational efficiency of the proposed COD approach.

eess.SY↗

arXiv2025

CameraMaster: Unified Camera Semantic-Parameter Control for Photography Retouching

Text-guided diffusion models have greatly advanced image editing and generation. However, achieving physically consistent image retouching with precise parameter control (e.g., exposure, white balance, zoom) remains challenging. Existing methods either rely solely on ambiguous and entangled text prompts, which hinders precise camera control, or train separate heads/weights for parameter adjustment, which compromises scalability, multi-parameter composition, and sensitivity to subtle variations. To address these limitations, we propose CameraMaster, a unified camera-aware framework for image retouching. The key idea is to explicitly decouple the camera directive and then coherently integrate two critical information streams: a directive representation that captures the photographer's intent, and a parameter embedding that encodes precise camera settings. CameraMaster first uses the camera parameter embedding to modulate both the camera directive and the content semantics. The modulated directive is then injected into the content features via cross-attention, yielding a strongly camera-sensitive semantic context. In addition, the directive and camera embeddings are injected as conditioning and gating signals into the time embedding, enabling unified, layer-wise modulation throughout the denoising process and enforcing tight semantic-parameter alignment. To train and evaluate CameraMaster, we construct a large-scale dataset of 78K image-prompt pairs annotated with camera parameters. Extensive experiments show that CameraMaster produces monotonic and near-linear responses to parameter variations, supports seamless multi-parameter composition, and significantly outperforms existing methods.

cs.CV↗

arXiv2025

Decoupling Urban Food Accessibility Resilience during Disasters through Time-Series Analysis of Human Mobility and Power Outages

Disaster-induced power outages create cascading disruptions across urban lifelines, yet the timed coupling between grid failure and essential service access remains poorly quantified. Focusing on Hurricane Beryl in Houston (2024), this study integrates approximately 173000 15-minute outage records with over 1.25 million visits to 3187 food facilities to quantify how infrastructure performance and human access co-evolve. We construct daily indices for outage characteristics (intensity, duration) and food access metrics (redundancy, frequency, proximity), estimate cross-system lags through lagged correlations over zero to seven days, and identify recovery patterns using DTW k-means clustering. Overlaying these clusters yields compound power-access typologies and enables facility-level criticality screening. The analysis reveals a consistent two-day lag: food access reaches its nadir on July 8 at landfall while outage severity peaks around July 10, with negative correlations strongest at a two-day lag and losing significance by day four. We identify four compound typologies from high/low outage crossed with high/low access disruption levels. Road network sparsity, more than income, determines the depth and persistence of access loss. Through this analysis, we enumerate 294 critical food facilities in the study area requiring targeted continuity measures including backup power, microgrids, and feeder prioritization. The novelty lies in measuring interdependency at daily operational resolution while bridging scales from communities to individual facilities, converting dynamic coupling patterns into actionable interventions for phase-sensitive restoration and equity-aware preparedness. The framework is transferable to other lifelines and hazards, offering a generalizable template for diagnosing and mitigating cascading effects on community access during disaster recovery.

stat.AP↗