arXiv subjects

Bo Zhou

Publications and source records attributed to Bo Zhou.

At least 73 records · Page 4Linked to original sources

arXiv2025

HSCP: A Two-Stage Spectral Clustering Framework for Resource-Constrained UAV Identification

With the rapid development of Unmanned Aerial Vehicles (UAVs) and the increasing complexity of low-altitude security threats, traditional UAV identification methods struggle to extract reliable signal features and meet real-time requirements in complex environments. Recently, deep learning based Radio Frequency Fingerprint Identification (RFFI) approaches have greatly improved recognition accuracy. However, their large model sizes and high computational demands hinder deployment on resource-constrained edge devices. While model pruning offers a general solution for complexity reduction, existing weight, channel, and layer pruning techniques struggle to concurrently optimize compression rate, hardware acceleration, and recognition accuracy. To this end, in this paper, we introduce HSCP, a Hierarchical Spectral Clustering Pruning framework that combines layer pruning with channel pruning to achieve extreme compression, high performance, and efficient inference. In the first stage, HSCP employs spectral clustering guided by Centered Kernel Alignment (CKA) to identify and remove redundant layers. Subsequently, the same strategy is applied to the channel dimension to eliminate a finer redundancy. To ensure robustness, we further employ a noise-robust fine-tuning strategy. Experiments on the UAV-M100 benchmark demonstrate that HSCP outperforms existing channel and layer pruning methods. Specifically, HSCP achieves $86.39\%$ parameter reduction and $84.44\%$ FLOPs reduction on ResNet18 while improving accuracy by $1.49\%$ compared to the unpruned baseline, and maintains superior robustness even in low signal-to-noise ratio environments.

cs.CV↗

arXiv2025

HiFusion: Hierarchical Intra-Spot Alignment and Regional Context Fusion for Spatial Gene Expression Prediction from Histopathology

Spatial transcriptomics (ST) bridges gene expression and tissue morphology but faces clinical adoption barriers due to technical complexity and prohibitive costs. While computational methods predict gene expression from H&E-stained whole-slide images (WSIs), existing approaches often fail to capture the intricate biological heterogeneity within spots and are susceptible to morphological noise when integrating contextual information from surrounding tissue. To overcome these limitations, we propose HiFusion, a novel deep learning framework that integrates two complementary components. First, we introduce the Hierarchical Intra-Spot Modeling module that extracts fine-grained morphological representations through multi-resolution sub-patch decomposition, guided by a feature alignment loss to ensure semantic consistency across scales. Concurrently, we present the Context-aware Cross-scale Fusion module, which employs cross-attention to selectively incorporate biologically relevant regional context, thereby enhancing representational capacity. This architecture enables comprehensive modeling of both cellular-level features and tissue microenvironmental cues, which are essential for accurate gene expression prediction. Extensive experiments on two benchmark ST datasets demonstrate that HiFusion achieves state-of-the-art performance across both 2D slide-wise cross-validation and more challenging 3D sample-specific scenarios. These results underscore HiFusion's potential as a robust, accurate, and scalable solution for ST inference from routine histopathology.

cs.CV↗

arXiv2025

SynLlama: Generating Synthesizable Molecules and Their Analogs with Large Language Models

Generative machine learning models for exploring chemical space have shown immense promise, but many molecules they generate are too difficult to synthesize, making them impractical for further investigation or development. In this work, we present a novel approach by fine-tuning Meta's Llama3 Large Language Models (LLMs) to create SynLlama, which generates full synthetic pathways made of commonly accessible building blocks and robust organic reaction templates. SynLlama explores a large synthesizable space using significantly less data, and offers strong performance in both forward and bottom-up synthesis planning compared to other state-of-the-art methods. We find that SynLlama, even without training on external building blocks, can effectively generalize to unseen yet purchasable building blocks, meaning that its reconstruction capabilities extend to a broader synthesizable chemical space than the training data. We also demonstrate the use of SynLlama in a pharmaceutical context for synthesis planning of analog molecules and hit expansion leads for proposed inhibitors of target proteins, offering medicinal chemists a valuable tool for discovery.

cs.LG↗

arXiv2025

Low-probability Tokens Sustain Exploration in Reinforcement Learning with Verifiable Reward

Reinforcement Learning with Verifiable Rewards (RLVR) has propelled Large Language Models in complex reasoning, yet its scalability is often hindered by a training bottleneck where performance plateaus as policy entropy collapses, signaling a loss of exploration. Previous methods typically address this by maintaining high policy entropy, yet the precise mechanisms that govern meaningful exploration have remained underexplored. Our analysis suggests that an unselective focus on entropy risks amplifying irrelevant tokens and destabilizing training. This paper investigates the exploration dynamics within RLVR and identifies a key issue: the gradual elimination of valuable low-probability exploratory tokens, which we term \textbf{\textit{reasoning sparks}}. We find that while abundant in pre-trained models, these sparks are systematically extinguished during RLVR due to over-penalization, leading to a degeneracy in exploration. To address this, we introduce Low-probability Regularization (Lp-Reg). Its core mechanism regularizes the policy towards a heuristic proxy distribution. This proxy is constructed by filtering out presumed noise tokens and re-normalizing the distribution over the remaining candidates. The result is a less-noisy proxy where the probability of \textit{reasoning sparks} is amplified, which then serves as a soft regularization target to shield these valuable tokens from elimination via KL divergence. Experiments show that Lp-Reg enables stable on-policy RL, sustaining continuous scaling across $3,000$ training steps and $81,204$ GPU-hours, where baseline entropy-control methods collapse. This sustained exploration leads to state-of-the-art performance, achieving a $60.17\%$ average accuracy on five math benchmarks, an improvement of $2.66\%$ over prior methods. Code is available at https://github.com/CarlanLark/Lp-Reg.

cs.LG↗

arXiv2025

Game-TARS: Pretrained Foundation Models for Scalable Generalist Multimodal Game Agents

We present Game-TARS, a generalist game agent trained with a unified, scalable action space anchored to human-aligned native keyboard-mouse inputs. Unlike API- or GUI-based approaches, this paradigm enables large-scale continual pre-training across heterogeneous domains, including OS, web, and simulation games. Game-TARS is pre-trained on over 500B tokens with diverse trajectories and multimodal data. Key techniques include a decaying continual loss to reduce causal confusion and an efficient Sparse-Thinking strategy that balances reasoning depth and inference cost. Experiments show that Game-TARS achieves about 2 times the success rate over the previous sota model on open-world Minecraft tasks, is close to the generality of fresh humans in unseen web 3d games, and outperforms GPT-5, Gemini-2.5-Pro, and Claude-4-Sonnet in FPS benchmarks. Scaling results on training-time and test-time confirm that the unified action space sustains improvements when scaled to cross-game and multimodal data. Our results demonstrate that simple, scalable action representations combined with large-scale pre-training provide a promising path toward generalist agents with broad computer-use abilities.

cs.AI↗

arXiv2025

The universal size compression effect of nucleon pair in finite nuclei

We systematically investigate the size evolution of the di-neutron (2n), di-proton (2p), and deuteron (d) in 6He, 14Be, 17B, 6Be, 17Ne, and 6Li using microscopic calculations. Remarkably, all nucleon pairs exhibit a universal size compression at the nuclear surface, regardless of their species and binding energies. These features correspond to the BCS- and BEC-like nucleon pairs, which recent experimental techniques can further investigate.

nucl-th↗

arXiv2025

ReLook: Vision-Grounded RL with a Multimodal LLM Critic for Agentic Web Coding

While Large Language Models (LLMs) excel at algorithmic code generation, they struggle with front-end development, where correctness is judged on rendered pixels and interaction. We present ReLook, an agentic, vision-grounded reinforcement learning framework that empowers an agent to close a robust generate--diagnose--refine loop by invoking a multimodal LLM (MLLM) as a tool. During training, the agent uses the MLLM-in-the-loop both as a visual critic--scoring code with screenshots--and as a source of actionable, vision-grounded feedback; a strict zero-reward rule for invalid renders anchors renderability and prevents reward hacking. To prevent behavioral collapse, we introduce Forced Optimization, a strict acceptance rule that admits only improving revisions, yielding monotonically better trajectories. At inference, we decouple the critic and run a lightweight, critic-free self-edit cycle, keeping latency comparable to base decoding while retaining most of the gains. Across three widely used benchmarks, ReLook consistently outperforms strong baselines in vision-grounded front-end code generation, highlighting the benefits of agentic perception, visual rewards, and training-inference decoupling.

cs.LG↗

arXiv2025

Reinforcement Learning on Pre-Training Data

The growing disparity between the exponential scaling of computational resources and the finite growth of high-quality text data now constrains conventional scaling approaches for large language models (LLMs). To address this challenge, we introduce Reinforcement Learning on Pre-Training data (RLPT), a new training-time scaling paradigm for optimizing LLMs. In contrast to prior approaches that scale training primarily through supervised learning, RLPT enables the policy to autonomously explore meaningful trajectories to learn from pre-training data and improve its capability through reinforcement learning (RL). While existing RL strategies such as reinforcement learning from human feedback (RLHF) and reinforcement learning with verifiable rewards (RLVR) rely on human annotation for reward construction, RLPT eliminates this dependency by deriving reward signals directly from pre-training data. Specifically, it adopts a next-segment reasoning objective, rewarding the policy for accurately predicting subsequent text segments conditioned on the preceding context. This formulation allows RL to be scaled on pre-training data, encouraging the exploration of richer trajectories across broader contexts and thereby fostering more generalizable reasoning skills. Extensive experiments on both general-domain and mathematical reasoning benchmarks across multiple models validate the effectiveness of RLPT. For example, when applied to Qwen3-4B-Base, RLPT yields absolute improvements of $3.0$, $5.1$, $8.1$, $6.0$, $6.6$, and $5.3$ on MMLU, MMLU-Pro, GPQA-Diamond, KOR-Bench, AIME24, and AIME25, respectively. The results further demonstrate favorable scaling behavior, suggesting strong potential for continued gains with more compute. In addition, RLPT provides a solid foundation, extending the reasoning boundaries of LLMs and enhancing RLVR performance.

cs.CL↗

arXiv2025

Learning from Gene Names, Expression Values and Images: Contrastive Masked Text-Image Pretraining for Spatial Transcriptomics Representation Learning

Spatial transcriptomics aims to connect high-resolution histology images with spatially resolved gene expression. To achieve better performance on downstream tasks such as gene expression prediction, large-scale pre-training is required to obtain generalisable representations that can bridge histology and transcriptomics across tissues, protocols, and laboratories. Existing cross-modal pre-training approaches for spatial transcriptomics rely on either gene names or expression values in isolation, which strips the gene branch of essential semantics and breaks the association between each gene and its quantitative magnitude. In addition, by restricting supervision to image-text alignment, these methods ignore intrinsic visual cues that are critical for learning robust image features. We present CoMTIP, the first Contrastive Masked Text-Image Pretraining framework that jointly learns from images, gene names, and expression values while capturing fine-grained visual context for spatial transcriptomics. The vision branch uses Masked Feature Modeling to reconstruct occluded patches and learn context-aware image embeddings. The text branch applies a scalable Gene-Text Encoder that processes all gene sentences in parallel, enriches each gene and its numerical value with dedicated embeddings, and employs Pair-aware Adversarial Training (PAAT) to preserve correct gene-value associations. Image and text representations are aligned in a shared InfoNCE-optimised space. Experiments on public spatial transcriptomics datasets show that CoMTIP not only surpasses previous methods on diverse downstream tasks but also achieves zero-shot gene expression prediction, a capability that existing approaches do not provide.

cs.CV↗

arXiv2025

Global Spin Alignment of (Anti-)$^4$Li in Non-Central Heavy-Ion Collisions

Non-central heavy-ion collisions produce hot and dense nuclear matter with significant fluid vorticity, which can induce global polarizations or alignments of particles with non-zero spins along the direction of the total orbital angular momentum. This phenomenon has been observed for hyperons and vector mesons in experiments. In the present study, we demonstrate that polarized nucleons lead to global spin alignment of the unstable nucleus $^4$Li, which can be measured through its strong decays via $^4\text{Li} \rightarrow {^3\text{He}} + p$. Assuming that $^4$Li is formed through the coalescence of polarized nucleons at kinetic freeze-out, we obtain the angular distribution of the daughter particle $^3$He in the rest frame of the polarized $^4$Li. Taking kinetically freeze-out nucleons from an isotropic and thermalized fireball of constant vorticity and including quantum corrections up to $\hbar^2$ in the coalescence calculation through the Moyal star product, we find that the angular distribution of $^3$He has a $\cos(2θ^*)$ dependence with $θ^*$ being its angle with respect to the quantization axis of $^4$Li. We also find that the $^3$He angular distribution depends on both the vorticity and the polarization of kinetically freeze-out nucleons. Future measurements on the spin alignment of $^4$Li in heavy-ion collisions thus offer a promising method to probe the spin dynamics, vortical structure, and spin-dependent equation-of-state of the nuclear matter produced in these collisions.

nucl-th↗

arXiv2025

Bridging Generalization and Personalization in Human Activity Recognition via On-Device Few-Shot Learning

Human Activity Recognition (HAR) with different sensing modalities requires both strong generalization across diverse users and efficient personalization for individuals. However, conventional HAR models often fail to generalize when faced with user-specific variations, leading to degraded performance. To address this challenge, we propose a novel on-device few-shot learning framework that bridges generalization and personalization in HAR. Our method first trains a generalizable representation across users and then rapidly adapts to new users with only a few labeled samples, updating lightweight classifier layers directly on resource-constrained devices. This approach achieves robust on-device learning with minimal computation and memory cost, making it practical for real-world deployment. We implement our framework on the energy-efficient RISC-V GAP9 microcontroller and evaluate it on three benchmark datasets (RecGym, QVAR-Gesture, Ultrasound-Gesture). Across these scenarios, post-deployment adaptation improves accuracy by 3.73\%, 17.38\%, and 3.70\%, respectively. These results demonstrate that few-shot on-device learning enables scalable, user-aware, and energy-efficient wearable human activity recognition by seamlessly uniting generalization and personalization. The related framework is open sourced for further research\footnote{https://github.com/kangpx/onlineTiny2023}.

cs.LG↗

arXiv2025

UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning

The development of autonomous agents for graphical user interfaces (GUIs) presents major challenges in artificial intelligence. While recent advances in native agent models have shown promise by unifying perception, reasoning, action, and memory through end-to-end learning, open problems remain in data scalability, multi-turn reinforcement learning (RL), the limitations of GUI-only operation, and environment stability. In this technical report, we present UI-TARS-2, a native GUI-centered agent model that addresses these challenges through a systematic training methodology: a data flywheel for scalable data generation, a stabilized multi-turn RL framework, a hybrid GUI environment that integrates file systems and terminals, and a unified sandbox platform for large-scale rollouts. Empirical evaluation demonstrates that UI-TARS-2 achieves significant improvements over its predecessor UI-TARS-1.5. On GUI benchmarks, it reaches 88.2 on Online-Mind2Web, 47.5 on OSWorld, 50.6 on WindowsAgentArena, and 73.3 on AndroidWorld, outperforming strong baselines such as Claude and OpenAI agents. In game environments, it attains a mean normalized score of 59.8 across a 15-game suite-roughly 60% of human-level performance-and remains competitive with frontier proprietary models (e.g., OpenAI o3) on LMGame-Bench. Additionally, the model can generalize to long-horizon information-seeking tasks and software engineering benchmarks, highlighting its robustness across diverse agent tasks. Detailed analyses of training dynamics further provide insights into achieving stability and efficiency in large-scale agent RL. These results underscore UI-TARS-2's potential to advance the state of GUI agents and exhibit strong generalization to real-world interactive scenarios.

cs.AI↗

arXiv2025

Implementing General-Order Frequency Dynamic Response Model and Frequency Excursion Duration Criterion in Unit Commitment Problem

This paper introduces a novel approach for incorporating frequency dynamics into the unit commitment (UC) problem through a general-order differential equation model, solved using Bernstein polynomial approximation. Traditional frequency-constrained UC (FCUC) models typically rely on simplified first-order assumptions or scalar frequency metrics, such as frequency nadir, to indirectly enforce dynamic behavior. In contrast, our formulation explicitly models time-domain frequency response using second-order dynamics, enabling a more accurate and flexible representation of generator behavior. The resulting differential equations are approximated with high fidelity using Bernstein polynomials, leading to a mixed-integer linear programming (MILP) formulation that remains computationally tractable for small-scale power systems. Additionally, we introduce a new constraint based on the duration of frequency excursions below a critical threshold, motivated by practical concerns such as relay operation and equipment protection. A data-driven method is employed to relate the area under this threshold-computed as the integral of the Bernstein approximation-to the duration of frequency deviation. The proposed framework is validated using real-world data from an island system in Spain, demonstrating enhanced frequency security with a moderate increase in operational cost. These results suggest the method's strong potential for application in low-inertia, small-scale power systems.

eess.SY↗

arXiv2025

Bilevel Mixed-Integer Linear Program with Binary Tender

Bilevel programs model sequential decision interactions between two sets of players and find wide applications in real-world complex systems. In this paper, we consider a bilevel mixed-integer linear program with binary tender, wherein the upper and lower levels are linked via binary decision variables and both levels may involve additional mixed-integer decisions. We recast this bilevel program as a single-level formulation through a value function for the lower-level problem and then propose valid inequalities to replace and iteratively approximate the value function. We first derive a family of Lagrangian-based valid inequalities that give a complete description of the value function, providing a baseline method to obtain exact solutions for the considered class of bilevel programs. To enhance the strength of this approach, we further investigate another two types of valid inequalities. First, when the lower-level value function has intrinsic special properties such as supermodularity or submodularity, we exploit such properties to separate the Lagrangian-based inequalities quickly. Second, we derive decision rule-based valid inequalities, where linear decision rules and learning techniques are explored respectively. We demonstrate the effectiveness and efficiency of the proposed methods in extensive numerical experiments, including instances of general bilevel mixed-integer programs and those of a facility location interdiction problem.

math.OC↗

arXiv2025

Spectrum Prediction in the Fractional Fourier Domain with Adaptive Filtering

Accurate spectrum prediction is crucial for dynamic spectrum access (DSA) and resource allocation. However, due to the unique characteristics of spectrum data, existing methods based on the time or frequency domain often struggle to separate predictable patterns from noise. To address this, we propose the Spectral Fractional Filtering and Prediction (SFFP) framework. SFFP first employs an adaptive fractional Fourier transform (FrFT) module to transform spectrum data into a suitable fractional Fourier domain, enhancing the separability of predictable trends from noise. Subsequently, an adaptive Filter module selectively suppresses noise while preserving critical predictive features within this domain. Finally, a prediction module, leveraging a complex-valued neural network, learns and forecasts these filtered trend components. Experiments on real-world spectrum data show that the SFFP outperforms leading spectrum and general forecasting methods.

cs.LG↗

arXiv2025

ChairPose: Pressure-based Chair Morphology Grounded Sitting Pose Estimation through Simulation-Assisted Training

Prolonged seated activity is increasingly common in modern environments, raising concerns around musculoskeletal health, ergonomics, and the design of responsive interactive systems. Existing posture sensing methods such as vision-based or wearable approaches face limitations including occlusion, privacy concerns, user discomfort, and restricted deployment flexibility. We introduce ChairPose, the first full body, wearable free seated pose estimation system that relies solely on pressure sensing and operates independently of chair geometry. ChairPose employs a two stage generative model trained on pressure maps captured from a thin, chair agnostic sensing mattress. Unlike prior approaches, our method explicitly incorporates chair morphology into the inference process, enabling accurate, occlusion free, and privacy preserving pose estimation. To support generalization across diverse users and chairs, we introduce a physics driven data augmentation pipeline that simulates realistic variations in posture and seating conditions. Evaluated across eight users and four distinct chairs, ChairPose achieves a mean per joint position error of 89.4 mm when both the user and the chair are unseen, demonstrating robust generalization to novel real world generalizability. ChairPose expands the design space for posture aware interactive systems, with potential applications in ergonomics, healthcare, and adaptive user interfaces.

cs.HC↗

arXiv2025

ST-DAI: Single-shot 2.5D Spatial Transcriptomics with Intra-Sample Domain Adaptive Imputation for Cost-efficient 3D Reconstruction

For 3D spatial transcriptomics (ST), the high per-section acquisition cost of fully sampling every tissue section remains a significant challenge. Although recent approaches predict gene expression from histology images, these methods require large external datasets, which leads to high-cost and suffers from substantial domain discrepancies that lead to poor generalization on new samples. In this work, we introduce ST-DAI, a single-shot framework for 3D ST that couples a cost-efficient 2.5D sampling scheme with an intra-sample domain-adaptive imputation framework. First, in the cost-efficient 2.5D sampling stage, one reference section (central section) is fully sampled while other sections (adjacent sections) is sparsely sampled, thereby capturing volumetric context at significantly reduced experimental cost. Second, we propose a single-shot 3D imputation learning method that allows us to generate fully sampled 3D ST from this cost-efficient 2.5D ST scheme, using only sample-specific training. We observe position misalignment and domain discrepancy between sections. To address those issues, we adopt a pipeline that first aligns the central section to the adjacent section, thereafter generates dense pseudo-supervision on the central section, and then performs Fast Multi-Domain Refinement (FMDR), which adapts the network to the domain of the adjacent section while fine-tuning only a few parameters through the use of Parameter-Efficient Domain-Alignment Layers (PDLs). During this refinement, a Confidence Score Generator (CSG) reweights the pseudo-labels according to their estimated reliability, thereby directing imputation toward trustworthy regions. Our experimental results demonstrate that ST-DAI achieves gene expression prediction performance comparable to fully sampled approaches while substantially reducing the measurement burden.

eess.IV↗

arXiv2025

A Differential Evolution Algorithm with Neighbor-hood Mutation for DOA Estimation

Two-dimensional (2D) Multiple Signal Classification algorithm is a powerful technique for high-resolution direction-of-arrival (DOA) estimation in array signal processing. However, the exhaustive search over the 2D an-gular domain leads to high computa-tional cost, limiting its applicability in real-time scenarios. In this work, we reformulate the peak-finding process as a multimodal optimization prob-lem, and propose a Differential Evolu-tion algorithm with Neighborhood Mutation (DE-NM) to efficiently lo-cate multiple spectral peaks without requiring dense grid sampling. Simu-lation results demonstrate that the proposed method achieves comparable estimation accuracy to the traditional grid search, while significantly reduc-ing computation time. This strategy presents a promising solution for real-time, high-resolution DOA estimation in practical applications. The imple-mentation code is available at https://github.com/zzb-nice/DOA_multimodel_optimize.

eess.SP↗