Hasil untuk "Vocational guidance. Career development"

Duane Brown

Jinsheng Wei, Zhaodi Xu, Guanming Lu et al.

Micro-gesture recognition (MGR) is challenging due to subtle inter-class variations. Existing methods rely on category-level supervision, which is insufficient for capturing subtle and localized motion differences. Thus, this paper proposes a Fine-Grained Semantic Guidance Learning (FG-SGL) framework that jointly integrates fine-grained and category-level semantics to guide vision--language models in perceiving local MG motions. FG-SA adopts fine-grained semantic cues to guide the learning of local motion features, while CP-A enhances the separability of MG features through category-level semantic guidance. To support fine-grained semantic guidance, this work constructs a fine-grained textual dataset with human annotations that describes the dynamic process of MGs in four refined semantic dimensions. Furthermore, a Multi-Level Contrastive Optimization strategy is designed to jointly optimize both modules in a coarse-to-fine pattern. Experiments show that FG-SGL achieves competitive performance, validating the effectiveness of fine-grained semantic guidance for MGR.

Yuanyang Yin, Yufan Deng, Shenghai Yuan et al.

The task of Image-to-Video (I2V) generation aims to synthesize a video from a reference image and a text prompt. This requires diffusion models to reconcile high-frequency visual constraints and low-frequency textual guidance during the denoising process. However, while existing I2V models prioritize visual consistency, how to effectively couple this dual guidance to ensure strong adherence to the text prompt remains underexplored. In this work, we observe that in Diffusion Transformer (DiT)-based I2V models, certain intermediate layers exhibit weak semantic responses (termed Semantic-Weak Layers), as indicated by a measurable drop in text-visual similarity. We attribute this to a phenomenon called Condition Isolation, where attention to visual features becomes partially detached from text guidance and overly relies on learned visual priors. To address this, we propose Focal Guidance (FG), which enhances the controllability from Semantic-Weak Layers. FG comprises two mechanisms: (1) Fine-grained Semantic Guidance (FSG) leverages CLIP to identify key regions in the reference frame and uses them as anchors to guide Semantic-Weak Layers. (2) Attention Cache transfers attention maps from semantically responsive layers to Semantic-Weak Layers, injecting explicit semantic signals and alleviating their over-reliance on the model's learned visual priors, thereby enhancing adherence to textual instructions. To further validate our approach and address the lack of evaluation in this direction, we introduce a benchmark for assessing instruction following in I2V models. On this benchmark, Focal Guidance proves its effectiveness and generalizability, raising the total score on Wan2.1-I2V to 0.7250 (+3.97\%) and boosting the MMDiT-based HunyuanVideo-I2V to 0.5571 (+7.44\%).

هانیه غفاری, محمد قهرمانی, فرنوش اعلامی

هدف: هدف از پژوهش حاضر امکان‌سنجی طراحی یک شبکه ارتباطی آنلاین جهت بهبود کیفیت مدیریت مدارس بود.روش: پژوهش حاضر به لحاظ هدف از نوع پژوهش‌های کاربردی و توسعه‌ای بود و روش جمع‌آوری داده‌ها به‌صورت روش ترکیبی بود. در این راستا مشارکت‌کنندگان بخش کیفی شامل اساتید هیئت‌علمی علوم تربیتی دانشگاه‌های دولتی و مدیران مدارس متوسطه 10 نفر که به روش هدفمند انتخاب شدند مصاحبه‌های نیمه ساختاریافته انجام شد و در بخش کمی تحقیق بر اساس مضامین استخراج‌شده از مصاحبه‌ها و اهداف پژوهشی پرسشنامه 28 گویه‌ای طراحی شد و بین 116 نفر از مدیران که به روش تصادفی طبقه ای انتخاب شده بودند توزیع شد. یافته ها و نتیجه گیری: نتایج نشان می‌دهد که مدیران مدارس متوسطه اهمیت و ضرورت وجود شبکه‌های توسعه حرفه‌ای را با میانگین 3.83 درک می‌کنند و منابع انسانی با میانگین 4.33، فنی 4.33، مالی 4.11 و قانونی 4.27 مطرح‌شده را با میانگین بالاتر از متوسط جامعه (3) لازم می‌دانند؛ ولی درعین‌حال چالش‌ها و موانع به‌خصوص موانع سازمانی و بروکراتیک را تأثیرگذار می‌دانند

Thomas Staunton

The purpose of this paper is to explore the process through which higher education students in the U.K. are inducted into using digital platforms for their careers. This paper will present data from a longitudinal study looking at how students use digital social media platforms as part of their career transitions after graduating. A key finding from this study is that the university plays a central role in why and how students start using digital platforms for their career. Qualitative data was gathered from students both before and after graduating about their experiences of using digital platforms for their careers. This data was analysed using an interpretative phenomenological analysis. The themes developed from this pointed to four main themes: induction, hybridity, imaginations and exposure to e-safety. This is contextualised through exploring this process as an example of the platformisation of career in a higher education context. Abstrakti Tämän artikkelin tarkoitus on tarkastella prosessia, jonka kautta Britannian korkeakouluopiskelijat perehdytetään käyttämään digitaalisia alustoja urasuunnittelussaan. Artikkeli esittelee aineiston pitkittäistutkimuksesta, jonka aiheena on miten opiskelijat käyttävät digitaalisia sosiaalisen median alustoja osana valmistumisen jälkeisiä urasiirtymiä. Keskeinen tulos tästä tutkimuksesta on, että yliopistolla on keskeinen rooli siinä miksi ja miten opiskelijat alkavat käyttää digitaalisia alustoja osana urasuunnitteluaan. Opiskelijoiden keskuudesta kerättiin laadullista aineistoa sekä ennen että jälkeen valmistumisen koskien heidän kokemuksiaan urasuunnitteluun liittyvistä digitaalisista alustoista. Tämä aineisto analysoitiin tulkitsevan fenomenologisen analyysin avulla. Analyysin perusteella aineistosta erotettiin neljä pääteemaa: perehdytys, hybriditeetti, ennakkouskomukset sekä digitaalinen turvallisuus. Nämä pääteemat kontekstualisoitiin tarkastelemalla opiskelijoiden prosessia esimerkkinä uran nivoutumisesta alustoihin korkeakoulukontekstissa. Asiasanat: urasuunnittelu; uraohjaus; digitalisuus; korkea-aste

Raquel Bulgarelli-Bolaños

Objetivo: Avaliar os principais resultados do teste piloto do Programa de Treinamento Integral: Projetando Meu Futuro, especificamente o MOOC Tempo para Tudo, para alunos do primeiro semestre da Universidade LCI Veritas. Metodologia: Trata-se de um estudo misto com delineamento explicativo sequencial. Foram considerados dados quantitativos coletados da pesquisa de avaliação e de um instrumento de hábitos de gerenciamento de tempo (pós e pré-teste) e, então, com dados qualitativos dessa avaliação, os produtos do curso e as informações das entrevistas com os professores foram explorados em profundidade. A principal fonte foram os alunos participantes (n=161) e as demais fontes foram dois professores representantes dos cursos participantes do teste piloto e 483 documentos. Resultados: Há uma alta qualidade pedagógica e um aumento nas habilidades esperadas de gerenciamento de tempo. 76% dos alunos estão muito satisfeitos, especialmente com os recursos de aprendizagem, o design instrucional e o conteúdo. De fato, 86% dos alunos recomendariam o MOOC a outro colega, e 87% da população atingiu uma alta frequência de hábitos de gerenciamento de tempo ao final do MOOC, principalmente no planejamento. Conclusões: Os resultados do piloto do MOOC Tiempo para Todo são avaliados como altamente positivos tanto para a formação dos alunos quanto para o trabalho dos professores e do Gabinete de Orientação. Sua aplicação é recomendada em todos os cursos universitários, planejando ações para melhorar a motivação dos alunos e dando maior ênfase ao uso de estratégias de gestão do tempo.

Xuesong Li, Dianye Huang, Yameng Zhang et al.

Understanding medical ultrasound imaging remains a long-standing challenge due to significant visual variability caused by differences in imaging and acquisition parameters. Recent advancements in large language models (LLMs) have been used to automatically generate terminology-rich summaries orientated to clinicians with sufficient physiological knowledge. Nevertheless, the increasing demand for improved ultrasound interpretability and basic scanning guidance among non-expert users, e.g., in point-of-care settings, has not yet been explored. In this study, we first introduce the scene graph (SG) for ultrasound images to explain image content to ordinary and provide guidance for ultrasound scanning. The ultrasound SG is first computed using a transformer-based one-stage method, eliminating the need for explicit object detection. To generate a graspable image explanation for ordinary, the user query is then used to further refine the abstract SG representation through LLMs. Additionally, the predicted SG is explored for its potential in guiding ultrasound scanning toward missing anatomies within the current imaging view, assisting ordinary users in achieving more standardized and complete anatomical exploration. The effectiveness of this SG-based image explanation and scanning guidance has been validated on images from the left and right neck regions, including the carotid and thyroid, across five volunteers. The results demonstrate the potential of the method to maximally democratize ultrasound by enhancing its interpretability and usability for ordinaries.

Mingkang Zhu, Xi Chen, Zhongdao Wang et al.

Recent advancements in reinforcement learning from human feedback have shown that utilizing fine-grained token-level reward models can substantially enhance the performance of Proximal Policy Optimization (PPO) in aligning large language models. However, it is challenging to leverage such token-level reward as guidance for Direct Preference Optimization (DPO), since DPO is formulated as a sequence-level bandit problem. To address this challenge, this work decomposes the sequence-level PPO into a sequence of token-level proximal policy optimization problems and then frames the problem of token-level PPO with token-level reward guidance, from which closed-form optimal token-level policy and the corresponding token-level reward can be derived. Using the obtained reward and Bradley-Terry model, this work establishes a framework of computable loss functions with token-level reward guidance for DPO, and proposes a practical reward guidance based on the induced DPO reward. This formulation enables different tokens to exhibit varying degrees of deviation from reference policy based on their respective rewards. Experiment results demonstrate that our method achieves substantial performance improvements over DPO, with win rate gains of up to 7.5 points on MT-Bench, 6.2 points on AlpacaEval 2, and 4.3 points on Arena-Hard. Code is available at https://github.com/dvlab-research/TGDPO.

Chenyu Wang, Cai Zhou, Sharut Gupta et al.

Diffusion models can be improved with additional guidance towards more effective representations of input. Indeed, prior empirical work has already shown that aligning internal representations of the diffusion model with those of pre-trained models improves generation quality. In this paper, we present a systematic framework for incorporating representation guidance into diffusion models. We provide alternative decompositions of denoising models along with their associated training criteria, where the decompositions determine when and how the auxiliary representations are incorporated. Guided by our theoretical insights, we introduce two new strategies for enhancing representation alignment in diffusion models. First, we pair examples with target representations either derived from themselves or arisen from different synthetic modalities, and subsequently learn a joint model over the multimodal pairs. Second, we design an optimal training curriculum that balances representation learning and data generation. Our experiments across image, protein sequence, and molecule generation tasks demonstrate superior performance as well as accelerated training. In particular, on the class-conditional ImageNet $256\times 256$ benchmark, our guidance results in $23.3$ times faster training than the original SiT-XL as well as four times speedup over the state-of-the-art method REPA. The code is available at https://github.com/ChenyuWang-Monica/REED.

Debodeep Banerjee, Burcu Sayin, Stefano Teso et al.

Medical decision-making is a critical task, where errors can result in serious, potentially life-threatening consequences. While full automation remains challenging, hybrid frameworks that combine machine intelligence with human oversight offer a practical alternative. In this paper, we present MedGellan, a lightweight, annotation-free framework that uses a Large Language Model (LLM) to generate clinical guidance from raw medical records, which is then used by a physician to predict diagnoses. MedGellan uses a Bayesian-inspired prompting strategy that respects the temporal order of clinical data. Preliminary experiments show that the guidance generated by the LLM with MedGellan improves diagnostic performance, particularly in recall and $F_1$ score.

Mahdis Tajdari, Jason Forsyth, Sol Lim

Navigating peripersonal space requires reaching targets in both horizontal (e.g., desks) and vertical (e.g., shelves) layouts with high precision. We developed a haptic glove to aid peri-personal target navigation and investigated the effectiveness of different feedback delivery methods. Twenty-two participants completed target navigation tasks under various conditions, including scene layout (horizontal or vertical), guidance approach (two-tactor or worst-axis first), guidance metaphor (push or pull), and intensity mode (linear or zone) for conveying distance cues. Task completion time, hand trajectory distance, and the percentage of hand trajectory in a critical area were measured as performance outcomes, along with subjective feedback. Participants achieved significantly faster task completion times and covered less hand trajectory distance in the horizontal layout, worst-axis first approach, and pull metaphor conditions. Additionally, male participants demonstrated superior performance and reported lower levels of frustration compared to their female counterparts throughout the study. Intensity mode had no significant effect on the results. In summary, vibrating one tactor at a time (worst-axis first) and using the pull metaphor were the most effective methods of delivering vibrotactile feedback for peripersonal target navigation in both horizontal and vertical settings. Findings from this work can guide future development of haptic gloves for individuals with vision impairments, environments with visual limitations, and for accessibility and rehabilitation applications.

Gilbert Bahati, Ryan M. Bena, Meg Wilkinson et al.

Robotic systems navigating in real-world settings require a semantic understanding of their environment to properly determine safe actions. This work aims to develop the mathematical underpinnings of such a representation -- specifically, the goal is to develop safety filters that are risk-aware. To this end, we take a two step approach: encoding an understanding of the environment via Poisson's equation, and associated risk via Laplace guidance fields. That is, we first solve a Dirichlet problem for Poisson's equation to generate a safety function that encodes system safety as its 0-superlevel set. We then separately solve a Dirichlet problem for Laplace's equation to synthesize a safe \textit{guidance field} that encodes variable levels of caution around obstacles -- by enforcing a tunable flux boundary condition. The safety function and guidance fields are then combined to define a safety constraint and used to synthesize a risk-aware safety filter which, given a semantic understanding of an environment with associated risk levels of environmental features, guarantees safety while prioritizing avoidance of higher risk obstacles. We demonstrate this method in simulation and discuss how \textit{a priori} understandings of obstacle risk can be directly incorporated into the safety filter to generate safe behaviors that are risk-aware.

مریم جعغری, محمد رضا عابدی, فاطمه سمیعی et al.

با توجه به افزایش روزافزون نرخ بیکاری، مداخلات مشاوره‌ای شغلی می‌توانند نقش حیاتی در انجام اقدامات لازم جهت مقابله با این بحران ایفا کنند. علی‌رغم تحقیقات گسترده در مورد مداخله‌های مشاوره شغلی برای افراد بیکار در یک قرن گذشته در جهان، این حیطه در داخل مرزهای کشور فاقد مطالعه منظم و سیستماتیک بوده و اغلب بر نظریات محدود تأکید داشته است. جهت دست‌یابی به طبقه‌بندی و شناخت بهتر اقداماتی که تاکنون در این زمینه صورت گرفته، یک مرور سیستماتیک از سال‌های 1398 تا تیرماه 1403 و 2018 تا 2023 میلادی صورت گرفت. نتایج کمک کرد تا دیدگاه‌های نظری برجسته اتخاذشده در مداخلات طی این پنج سال بررسی و مهم‌ترین مداخلات و متغیرها شناسایی و همچنین گروه‌های هدف در مداخلات ویژه بیکاران مشخص شوند. این بررسی همچنین کمک کرد محدودیت‌های تحقیقاتی مهم را شناسایی و شکاف‌های پژوهشی امیدوارکننده پیشنهاد شود. با تکیه بر این بینش‌ها، یک دستور کار تحقیقاتی آینده ایجاد می‌شود که فرصت‌هایی را برای پیشرفت مفهومی، نظری و تجربی تحقیقات مداخلات مشاوره‌ای شغلی برای بیکاران برجسته می‌کند.

,وحید قهرمانی, امید نوری, رضا دیهیم فرد et al.

شغل و ماهیت آن عامل مهمی است که بر زندگی فرد تاثیر مستقیم می‌گذارد. ماهیت شغل دربرگیرنده محیط فعالیت، فواید و زیان‌های حاصل از آن است. نوع شغل می‌تواند بر سلامت روانی و تاب‌آوری افراد تاثیرگذار باشد. در این پژوهش اثر سه گروه شغلی کشاورزی شهری، شغل دولتی و آزاد بر تاب‌آوری روان‌شناختی افراد بررسی شد. برای این منظور از 5 مولفه مهارت اجتماعی، ظرفیت مقابله با استرس‎ها، مثبت‌اندیشی، احساس سرزندگی و انگیزه استفاده شد. پایایی پرسشنامه 81/0 و روایی آن به روش صوری محتوایی تایید شد. 156 نفر از سه گروه شغلی، به‌صورت هدفمند و به روش گلوله‌برفی انتخاب شدند. برای تجزیه و تحلیل داده‌ها در سطح آمار استنباطی از تحلیل واریانس دوطرفه و تحلیل چند متغیره و نرم‌افزار SPSS استفاده شد. نتایج نشان داد تاب‌آوری روان‌شناختی افراد در سه گروه‌ شغلی متفاوت بوده و افرادی که در زمینه کشاورزی شهری فعالیت دارند دارای تاب‎آوری بیش‌تر نسبت به دو گروه شغلی هستند (000/0=P). نمره کل تاب‌آوری در گروه‌های سنی تفاوت معنی‌داری دارند و بیش‌ترین مقدار میانگین (85/19) برای گروه کشاورزان شهری در گروه سنی بالای 50 سال است. اثر تعاملی برای گروه‌های سنی و نوع اشتغال نیز معنی‌دار بود (004/0=P). نمره کل تاب‌آوری روان‌شناختی با توجه به جنسیت افراد تفاوت معنی‌داری نشان نداد. هم‌چنین نتایج تحلیل واریانس دوراهه بین‌گروهی نشان داد که اثر تعاملی بین گروه شغلی و جنسیت افراد معنی‌دار نیست. در راستای بهبود تاب‌آوری افراد و همبستگی آن با شغل، پیشنهاد می‌شود سایر ابعاد تاب‌آوری و ارتباط آن با نوع شغل، بخصوص در گروه زنان بررسی شود.

سپیده عبداللهی, آزاده چوب فروش زاده, حمیدرضا آریانپور et al.

مقدمه: انتخاب شغل یکی از دغدغه‌های همیشگی دانش‌آموزان بوده است. این امر باید در نهایت آگاهی و شناخت از خود و توانمندی‌ها انجام شود. در صورت انتخاب نامناسب و ناآگاهانه در زمینه‌ تحصیلی و شغلی، فرد امید شغلی خود را از دست داده و هیجان‌های منفی مانند استرس و اضطراب را تجربه خواهد کرد. هدف: این پژوهش با هدف تعیین اثربخشی مداخله‌ مبتنی بر حکایت درمانی مسیر شغلی کوکران بر پریشانی ‌روان‌شناختی و امید شغلی در دانش‌آموزان مقطع متوسطه انجام گرفت. روش: پژوهش حاضر از نوع شبه‌آزمایشی با طرح پیش‌آزمون، پس‌آزمون با گروه کنترل بود. جامعه‌ این پژوهش کلیه‌ دانش‌آموزان دختر و پسر مقطع متوسطه دوم شهرستان اصفهان در سال تحصیلی 1403-1402 بودند که از بین آن‌ها 30 دانش‌آموز به شیوه‌ نمونه‌گیری در دسترس انتخاب و به صورت تصادفی در دو گروه آزمایش و کنترل (هرگروه 15 نفر) قرار گرفتند. گروه آزمایش به صورت هفتگی در هفت جلسه (90 دقیقه‌ای) تحت آموزش مشاوره مسیر شغلی کوکران قرار گرفتند. داده‌ها با استفاده از پرسشنامه افسردگی، اضطراب، استرس و پرسشنامه‌ امید شغلی گردآوری شدند و با استفاده از نرم افزار SPSS-26 و روش تحلیل کوواریانس چند متغیره تجزیه و تحلیل شدند.یافته‌ها: نتایج نشان داد که مداخله‌ مبتنی بر حکایت درمانی مسیر شغلی کوکران منجر به کاهش پریشانی‌ روان شناختی و بهبود امید شغلی در دانش‌آموزان گروه آزمایش نسبت به دانش‌آموزان گروه کنترل گشته است (05/0p<). نتیجه‌گیری: براساس یافته‌ها می‌توان نتیجه گرفت که حکایت درمانی مسیر شغلی کوکران برای افزایش امید شغلی و کاهش پریشانی روان-شناختی در دانش‌آموزان تأثیرگذار بوده است.

Linxuan Xin, Zheng Zhang, Jinfu Wei et al.

Prior material creation methods had limitations in producing diverse results mainly because reconstruction-based methods relied on real-world measurements and generation-based methods were trained on relatively small material datasets. To address these challenges, we propose DreamPBR, a novel diffusion-based generative framework designed to create spatially-varying appearance properties guided by text and multi-modal controls, providing high controllability and diversity in material generation. Key to achieving diverse and high-quality PBR material generation lies in integrating the capabilities of recent large-scale vision-language models trained on billions of text-image pairs, along with material priors derived from hundreds of PBR material samples. We utilize a novel material Latent Diffusion Model (LDM) to establish the mapping between albedo maps and the corresponding latent space. The latent representation is then decoded into full SVBRDF parameter maps using a rendering-aware PBR decoder. Our method supports tileable generation through convolution with circular padding. Furthermore, we introduce a multi-modal guidance module, which includes pixel-aligned guidance, style image guidance, and 3D shape guidance, to enhance the control capabilities of the material LDM. We demonstrate the effectiveness of DreamPBR in material creation, showcasing its versatility and user-friendliness on a wide range of controllable generation and editing applications.

Jie Qi, Ronghua Wanga, Nailong Wu

The focus of this paper is to develop a methodology that enables an unmanned surface vehicle (USV) to efficiently track a planned path. The introduction of a vector field-based adaptive line of-sight guidance law (VFALOS) for accurate trajectory tracking and minimizing the overshoot response time during USV tracking of curved paths improves the overall line-of-sight (LOS) guidance method. These improvements contribute to faster convergence to the desired path, reduce oscillations, and can mitigate the effects of persistent external disturbances. It is shown that the proposed guidance law exhibits k-exponential stability when converging to the desired path consisting of straight and curved lines. The results in the paper show that the proposed method effectively improves the accuracy of the USV tracking the desired path while ensuring the safety of the USV work.

Boyao Peng, Lexing Zhang, Enkai Li

In the upstream of the exit ramp of the expressway, the speed limit difference leads to a significant deceleration of the vehicle in the area adjacent to the off-ramp. The friction coefficient of the road surface decreases under rainy weather, and the above deceleration process can easily lead to sideslip and rollover of the vehicle. Dynamic speed guidance is an effective way to improve the status quo. Currently, there is an emerging trend to utilize I2V technology and high-precision map technology for lane level speed guidance control. This paper presents an optimized joint control strategy for main line-off-ramp speed guidance, which can adjust the guidance speed in real time according to the rainfall intensity. At the same time, this paper designs a progressive deceleration strategy, which works together with the speed guidance control to ensure the safe deceleration of vehicles. The simulation results show that the proposed control strategy outperforms the fixed speed limit control in terms of improving the total traveled time (TTT), total traveled distance (TTD) and standard deviation of speed (SD). Sensitivity analysis shows that the proposed control strategy can improve performance with the increase of the compliance rate of drivers. The speed guidance control method established in this paper can improve the vehicle operation efficiency in the off-ramp area of the expressway and reduce the speed difference of each vehicle in rainy weather, which guarantee the safety of expressway driving in the rainy day.

Jun Xiao, Zihang Lyu, Hao Xie et al.

Blind image restoration remains a significant challenge in low-level vision tasks. Recently, denoising diffusion models have shown remarkable performance in image synthesis. Guided diffusion models, leveraging the potent generative priors of pre-trained models along with a differential guidance loss, have achieved promising results in blind image restoration. However, these models typically consider data consistency solely in the spatial domain, often resulting in distorted image content. In this paper, we propose a novel frequency-aware guidance loss that can be integrated into various diffusion models in a plug-and-play manner. Our proposed guidance loss, based on 2D discrete wavelet transform, simultaneously enforces content consistency in both the spatial and frequency domains. Experimental results demonstrate the effectiveness of our method in three blind restoration tasks: blind image deblurring, imaging through turbulence, and blind restoration for multiple degradations. Notably, our method achieves a significant improvement in PSNR score, with a remarkable enhancement of 3.72\,dB in image deblurring. Moreover, our method exhibits superior capability in generating images with rich details and reduced distortion, leading to the best visual quality.

Bingchen Yang, Haiyong Jiang, Hao Pan et al.

Reverse engineering CAD models from raw geometry is a classic but challenging research problem. In particular, reconstructing the CAD modeling sequence from point clouds provides great interpretability and convenience for editing. To improve upon this problem, we introduce geometric guidance into the reconstruction network. Our proposed model, PS-CAD, reconstructs the CAD modeling sequence one step at a time. At each step, we provide two forms of geometric guidance. First, we provide the geometry of surfaces where the current reconstruction differs from the complete model as a point cloud. This helps the framework to focus on regions that still need work. Second, we use geometric analysis to extract a set of planar prompts, that correspond to candidate surfaces where a CAD extrusion step could be started. Our framework has three major components. Geometric guidance computation extracts the two types of geometric guidance. Single-step reconstruction computes a single candidate CAD modeling step for each provided prompt. Single-step selection selects among the candidate CAD modeling steps. The process continues until the reconstruction is completed. Our quantitative results show a significant improvement across all metrics. For example, on the dataset DeepCAD, PS-CAD improves upon the best published SOTA method by reducing the geometry errors (CD and HD) by 10%, and the structural error (ECD metric) by about 15%.