arXiv subjects

Stefano Petrangeli

Publications and source records attributed to Stefano Petrangeli.

13 recordsLinked to original sources

arXiv2026

CamPilot: A Multi-Agent Cinematic Assistant for Camera-Controlled Movie Generation

The integration of large language models (LLMs) into video generation has enabled rapid text-to-video creation and improved visual quality. However, it still falls short of professional filmmaking, where cinematographic language is less refined than human-crafted camera work and multi-shot continuity remains challenging. To address these limitations, we introduce CamPilot, a multi-agent framework that integrates cinematographic planning and camera-work control to produce more coherent, logically structured, and human-aesthetic movies. CamPilot adopts a GRPO-based learning paradigm to learn camera work planning from 14K real-world professional movies, internalizing motion patterns and composition principles that support reasoning over shooting techniques (e.g., camera angle, motion, and focal behavior) and cross-shot relationships for controllable camera-viewpoint generation. Multiple agents further collaborate and evolve to improve overall output quality. To support this work and further studies in this domain, we establish CamEval, a benchmark for evaluating camera work quality and cinematic engagement. Empirical results show that CamPilot outperforms state-of-the-art text-to-movie generation methods on cinematographic control and quality, highlighting the impact of professional camera design on movie generation.

cs.CV↗

arXiv2026

Agentic Planning with Reasoning for Image Styling via Offline RL

Direct prompt-based editing often fails on complex transformations because vague and subjective prompts often require nuanced understanding of what should be changed in the image. Our core intuition is that leveraging compositional image editing tools rather than direct prompting profits from structured agent-level planning with explicit reasoning, leading to better results. This structured planning framework enables efficient offline RL post-training on quality-scored trajectories to improve performance. We present a tool-based agentic RL post-training framework that addresses this through structured planning with chain-of-thought reasoning. Our key contributions include: (1) A tool-based agentic planning methodology that combines a compositional library of orthogonal primitive transformations, structured context representation, and explicit per-step reasoning to decompose complex styling into interpretable tool sequences. (2) A synthetic data generation pipeline producing three large-scale datasets (each $\sim$10K trajectories) with reasoning chains, plans, and quality scores, as no existing datasets provide such supervision. Our datasets and code are publicly available at the HuggingFace repository. (3) Offline RL training methods for learning planners with reasoning as our core algorithmic contributions, which consistently improve over the Edit-Only baseline in visual quality and instruction following. (4) Comprehensive evaluation across 4B and 8B parameter Qwen3-VL models showing that our methods outperform other baselines in the majority of compositional tasks, validated by human evaluations.

cs.LG↗

arXiv2026

From Pixels to Policies: Reinforcing Spatial Reasoning in Language Models for Content-Aware Layout Design

We introduce LaySPA, a reinforcement learning framework that equips large language models (LLMs) with explicit and interpretable spatial reasoning for content-aware graphic layout design. LaySPA addresses two key challenges: LLMs' limited spatial reasoning and the lack of opacity in design decision making. Instead of operating at the pixel level, we reformulate layout design as a policy learning problem over a structured textual spatial environment that explicitly encodes canvas geometry, element attributes, and inter-element relationships. LaySPA produces dual-level outputs comprising interpretable reasoning traces and structured layout specifications, enabling transparent and controllable design decision making. Layout design policy is optimized via a multi-objective spatial critique that decomposes layout quality into geometric validity, relational coherence, and aesthetic consistency, and is trained using relative group optimization to stabilize learning in open-ended design spaces. Experiments demonstrate that LaySPA improves structural validity and visual quality, outperforming larger proprietary LLMs and achieving performance comparable to specialized SOTA layout generators while requiring fewer annotated samples and reduced latency.

cs.AI↗

arXiv2026

LLMs as Layout Designers: Enhanced Spatial Reasoning for Content-Aware Layout Generation

While Large Language Models (LLMs) have demonstrated impressive reasoning and planning abilities in textual domains and can effectively follow instructions for complex tasks, their ability to understand and manipulate spatial relationships remains limited. Such capabilities are crucial for content-aware graphic layout design, where the goal is to arrange heterogeneous elements onto a canvas so that final design remains visually balanced and structurally feasible. This problem requires precise coordination of placement, alignment, and structural organization of multiple elements within a constrained visual space. To address this limitation, we introduce LaySPA, a reinforcement learning-based framework that augments LLM-based agents with explicit spatial reasoning capabilities for layout design. LaySPA employs hybrid reward signals that jointly capture geometric constraints, structural fidelity, and visual quality, enabling agents to navigate the canvas, model inter-element relationships, and optimize spatial arrangements. Through group-relative policy optimization, the agent generates content-aware layouts that reflect salient regions, respect spatial constraints, and produces an interpretable reasoning trace explaining placement decisions and a structured layout specification. Experimental results show that LaySPA substantially improves the generation of structurally valid and visually appealing layouts, outperforming larger general-purpose LLMs and achieving performance comparable to state-of-the-art specialized layout models.

cs.AI↗

arXiv2026

Proc3D: Procedural 3D Generation and Parametric Editing of 3D Shapes with Large Language Models

Generating 3D models has traditionally been a complex task requiring specialized expertise. While recent advances in generative AI have sought to automate this process, existing methods produce non-editable representation, such as meshes or point clouds, limiting their adaptability for iterative design. In this paper, we introduce Proc3D, a system designed to generate editable 3D models while enabling real-time modifications. At its core, Proc3D introduces procedural compact graph (PCG), a graph representation of 3D models, that encodes the algorithmic rules and structures necessary for generating the model. This representation exposes key parameters, allowing intuitive manual adjustments via sliders and checkboxes, as well as real-time, automated modifications through natural language prompts using Large Language Models (LLMs). We demonstrate Proc3D's capabilities using two generative approaches: GPT-4o with in-context learning (ICL) and a fine-tuned LLAMA-3 model. Experimental results show that Proc3D outperforms existing methods in editing efficiency, achieving more than 400x speedup over conventional approaches that require full regeneration for each modification. Additionally, Proc3D improves ULIP scores by 28%, a metric that evaluates the alignment between generated 3D models and text prompts. By enabling text-aligned 3D model generation along with precise, real-time parametric edits, Proc3D facilitates highly accurate text-based image editing applications.

cs.GR↗

arXiv2026

PRISM: Learning Design Knowledge from Data for Stylistic Design Improvement

Graphic design often involves exploring different stylistic directions, which can be time-consuming for non-experts. We address this problem of stylistically improving designs based on natural language instructions. While VLMs have shown initial success in graphic design, their pretrained knowledge on styles is often too general and misaligned with specific domain data. For example, VLMs may associate minimalism with abstract designs, whereas designers emphasize shape and color choices. Our key insight is to leverage design data -- a collection of real-world designs that implicitly capture designer's principles -- to learn design knowledge and guide stylistic improvement. We propose PRISM (PRior-Informed Stylistic Modification) that constructs and applies a design knowledge base through three stages: (1) clustering high-variance designs to capture diversity within a style, (2) summarizing each cluster into actionable design knowledge, and (3) retrieving relevant knowledge during inference to enable style-aware improvement. Experiments on the Crello dataset show that PRISM achieves the highest average rank of 1.49 (closer to 1 is better) over baselines in style alignment. User studies further validate these results, showing that PRISM is consistently preferred by designers.

cs.AI↗

arXiv2025

X-Reflect: Cross-Reflection Prompting for Multimodal Recommendation

Large Language Models (LLMs) have been shown to enhance the effectiveness of enriching item descriptions, thereby improving the accuracy of recommendation systems. However, most existing approaches either rely on text-only prompting or employ basic multimodal strategies that do not fully exploit the complementary information available from both textual and visual modalities. This paper introduces a novel framework, Cross-Reflection Prompting, termed X-Reflect, designed to address these limitations by prompting Multimodal Large Language Models (MLLMs) to explicitly identify and reconcile supportive and conflicting information between text and images. By capturing nuanced insights from both modalities, this approach generates more comprehensive and contextually rich item representations. Extensive experiments conducted on two widely used benchmarks demonstrate that our method outperforms existing prompting baselines in downstream recommendation accuracy. Furthermore, we identify a U-shaped relationship between text-image dissimilarity and recommendation performance, suggesting the benefit of applying multimodal prompting selectively. To support efficient real-time inference, we also introduce X-Reflect-keyword, a lightweight variant that summarizes image content using keywords and replaces the base model with a smaller backbone, achieving nearly 50% reduction in input length while maintaining competitive performance. This work underscores the importance of integrating multimodal information and presents an effective solution for improving item understanding in multimodal recommendation systems.

cs.IR↗

arXiv2025

ScreenLLM: Stateful Screen Schema for Efficient Action Understanding and Prediction

Graphical User Interface (GUI) agents are autonomous systems that interpret and generate actions, enabling intelligent user assistance and automation. Effective training of these agent presents unique challenges, such as sparsity in supervision signals, scalability for large datasets, and the need for nuanced user understanding. We propose stateful screen schema, an efficient representation of GUI interactions that captures key user actions and intentions over time. Building on this foundation, we introduce ScreenLLM, a set of multimodal large language models (MLLMs) tailored for advanced UI understanding and action prediction. Extensive experiments on both open-source and proprietary models show that ScreenLLM accurately models user behavior and predicts actions. Our work lays the foundation for scalable, robust, and intelligent GUI agents that enhance user interaction in diverse software environments.

cs.CL↗

arXiv2024

HighlightMe: Detecting Highlights from Human-Centric Videos

We present a domain- and user-preference-agnostic approach to detect highlightable excerpts from human-centric videos. Our method works on the graph-based representation of multiple observable human-centric modalities in the videos, such as poses and faces. We use an autoencoder network equipped with spatial-temporal graph convolutions to detect human activities and interactions based on these modalities. We train our network to map the activity- and interaction-based latent structural representations of the different modalities to per-frame highlight scores based on the representativeness of the frames. We use these scores to compute which frames to highlight and stitch contiguous frames to produce the excerpts. We train our network on the large-scale AVA-Kinetics action dataset and evaluate it on four benchmark video highlight datasets: DSH, TVSum, PHD2, and SumMe. We observe a 4-12% improvement in the mean average precision of matching the human-annotated highlights over state-of-the-art methods in these datasets, without requiring any user-provided preferences or dataset-specific fine-tuning.

cs.CV↗

arXiv2024

Show Me What I Like: Detecting User-Specific Video Highlights Using Content-Based Multi-Head Attention

We propose a method to detect individualized highlights for users on given target videos based on their preferred highlight clips marked on previous videos they have watched. Our method explicitly leverages the contents of both the preferred clips and the target videos using pre-trained features for the objects and the human activities. We design a multi-head attention mechanism to adaptively weigh the preferred clips based on their object- and human-activity-based contents, and fuse them using these weights into a single feature representation for each user. We compute similarities between these per-user feature representations and the per-frame features computed from the desired target videos to estimate the user-specific highlight clips from the target videos. We test our method on a large-scale highlight detection dataset containing the annotated highlights of individual users. Compared to current baselines, we observe an absolute improvement of 2-4% in the mean average precision of the detected highlights. We also perform extensive ablation experiments on the number of preferred highlight clips associated with each user as well as on the object- and human-activity-based feature representations to validate that our method is indeed both content-based and user-specific.

cs.CV↗

arXiv2024

Large Content And Behavior Models To Understand, Simulate, And Optimize Content And Behavior

Shannon and Weaver's seminal information theory divides communication into three levels: technical, semantic, and effectiveness. While the technical level deals with the accurate reconstruction of transmitted symbols, the semantic and effectiveness levels deal with the inferred meaning and its effect on the receiver. Large Language Models (LLMs), with their wide generalizability, make some progress towards the second level. However, LLMs and other communication models are not conventionally designed for predicting and optimizing communication for desired receiver behaviors and intents. As a result, the effectiveness level remains largely untouched by modern communication systems. In this paper, we introduce the receivers' "behavior tokens," such as shares, likes, clicks, purchases, and retweets, in the LLM's training corpora to optimize content for the receivers and predict their behaviors. Other than showing similar performance to LLMs on content understanding tasks, our trained models show generalization capabilities on the behavior dimension for behavior simulation, content simulation, behavior understanding, and behavior domain adaptation. We show results on all these capabilities using a wide range of tasks on three corpora. We call these models Large Content and Behavior Models (LCBMs). Further, to spur more research on LCBMs, we release our new Content Behavior Corpus (CBC), a repository containing communicator, message, and corresponding receiver behavior (https://behavior-in-the-wild.github.io/LCBM).

cs.CL↗

arXiv2023

VADER: Video Alignment Differencing and Retrieval

We propose VADER, a spatio-temporal matching, alignment, and change summarization method to help fight misinformation spread via manipulated videos. VADER matches and coarsely aligns partial video fragments to candidate videos using a robust visual descriptor and scalable search over adaptively chunked video content. A transformer-based alignment module then refines the temporal localization of the query fragment within the matched video. A space-time comparator module identifies regions of manipulation between aligned content, invariant to any changes due to any residual temporal misalignments or artifacts arising from non-editorial changes of the content. Robustly matching video to a trusted source enables conclusions to be drawn on video provenance, enabling informed trust decisions on content encountered.

cs.CV↗

arXiv2022

Instant Reality: Gaze-Contingent Perceptual Optimization for 3D Virtual Reality Streaming

Media streaming has been adopted for a variety of applications such as entertainment, visualization, and design. Unlike video/audio streaming where the content is usually consumed sequentially, 3D applications such as gaming require streaming 3D assets to facilitate client-side interactions such as object manipulation and viewpoint movement. Compared to audio and video streaming, 3D streaming often requires larger data sizes and yet lower latency to ensure sufficient rendering quality, resolution, and latency for perceptual comfort. Thus, streaming 3D assets can be even more challenging than streaming audios/videos, and existing solutions often suffer from long loading time or limited quality. To address this critical and timely issue, we propose a perceptually-optimized progressive 3D streaming method for spatial quality and temporal consistency in immersive interactions. Based on the human visual mechanisms in the frequency domain, our model selects and schedules the streaming dataset for optimal spatial-temporal quality. We also train a neural network for our model to accelerate this decision process for real-time client-server applications. We evaluate our method via subjective studies and objective analysis under varying network conditions (from 3G to 5G) and client devices (HMD and traditional displays), and demonstrate better visual quality and temporal consistency than alternative solutions.

cs.HC↗