Mike Roberts

Matt Deitke, Dhruv Batra, Yonatan Bisk et al. · allen-ai, cmu

We present a retrospective on the state of Embodied AI research. Our analysis focuses on 13 challenges presented at the Embodied AI Workshop at CVPR. These challenges are grouped into three themes: (1) visual navigation, (2) rearrangement, and (3) embodied vision-and-language. We discuss the dominant datasets within each theme, evaluation metrics for the challenges, and the performance of state-of-the-art models. We highlight commonalities between top approaches to the challenges and identify potential future directions for Embodied AI research.

Yicong Hong, Yiqun Mei, Chongjian Ge et al.

A truly interactive world model requires three key ingredients: real-time long-horizon streaming, consistent spatial memory, and precise user control. However, most existing approaches address only one of these aspects in isolation, as achieving all three simultaneously is highly challenging-for example, long-term memory mechanisms often degrade real-time performance. In this work, we present RELIC, a unified framework that tackles these three challenges altogether. Given a single image and a text description, RELIC enables memory-aware, long-duration exploration of arbitrary scenes in real time. Built upon recent autoregressive video-diffusion distillation techniques, our model represents long-horizon memory using highly compressed historical latent tokens encoded with both relative actions and absolute camera poses within the KV cache. This compact, camera-aware memory structure supports implicit 3D-consistent content retrieval and enforces long-term coherence with minimal computational overhead. In parallel, we fine-tune a bidirectional teacher video model to generate sequences beyond its original 5-second training horizon, and transform it into a causal student generator using a new memory-efficient self-forcing paradigm that enables full-context distillation over long-duration teacher as well as long student self-rollouts. Implemented as a 14B-parameter model and trained on a curated Unreal Engine-rendered dataset, RELIC achieves real-time generation at 16 FPS while demonstrating more accurate action following, more stable long-horizon streaming, and more robust spatial-memory retrieval compared with prior work. These capabilities establish RELIC as a strong foundation for the next generation of interactive world modeling.

Habib Slim, Shariq Farooq Bhat, Mohamed Elhoseiny et al.

Creating and editing high-quality 3D content remains a central challenge in computer graphics. We address this challenge by introducing CompoSE, a novel method for Compositional Synthesis and Editing of 3D shapes via part-aware control. Our method takes as input a set of coarse geometric primitives (e.g., bounding boxes) that represent distinct object parts arranged in a particular spatial configuration, and synthesizes as output part-separated 3D objects that support localized granular (i.e., compositional) editing of individual parts. The key insight that enables our method is our use of a diffusion transformer architecture that alternates between processing each part locally and aggregating contextual information across parts globally, and features a novel conditioning technique that ensures strong adherence to the user's input. Importantly, our method learns to infer part semantics and symmetries directly from the user's coarse layout guidance, and does not require part-level text prompts. We demonstrate that our method enables powerful part-level editing capabilities, including context-aware substitution, addition, deletion, and style-preserving resizing operations. We show through extensive experiments that our method significantly outperforms existing approaches on guided synthesis, as measured by objective metrics and LLM-based evaluations.

Mike Roberts, Jason Ramapuram, Anurag Ranjan et al.

For many fundamental scene understanding tasks, it is difficult or impossible to obtain per-pixel ground truth labels from real images. We address this challenge by introducing Hypersim, a photorealistic synthetic dataset for holistic indoor scene understanding. To create our dataset, we leverage a large repository of synthetic scenes created by professional artists, and we generate 77,400 images of 461 indoor scenes with detailed per-pixel labels and corresponding ground truth geometry. Our dataset: (1) relies exclusively on publicly available 3D assets; (2) includes complete scene geometry, material information, and lighting information for every scene; (3) includes dense per-pixel semantic instance segmentations and complete camera information for every image; and (4) factors every image into diffuse reflectance, diffuse illumination, and a non-diffuse residual term that captures view-dependent lighting effects. We analyze our dataset at the level of scenes, objects, and pixels, and we analyze costs in terms of money, computation time, and annotation effort. Remarkably, we find that it is possible to generate our entire dataset from scratch, for roughly half the cost of training a popular open-source natural language processing model. We also evaluate sim-to-real transfer performance on two real-world scene understanding tasks - semantic segmentation and 3D shape prediction - where we find that pre-training on our dataset significantly improves performance on both tasks, and achieves state-of-the-art performance on the most challenging Pix3D test set. All of our rendered image data, as well as all the code we used to generate our dataset and perform our experiments, is available online.

Chenyu Li, Oscar Michel, Xichen Pan et al.

Large-scale pre-trained video generation models excel in content creation but are not reliable as physically accurate world simulators out of the box. This work studies the process of post-training these models for accurate world modeling through the lens of the simple, yet fundamental, physics task of modeling object freefall. We show state-of-the-art video generation models struggle with this basic task, despite their visually impressive outputs. To remedy this problem, we find that fine-tuning on a relatively small amount of simulated videos is effective in inducing the dropping behavior in the model, and we can further improve results through a novel reward modeling procedure we introduce. Our study also reveals key limitations of post-training in generalization and distribution modeling. Additionally, we release a benchmark for this task that may serve as a useful diagnostic tool for tracking physical accuracy in large-scale video generative model development.

Xiaodan Xing, Fadong Shi, Jiahao Huang et al.

Generative Artificial Intelligence (AI) technologies and large models are producing realistic outputs across various domains, such as images, text, speech, and music. Creating these advanced generative models requires significant resources, particularly large and high-quality datasets. To minimise training expenses, many algorithm developers use data created by the models themselves as a cost-effective training solution. However, not all synthetic data effectively improve model performance, necessitating a strategic balance in the use of real versus synthetic data to optimise outcomes. Currently, the previously well-controlled integration of real and synthetic data is becoming uncontrollable. The widespread and unregulated dissemination of synthetic data online leads to the contamination of datasets traditionally compiled through web scraping, now mixed with unlabeled synthetic data. This trend, known as the AI autophagy phenomenon, suggests a future where generative AI systems may increasingly consume their own outputs without discernment, raising concerns about model performance, reliability, and ethical implications. What will happen if generative AI continuously consumes itself without discernment? What measures can we take to mitigate the potential adverse effects? To address these research questions, this study examines the existing literature, delving into the consequences of AI autophagy, analyzing the associated risks, and exploring strategies to mitigate its impact. Our aim is to provide a comprehensive perspective on this phenomenon advocating for a balanced approach that promotes the sustainable development of generative AI technologies in the era of large models.

Mike Roberts, Debadeepta Dey, Anh Truong et al.

Drones equipped with cameras are emerging as a powerful tool for large-scale aerial 3D scanning, but existing automatic flight planners do not exploit all available information about the scene, and can therefore produce inaccurate and incomplete 3D models. We present an automatic method to generate drone trajectories, such that the imagery acquired during the flight will later produce a high-fidelity 3D model. Our method uses a coarse estimate of the scene geometry to plan camera trajectories that: (1) cover the scene as thoroughly as possible; (2) encourage observations of scene geometry from a diverse set of viewing angles; (3) avoid obstacles; and (4) respect a user-specified flight time budget. Our method relies on a mathematical model of scene coverage that exhibits an intuitive diminishing returns property known as submodularity. We leverage this property extensively to design a trajectory planning algorithm that reasons globally about the non-additive coverage reward obtained across a trajectory, jointly with the cost of traveling between views. We evaluate our method by using it to scan three large outdoor scenes, and we perform a quantitative evaluation using a photorealistic video game simulator.

Niels Joubert, Jane L. E, Dan B Goldman et al.

We present a system to capture video footage of human subjects in the real world. Our system leverages a quadrotor camera to automatically capture well-composed video of two subjects. Subjects are tracked in a large-scale outdoor environment using RTK GPS and IMU sensors. Then, given the tracked state of our subjects, our system automatically computes static shots based on well-established visual composition principles and canonical shots from cinematography literature. To transition between these static shots, we calculate feasible, safe, and visually pleasing transitions using a novel real-time trajectory planning algorithm. We evaluate the performance of our tracking system, and experimentally show that RTK GPS significantly outperforms conventional GPS in capturing a variety of canonical shots. Lastly, we demonstrate our system guiding a consumer quadrotor camera autonomously capturing footage of two subjects in a variety of use cases. This is the first end-to-end system that enables people to leverage the mobility of quadrotors, as well as the knowledge of expert filmmakers, to autonomously capture high-quality footage of people in the real world.

Verena Kaynig, Amelio Vazquez-Reina, Seymour Knowles-Barley et al.

Automated sample preparation and electron microscopy enables acquisition of very large image data sets. These technical advances are of special importance to the field of neuroanatomy, as 3D reconstructions of neuronal processes at the nm scale can provide new insight into the fine grained structure of the brain. Segmentation of large-scale electron microscopy data is the main bottleneck in the analysis of these data sets. In this paper we present a pipeline that provides state-of-the art reconstruction performance while scaling to data sets in the GB-TB range. First, we train a random forest classifier on interactive sparse user annotations. The classifier output is combined with an anisotropic smoothing prior in a Conditional Random Field framework to generate multiple segmentation hypotheses per image. These segmentations are then combined into geometrically consistent 3D objects by segmentation fusion. We provide qualitative and quantitative evaluation of the automatic segmentation and demonstrate large-scale 3D reconstructions of neuronal processes from a $\mathbf{27,000}$ $\mathbf{μm^3}$ volume of brain tissue over a cube of $\mathbf{30 \; μm}$ in each dimension corresponding to 1000 consecutive image sections. We also introduce Mojo, a proofreading tool including semi-automated correction of merge errors based on sparse user scribbles.

Justin Thaler, Mike Roberts, Michael Mitzenmacher et al.

As the cloud computing paradigm has gained prominence, the need for verifiable computation has grown increasingly urgent. The concept of verifiable computation enables a weak client to outsource difficult computations to a powerful, but untrusted, server. Protocols for verifiable computation aim to provide the client with a guarantee that the server performed the requested computations correctly, without requiring the client to perform the computations herself. By design, these protocols impose a minimal computational burden on the client. However, existing protocols require the server to perform a large amount of extra bookkeeping in order to enable a client to easily verify the results. Verifiable computation has thus remained a theoretical curiosity, and protocols for it have not been implemented in real cloud computing systems. Our goal is to leverage GPUs to reduce the server-side slowdown for verifiable computation. To this end, we identify abundant data parallelism in a state-of-the-art general-purpose protocol for verifiable computation, originally due to Goldwasser, Kalai, and Rothblum, and recently extended by Cormode, Mitzenmacher, and Thaler. We implement this protocol on the GPU, obtaining 40-120x server-side speedups relative to a state-of-the-art sequential implementation. For benchmark problems, our implementation reduces the slowdown of the server to factors of 100-500x relative to the original computations requested by the client. Furthermore, we reduce the already small runtime of the client by 100x. Similarly, we obtain 20-50x server-side and client-side speedups for related protocols targeted at specific streaming problems. We believe our results demonstrate the immediate practicality of using GPUs for verifiable computation, and more generally that protocols for verifiable computation have become sufficiently mature to deploy in real cloud computing systems.