Klemen Kotar

Aaron Walsman, Muru Zhang, Klemen Kotar et al.

Visual understanding of geometric structures with complex spatial relationships is a fundamental component of human intelligence. As children, we learn how to reason about structure not only from observation, but also by interacting with the world around us -- by taking things apart and putting them back together again. The ability to reason about structure and compositionality allows us to not only build things, but also understand and reverse-engineer complex systems. In order to advance research in interactive reasoning for part-based geometric understanding, we propose a challenging new assembly problem using LEGO bricks that we call Break and Make. In this problem an agent is given a LEGO model and attempts to understand its structure by interactively inspecting and disassembling it. After this inspection period, the agent must then prove its understanding by rebuilding the model from scratch using low-level action primitives. In order to facilitate research on this problem we have built LTRON, a fully interactive 3D simulator that allows learning agents to assemble, disassemble and manipulate LEGO models. We pair this simulator with a new dataset of fan-made LEGO creations that have been uploaded to the internet in order to provide complex scenes containing over a thousand unique brick shapes. We take a first step towards solving this problem using sequence-to-sequence models that provide guidance for how to make progress on this challenging problem. Our simulator and data are available at github.com/aaronwalsman/ltron. Additional training code and PyTorch examples are available at github.com/aaronwalsman/ltron-torch-eccv22.

Klemen Kotar, Stephen Tian, Hong-Xing Yu et al. · stanford

The human visual system can effortlessly recognize an object under different extrinsic factors such as lighting, object poses, and background, yet current computer vision systems often struggle with these variations. An important step to understanding and improving artificial vision systems is to measure image similarity purely based on intrinsic object properties that define object identity. This problem has been studied in the computer vision literature as re-identification, though mostly restricted to specific object categories such as people and cars. We propose to extend it to general object categories, exploring an image similarity metric based on object intrinsics. To benchmark such measurements, we collect the Common paired objects Under differenT Extrinsics (CUTE) dataset of $18,000$ images of $180$ objects under different extrinsic factors such as lighting, poses, and imaging conditions. While existing methods such as LPIPS and CLIP scores do not measure object intrinsics well, we find that combining deep features learned from contrastive self-supervised learning with foreground filtering is a simple yet effective approach to approximating the similarity. We conduct an extensive survey of pre-trained features and foreground extraction methods to arrive at a strong baseline that best measures intrinsic object-centric image similarity among current methods. Finally, we demonstrate that our approach can aid in downstream applications such as acting as an analog for human subjects and improving generalizable re-identification. Please see our project website at https://s-tian.github.io/projects/cute/ for visualizations of the data and demos of our metric.

Rahul Venkatesh, Klemen Kotar, Lilian Naing Chen et al.

A central challenge in visual intelligence is learning the physical structure of scenes from raw videos: how regions form objects and the laws that govern their interactions. Solving these tasks requires world models capable of inferring distributional states of the world from partial observations - capabilities that current architectures do not provide. We introduce a new class of probabilistic world models that support estimation of the probability of any visual variable, such as appearance and dynamics, conditioned on any other variables. Here, we identify that these models can be trained efficiently with autoregressive sequence modeling, yielding world models from which rich object understanding emerges. First, we demonstrate that our model captures the physical laws governing how objects move by generating multiple plausible future states of the world through sequential inference. Then, by analyzing motion correlations across these futures, we extract objects and articulated object subparts. Having discovered these objects, we show that our world model can manipulate them in 3D. Finally, we demonstrate how physical relationships between objects can be computed from the world model, enabling applications such as Visual Jenga.

Klemen Kotar, Aaron Walsman, Roozbeh Mottaghi

We propose Embodied Navigation Trajectory Learner (ENTL), a method for extracting long sequence representations for embodied navigation. Our approach unifies world modeling, localization and imitation learning into a single sequence prediction task. We train our model using vector-quantized predictions of future states conditioned on current states and actions. ENTL's generic architecture enables sharing of the spatio-temporal sequence encoder for multiple challenging embodied tasks. We achieve competitive performance on navigation tasks using significantly less data than strong baselines while performing auxiliary tasks such as localization and future frame prediction (a proxy for world modeling). A key property of our approach is that the model is pre-trained without any explicit reward signal, which makes the resulting model generalizable to multiple tasks and environments.

Daniel M. Bear, Kevin Feigelis, Honglin Chen et al.

Leading approaches in machine vision employ different architectures for different tasks, trained on costly task-specific labeled datasets. This complexity has held back progress in areas, such as robotics, where robust task-general perception remains a bottleneck. In contrast, "foundation models" of natural language have shown how large pre-trained neural networks can provide zero-shot solutions to a broad spectrum of apparently distinct tasks. Here we introduce Counterfactual World Modeling (CWM), a framework for constructing a visual foundation model: a unified, unsupervised network that can be prompted to perform a wide variety of visual computations. CWM has two key components, which resolve the core issues that have hindered application of the foundation model concept to vision. The first is structured masking, a generalization of masked prediction methods that encourages a prediction model to capture the low-dimensional structure in visual data. The model thereby factors the key physical components of a scene and exposes an interface to them via small sets of visual tokens. This in turn enables CWM's second main idea -- counterfactual prompting -- the observation that many apparently distinct visual representations can be computed, in a zero-shot manner, by comparing the prediction model's output on real inputs versus slightly modified ("counterfactual") inputs. We show that CWM generates high-quality readouts on real-world images and videos for a diversity of tasks, including estimation of keypoints, optical flow, occlusions, object segments, and relative depth. Taken together, our results show that CWM is a promising path to unifying the manifold strands of machine vision in a conceptually simple foundation.

Wanhee Lee, Klemen Kotar, Rahul Mysore Venkatesh et al.

Understanding 3D scenes requires flexible combinations of visual reasoning tasks, including depth estimation, novel view synthesis, and object manipulation, all of which are essential for perception and interaction. Existing approaches have typically addressed these tasks in isolation, preventing them from sharing a common representation or transferring knowledge across tasks. A conceptually simpler but practically non-trivial alternative is to unify these diverse tasks into a single model, reducing different tasks from separate training objectives to merely different prompts and allowing for joint training across all datasets. In this work, we present a physical world model for unified 3D understanding and interaction (3WM), formulated as a probabilistic graphical model in which nodes represent multimodal scene elements such as RGB, optical flow, and camera pose. Diverse tasks emerge from different inference pathways through the graph: novel view synthesis from RGB and dense flow prompts, object manipulation from RGB and sparse flow prompts, and depth estimation from RGB and camera conditioning, all zero-shot without task-specific training. 3WM outperforms specialized baselines without the need for finetuning by offering precise controllability, strong geometric consistency, and robustness in real-world scenarios, achieving state-of-the-art performance on NVS and 3D object manipulation. Beyond predefined tasks, the model supports composable inference pathways, such as moving objects aside while navigating a 3D environment, enabling complex geometric reasoning. This demonstrates that a unified model can serve as a practical alternative to fragmented task-specific systems, taking a step towards a general-purpose visual world model.

Khai Loong Aw, Klemen Kotar, Wanhee Lee et al.

Young children demonstrate early abilities to understand their physical world, estimating depth, motion, object coherence, interactions, and many other aspects of physical scene understanding. Children are both data-efficient and flexible cognitive systems, creating competence despite extremely limited training data, while generalizing to myriad untrained tasks -- a major challenge even for today's best AI systems. Here we introduce a novel computational hypothesis for these abilities, the Zero-shot Visual World Model (ZWM). ZWM is based on three principles: a sparse temporally-factored predictor that decouples appearance from dynamics; zero-shot estimation through approximate causal inference; and composition of inferences to build more complex abilities. We show that ZWM can be learned from the first-person experience of a single child, rapidly generating competence across multiple physical understanding benchmarks. It also broadly recapitulates behavioral signatures of child development and builds brain-like internal representations. Our work presents a blueprint for efficient and flexible learning from human-scale data, advancing both a computational account for children's early physical understanding and a path toward data-efficient AI systems.

Klemen Kotar, Roozbeh Mottaghi

Over the years various methods have been proposed for the problem of object detection. Recently, we have witnessed great strides in this domain owing to the emergence of powerful deep neural networks. However, there are typically two main assumptions common among these approaches. First, the model is trained on a fixed training set and is evaluated on a pre-recorded test set. Second, the model is kept frozen after the training phase, so no further updates are performed after the training is finished. These two assumptions limit the applicability of these methods to real-world settings. In this paper, we propose Interactron, a method for adaptive object detection in an interactive setting, where the goal is to perform object detection in images observed by an embodied agent navigating in different environments. Our idea is to continue training during inference and adapt the model at test time without any explicit supervision via interacting with the environment. Our adaptive object detection model provides a 7.2 point improvement in AP (and 12.7 points in AP50) over DETR, a recent, high-performance object detector. Moreover, we show that our object detection model adapts to environments with completely different appearance characteristics, and performs well in them. The code is available at: https://github.com/allenai/interactron .

Klemen Kotar, Gabriel Ilharco, Ludwig Schmidt et al.

In the past few years, we have witnessed remarkable breakthroughs in self-supervised representation learning. Despite the success and adoption of representations learned through this paradigm, much is yet to be understood about how different training methods and datasets influence performance on downstream tasks. In this paper, we analyze contrastive approaches as one of the most successful and popular variants of self-supervised representation learning. We perform this analysis from the perspective of the training algorithms, pre-training datasets and end tasks. We examine over 700 training experiments including 30 encoders, 4 pre-training datasets and 20 diverse downstream tasks. Our experiments address various questions regarding the performance of self-supervised models compared to their supervised counterparts, current benchmarks used for evaluation, and the effect of the pre-training data on end task performance. Our Visual Representation Benchmark (ViRB) is available at: https://github.com/allenai/virb.

Rahul Venkatesh, Honglin Chen, Kevin Feigelis et al.

The ability to understand physical dynamics is critical for agents to act in the world. Here, we use Counterfactual World Modeling (CWM) to extract vision structures for dynamics understanding. CWM uses a temporally-factored masking policy for masked prediction of video data without annotations. This policy enables highly effective "counterfactual prompting" of the predictor, allowing a spectrum of visual structures to be extracted from a single pre-trained predictor without finetuning on annotated datasets. We demonstrate that these structures are useful for physical dynamics understanding, allowing CWM to achieve the state-of-the-art performance on the Physion benchmark.

Lukas Wolf, Greta Tuckute, Klemen Kotar et al.

Training on multiple modalities of input can augment the capabilities of a language model. Here, we ask whether such a training regime can improve the quality and efficiency of these systems as well. We focus on text--audio and introduce Whisbert, which is inspired by the text--image approach of FLAVA (Singh et al., 2022). In accordance with Babylm guidelines (Warstadt et al., 2023), we pretrain Whisbert on a dataset comprising only 100 million words plus their corresponding speech from the word-aligned version of the People's Speech dataset (Galvez et al., 2021). To assess the impact of multimodality, we compare versions of the model that are trained on text only and on both audio and text simultaneously. We find that while Whisbert is able to perform well on multimodal masked modeling and surpasses the Babylm baselines in most benchmark tasks, it struggles to optimize its complex objective and outperform its text-only Whisbert baseline.

Wanhee Lee, Klemen Kotar, Rahul Mysore Venkatesh et al.

3D scene understanding from single images is a pivotal problem in computer vision with numerous downstream applications in graphics, augmented reality, and robotics. While diffusion-based modeling approaches have shown promise, they often struggle to maintain object and scene consistency, especially in complex real-world scenarios. To address these limitations, we propose an autoregressive generative approach called Local Random Access Sequence (LRAS) modeling, which uses local patch quantization and randomly ordered sequence generation. By utilizing optical flow as an intermediate representation for 3D scene editing, our experiments demonstrate that LRAS achieves state-of-the-art novel view synthesis and 3D object manipulation capabilities. Furthermore, we show that our framework naturally extends to self-supervised depth estimation through a simple modification of the sequence design. By achieving strong performance on multiple 3D scene understanding tasks, LRAS provides a unified and effective framework for building the next generation of 3D vision models.

Klemen Kotar, Wanhee Lee, Rahul Venkatesh et al.

We present Probabilistic Structure Integration (PSI), a system for learning richly controllable and flexibly promptable world models from data. PSI consists of a three-step cycle. The first step, Probabilistic prediction, involves building a probabilistic graphical model Psi of the data, in the form of a random-access autoregressive sequence model. Psi supports a complete set of learned conditional distributions describing the dependence of any variables in the data on any other set of variables. In step 2, Structure extraction, we show how to extract underlying low-dimensional properties in the data, corresponding to a diverse set of meaningful "intermediate structures", in a zero-shot fashion via causal inference on Psi. Step 3, Integration, completes the cycle by converting these structures into new token types that are then continually mixed back into the training diet as conditioning signals and prediction targets. Each such cycle augments the capabilities of Psi, both allowing it to model the underlying data better, and creating new control handles -- akin to an LLM-like universal prompting language. We train an instance of Psi on 1.4 trillion tokens of internet video data; we use it to perform a variety of useful video prediction and understanding inferences; we extract state-of-the-art optical flow, self-supervised depth and object segmentation; and we use these structures to support a full cycle of predictive improvements.

Greta Tuckute, Klemen Kotar, Evelina Fedorenko et al.

We introduce AuriStream, a biologically inspired model for encoding speech via a two-stage framework inspired by the human auditory processing hierarchy. The first stage transforms raw audio into a time-frequency representation based on the human cochlea, from which we extract discrete \textbf{cochlear tokens}. The second stage applies an autoregressive sequence model over the cochlear tokens. AuriStream learns meaningful phoneme and word representations, and state-of-the-art lexical semantics. AuriStream shows competitive performance on diverse downstream SUPERB speech tasks. Complementing AuriStream's strong representational capabilities, it generates continuations of audio which can be visualized in a spectrogram space and decoded back into audio, providing insights into the model's predictions. In summary, we present a two-stage framework for speech representation learning to advance the development of more human-like models that efficiently handle a range of speech-based tasks.

Rahul Venkatesh, Klemen Kotar, Lilian Naing Chen et al.

Segments in computer vision are often defined by semantic considerations and are highly dependent on category-specific conventions. In contrast, developmental psychology suggests that humans perceive the world in terms of Spelke objects--groupings of physical things that reliably move together when acted on by physical forces. Spelke objects thus operate on category-agnostic causal motion relationships which potentially better support tasks like manipulation and planning. In this paper, we first benchmark the Spelke object concept, introducing the SpelkeBench dataset that contains a wide variety of well-defined Spelke segments in natural images. Next, to extract Spelke segments from images algorithmically, we build SpelkeNet, a class of visual world models trained to predict distributions over future motions. SpelkeNet supports estimation of two key concepts for Spelke object discovery: (1) the motion affordance map, identifying regions likely to move under a poke, and (2) the expected-displacement map, capturing how the rest of the scene will move. These concepts are used for "statistical counterfactual probing", where diverse "virtual pokes" are applied on regions of high motion-affordance, and the resultant expected displacement maps are used define Spelke segments as statistical aggregates of correlated motion statistics. We find that SpelkeNet outperforms supervised baselines like SegmentAnything (SAM) on SpelkeBench. Finally, we show that the Spelke concept is practically useful for downstream applications, yielding superior performance on the 3DEditBench benchmark for physical object manipulation when used in a variety of off-the-shelf object manipulation models.

Klemen Kotar, Greta Tuckute

Biological neural networks are shaped both by evolution across generations and by individual learning within an organism's lifetime, whereas standard artificial neural networks undergo a single, large training procedure without inherited constraints. In this preliminary work, we propose a framework that incorporates this crucial generational dimension - an "outer loop" of evolution that shapes the "inner loop" of learning - so that artificial networks better mirror the effects of evolution and individual learning in biological organisms. Focusing on language, we train a model that inherits a "model connectome" from the outer evolution loop before exposing it to a developmental-scale corpus of 100M tokens. Compared with two closely matched control models, we show that the connectome model performs better or on par on natural language processing tasks as well as alignment to human behavior and brain data. These findings suggest that a model connectome serves as an efficient prior for learning in low-data regimes - narrowing the gap between single-generation artificial models and biologically evolved neural networks.

Luca Weihs, Jordi Salvador, Klemen Kotar et al.

The domain of Embodied AI, in which agents learn to complete tasks through interaction with their environment from egocentric observations, has experienced substantial growth with the advent of deep reinforcement learning and increased interest from the computer vision, NLP, and robotics communities. This growth has been facilitated by the creation of a large number of simulated environments (such as AI2-THOR, Habitat and CARLA), tasks (like point navigation, instruction following, and embodied question answering), and associated leaderboards. While this diversity has been beneficial and organic, it has also fragmented the community: a huge amount of effort is required to do something as simple as taking a model trained in one environment and testing it in another. This discourages good science. We introduce AllenAct, a modular and flexible learning framework designed with a focus on the unique requirements of Embodied AI research. AllenAct provides first-class support for a growing collection of embodied environments, tasks and algorithms, provides reproductions of state-of-the-art models and includes extensive documentation, tutorials, start-up code, and pre-trained models. We hope that our framework makes Embodied AI more accessible and encourages new researchers to join this exciting area. The framework can be accessed at: https://allenact.org/