C. Lawrence Zitnick

Richard Tran, Janice Lan, Muhammed Shuaibi et al. · baidu, cmu

The development of machine learning models for electrocatalysts requires a broad set of training data to enable their use across a wide variety of materials. One class of materials that currently lacks sufficient training data is oxides, which are critical for the development of OER catalysts. To address this, we developed the OC22 dataset, consisting of 62,331 DFT relaxations (~9,854,504 single point calculations) across a range of oxide materials, coverages, and adsorbates. We define generalized total energy tasks that enable property prediction beyond adsorption energies; we test baseline performance of several graph neural networks; and we provide pre-defined dataset splits to establish clear benchmarks for future efforts. In the most general task, GemNet-OC sees a ~36% improvement in energy predictions when combining the chemically dissimilar OC20 and OC22 datasets via fine-tuning. Similarly, we achieved a ~19% improvement in total energy predictions on OC20 and a ~9% improvement in force predictions in OC22 when using joint training. We demonstrate the practical utility of a top performing model by capturing literature adsorption energies and important OER scaling relationships. We expect OC22 to provide an important benchmark for models seeking to incorporate intricate long-range electrostatic and magnetic interactions in oxide surfaces. Dataset and baseline models are open sourced, and a public leaderboard is available to encourage continued community developments on the total energy tasks and data.

Johannes Gasteiger, Muhammed Shuaibi, Anuroop Sriram et al. · baidu, cmu

Recent years have seen the advent of molecular simulation datasets that are orders of magnitude larger and more diverse. These new datasets differ substantially in four aspects of complexity: 1. Chemical diversity (number of different elements), 2. system size (number of atoms per sample), 3. dataset size (number of data samples), and 4. domain shift (similarity of the training and test set). Despite these large differences, benchmarks on small and narrow datasets remain the predominant method of demonstrating progress in graph neural networks (GNNs) for molecular simulation, likely due to cheaper training compute requirements. This raises the question -- does GNN progress on small and narrow datasets translate to these more complex datasets? This work investigates this question by first developing the GemNet-OC model based on the large Open Catalyst 2020 (OC20) dataset. GemNet-OC outperforms the previous state-of-the-art on OC20 by 16% while reducing training time by a factor of 10. We then compare the impact of 18 model components and hyperparameter choices on performance in multiple datasets. We find that the resulting model would be drastically different depending on the dataset used for making model choices. To isolate the source of this discrepancy we study six subsets of the OC20 dataset that individually test each of the above-mentioned four dataset aspects. We find that results on the OC-2M subset correlate well with the full OC20 dataset while being substantially cheaper to train on. Our findings challenge the common practice of developing GNNs solely on small datasets, but highlight ways of achieving fast development cycles and generalizable results via moderately-sized, representative datasets such as OC-2M and efficient models such as GemNet-OC. Our code and pretrained model weights are open-sourced.

Janice Lan, Aini Palizhati, Muhammed Shuaibi et al. · meta-ai

Computational catalysis is playing an increasingly significant role in the design of catalysts across a wide range of applications. A common task for many computational methods is the need to accurately compute the adsorption energy for an adsorbate and a catalyst surface of interest. Traditionally, the identification of low energy adsorbate-surface configurations relies on heuristic methods and researcher intuition. As the desire to perform high-throughput screening increases, it becomes challenging to use heuristics and intuition alone. In this paper, we demonstrate machine learning potentials can be leveraged to identify low energy adsorbate-surface configurations more accurately and efficiently. Our algorithm provides a spectrum of trade-offs between accuracy and efficiency, with one balanced option finding the lowest energy configuration 87.36% of the time, while achieving a 2000x speedup in computation. To standardize benchmarking, we introduce the Open Catalyst Dense dataset containing nearly 1,000 diverse surfaces and 100,000 unique configurations.

C. Lawrence Zitnick, Abhishek Das, Adeesh Kolluru et al. · baidu, cmu

Modeling the energy and forces of atomic systems is a fundamental problem in computational chemistry with the potential to help address many of the world's most pressing problems, including those related to energy scarcity and climate change. These calculations are traditionally performed using Density Functional Theory, which is computationally very expensive. Machine learning has the potential to dramatically improve the efficiency of these calculations from days or hours to seconds. We propose the Spherical Channel Network (SCN) to model atomic energies and forces. The SCN is a graph neural network where nodes represent atoms and edges their neighboring atoms. The atom embeddings are a set of spherical functions, called spherical channels, represented using spherical harmonics. We demonstrate, that by rotating the embeddings based on the 3D edge orientation, more information may be utilized while maintaining the rotational equivariance of the messages. While equivariance is a desirable property, we find that by relaxing this constraint in both message passing and aggregation, improved accuracy may be achieved. We demonstrate state-of-the-art results on the large-scale Open Catalyst dataset in both energy and force prediction for numerous tasks and metrics.

Anuroop Sriram, Abhishek Das, Brandon M. Wood et al. · baidu, cmu

Recent progress in Graph Neural Networks (GNNs) for modeling atomic simulations has the potential to revolutionize catalyst discovery, which is a key step in making progress towards the energy breakthroughs needed to combat climate change. However, the GNNs that have proven most effective for this task are memory intensive as they model higher-order interactions in the graphs such as those between triplets or quadruplets of atoms, making it challenging to scale these models. In this paper, we introduce Graph Parallelism, a method to distribute input graphs across multiple GPUs, enabling us to train very large GNNs with hundreds of millions or billions of parameters. We empirically evaluate our method by scaling up the number of parameters of the recently proposed DimeNet++ and GemNet models by over an order of magnitude. On the large-scale Open Catalyst 2020 (OC20) dataset, these graph-parallelized models lead to relative improvements of 1) 15% on the force MAE metric for the S2EF task and 2) 21% on the AFbT metric for the IS2RS task, establishing new state-of-the-art results.

Saro Passaro, C. Lawrence Zitnick

Graph neural networks that model 3D data, such as point clouds or atoms, are typically desired to be $SO(3)$ equivariant, i.e., equivariant to 3D rotations. Unfortunately equivariant convolutions, which are a fundamental operation for equivariant networks, increase significantly in computational complexity as higher-order tensors are used. In this paper, we address this issue by reducing the $SO(3)$ convolutions or tensor products to mathematically equivalent convolutions in $SO(2)$ . This is accomplished by aligning the node embeddings' primary axis with the edge vectors, which sparsifies the tensor product and reduces the computational complexity from $O(L^6)$ to $O(L^3)$, where $L$ is the degree of the representation. We demonstrate the potential implications of this improvement by proposing the Equivariant Spherical Channel Network (eSCN), a graph neural network utilizing our novel approach to equivariant convolutions, which achieves state-of-the-art results on the large-scale OC-20 and OC-22 datasets.

Nima Shoghi, Adeesh Kolluru, John R. Kitchin et al.

Foundation models have been transformational in machine learning fields such as natural language processing and computer vision. Similar success in atomic property prediction has been limited due to the challenges of training effective models across multiple chemical domains. To address this, we introduce Joint Multi-domain Pre-training (JMP), a supervised pre-training strategy that simultaneously trains on multiple datasets from different chemical domains, treating each dataset as a unique pre-training task within a multi-task framework. Our combined training dataset consists of $\sim$120M systems from OC20, OC22, ANI-1x, and Transition-1x. We evaluate performance and generalization by fine-tuning over a diverse set of downstream tasks and datasets including: QM9, rMD17, MatBench, QMOF, SPICE, and MD22. JMP demonstrates an average improvement of 59% over training from scratch, and matches or sets state-of-the-art on 34 out of 40 tasks. Our work highlights the potential of pre-training strategies that utilize diverse data to advance property prediction across chemical domains, especially for low-data tasks. Please visit https://nima.sh/jmp for further information.

Vahe Gharakhanyan, Yi Yang, Luis Barroso-Luque et al. · baidu, cmu

Crystal Structure Prediction (CSP) of molecular crystals plays a central role in applications, such as pharmaceuticals and organic electronics. CSP is challenging and computationally expensive due to the need to explore a large search space with sufficient accuracy to capture energy differences of a few kJ/mol between polymorphs. Dispersion-inclusive density functional theory (DFT) provides the required accuracy but its computational cost is impractical for a large number of putative structures. We introduce FastCSP, an open-source, high-throughput CSP workflow based on machine learning interatomic potentials (MLIPs). FastCSP combines random structure generation using Genarris 3.0 with geometry relaxation and free energy calculations powered entirely by the Universal Model for Atoms (UMA) MLIP. We benchmark FastCSP on a curated set of 28 mostly rigid molecules, demonstrating that our workflow consistently generates known experimental structures and ranks them within 5 kJ/mol per molecule of the global minimum. Our results demonstrate that universal MLIPs can be used across diverse compounds without requiring system-specific tuning. Moreover, the speed and accuracy afforded by UMA eliminate the need for classical force fields in the early stages of CSP and for final re-ranking with DFT. The open-source release of the entire FastCSP workflow significantly lowers the barrier to accessing CSP. CSP results for a single system can be obtained within hours on tens of modern GPUs, making high-throughput crystal structure prediction feasible for a broad range of scientific applications.

Drew A. Hudson, C. Lawrence Zitnick

We introduce the GANformer2 model, an iterative object-oriented transformer, explored for the task of generative modeling. The network incorporates strong and explicit structural priors, to reflect the compositional nature of visual scenes, and synthesizes images through a sequential process. It operates in two stages: a fast and lightweight planning phase, where we draft a high-level scene layout, followed by an attention-based execution phase, where the layout is being refined, evolving into a rich and detailed picture. Our model moves away from conventional black-box GAN architectures that feature a flat and monolithic latent space towards a transparent design that encourages efficiency, controllability and interpretability. We demonstrate GANformer2's strengths and qualities through a careful evaluation over a range of datasets, from multi-object CLEVR scenes to the challenging COCO images, showing it successfully achieves state-of-the-art performance in terms of visual quality, diversity and consistency. Further experiments demonstrate the model's disentanglement and provide a deeper insight into its generative process, as it proceeds step-by-step from a rough initial sketch, to a detailed layout that accounts for objects' depths and dependencies, and up to the final high-resolution depiction of vibrant and intricate real-world scenes. See https://github.com/dorarad/gansformer for model implementation.

Drew A. Hudson, C. Lawrence Zitnick

We introduce the GANformer, a novel and efficient type of transformer, and explore it for the task of visual generative modeling. The network employs a bipartite structure that enables long-range interactions across the image, while maintaining computation of linear efficiency, that can readily scale to high-resolution synthesis. It iteratively propagates information from a set of latent variables to the evolving visual features and vice versa, to support the refinement of each in light of the other and encourage the emergence of compositional representations of objects and scenes. In contrast to the classic transformer architecture, it utilizes multiplicative integration that allows flexible region-based modulation, and can thus be seen as a generalization of the successful StyleGAN network. We demonstrate the model's strength and robustness through a careful evaluation over a range of datasets, from simulated multi-object environments to rich real-world indoor and outdoor scenes, showing it achieves state-of-the-art results in terms of image quality and diversity, while enjoying fast learning and better data-efficiency. Further qualitative and quantitative experiments offer us an insight into the model's inner workings, revealing improved interpretability and stronger disentanglement, and illustrating the benefits and efficacy of our approach. An implementation of the model is available at https://github.com/dorarad/gansformer.

Songwei Ge, Vedanuj Goswami, C. Lawrence Zitnick et al.

Sketching or doodling is a popular creative activity that people engage in. However, most existing work in automatic sketch understanding or generation has focused on sketches that are quite mundane. In this work, we introduce two datasets of creative sketches -- Creative Birds and Creative Creatures -- containing 10k sketches each along with part annotations. We propose DoodlerGAN -- a part-based Generative Adversarial Network (GAN) -- to generate unseen compositions of novel part appearances. Quantitative evaluations as well as human studies demonstrate that sketches generated by our approach are more creative and of higher quality than existing approaches. In fact, in Creative Birds, subjects prefer sketches generated by DoodlerGAN over those drawn by humans! Our code can be found at https://github.com/facebookresearch/DoodlerGAN and a demo can be found at http://doodlergan.cloudcv.org.

Yuandong Tian, Jerry Ma, Qucheng Gong et al.

The AlphaGo, AlphaGo Zero, and AlphaZero series of algorithms are remarkable demonstrations of deep reinforcement learning's capabilities, achieving superhuman performance in the complex game of Go with progressively increasing autonomy. However, many obstacles remain in the understanding of and usability of these promising approaches by the research community. Toward elucidating unresolved mysteries and facilitating future research, we propose ELF OpenGo, an open-source reimplementation of the AlphaZero algorithm. ELF OpenGo is the first open-source Go AI to convincingly demonstrate superhuman performance with a perfect (20:0) record against global top professionals. We apply ELF OpenGo to conduct extensive ablation studies, and to identify and analyze numerous interesting phenomena in both the model training and in the gameplay inference procedures. Our code, models, selfplay datasets, and auxiliary data are publicly available at https://ai.facebook.com/tools/elf-opengo/.

Yuandong Tian, Qucheng Gong, Wenling Shang et al.

In this paper, we propose ELF, an Extensive, Lightweight and Flexible platform for fundamental reinforcement learning research. Using ELF, we implement a highly customizable real-time strategy (RTS) engine with three game environments (Mini-RTS, Capture the Flag and Tower Defense). Mini-RTS, as a miniature version of StarCraft, captures key game dynamics and runs at 40K frame-per-second (FPS) per core on a Macbook Pro notebook. When coupled with modern reinforcement learning methods, the system can train a full-game bot against built-in AIs end-to-end in one day with 6 CPUs and 1 GPU. In addition, our platform is flexible in terms of environment-agent communication topologies, choices of RL methods, changes in game parameters, and can host existing C/C++-based game environments like Arcade Learning Environment. Using ELF, we thoroughly explore training parameters and show that a network with Leaky ReLU and Batch Normalization coupled with long-horizon training and progressive curriculum beats the rule-based built-in AI more than $70\%$ of the time in the full game of Mini-RTS. Strong performance is also achieved on the other two games. In game replays, we show our agents learn interesting strategies. ELF, along with its RL platform, is open-sourced at https://github.com/facebookresearch/ELF.

Nate Gruver, Anuroop Sriram, Andrea Madotto et al. · baidu, cmu

We propose fine-tuning large language models for generation of stable materials. While unorthodox, fine-tuning large language models on text-encoded atomistic data is simple to implement yet reliable, with around 90% of sampled structures obeying physical constraints on atom positions and charges. Using energy above hull calculations from both learned ML potentials and gold-standard DFT calculations, we show that our strongest model (fine-tuned LLaMA-2 70B) can generate materials predicted to be metastable at about twice the rate (49% vs 28%) of CDVAE, a competing diffusion model. Because of text prompting's inherent flexibility, our models can simultaneously be used for unconditional generation of stable material, infilling of partial structures and text-conditional generation. Finally, we show that language models' ability to capture key symmetries of crystal structures improves with model scale, suggesting that the biases of pretrained LLMs are surprisingly well-suited for atomistic data.

Xiang Fu, Brandon M. Wood, Luis Barroso-Luque et al.

Machine learning interatomic potentials (MLIPs) have become increasingly effective at approximating quantum mechanical calculations at a fraction of the computational cost. However, lower errors on held out test sets do not always translate to improved results on downstream physical property prediction tasks. In this paper, we propose testing MLIPs on their practical ability to conserve energy during molecular dynamic simulations. If passed, improved correlations are found between test errors and their performance on physical property prediction tasks. We identify choices which may lead to models failing this test, and use these observations to improve upon highly-expressive models. The resulting model, eSEN, provides state-of-the-art results on a range of physical property prediction tasks, including materials stability prediction, thermal conductivity prediction, and phonon calculations.

Luis Barroso-Luque, Muhammed Shuaibi, Xiang Fu et al.

The ability to discover new materials with desirable properties is critical for numerous applications from helping mitigate climate change to advances in next generation computing hardware. AI has the potential to accelerate materials discovery and design by more effectively exploring the chemical space compared to other computational methods or by trial-and-error. While substantial progress has been made on AI for materials data, benchmarks, and models, a barrier that has emerged is the lack of publicly available training data and open pre-trained models. To address this, we present a Meta FAIR release of the Open Materials 2024 (OMat24) large-scale open dataset and an accompanying set of pre-trained models. OMat24 contains over 110 million density functional theory (DFT) calculations focused on structural and compositional diversity. Our EquiformerV2 models achieve state-of-the-art performance on the Matbench Discovery leaderboard and are capable of predicting ground-state stability and formation energies to an F1 score above 0.9 and an accuracy of 20 meV/atom, respectively. We explore the impact of model size, auxiliary denoising objectives, and fine-tuning on performance across a range of datasets including OMat24, MPtraj, and Alexandria. The open release of the OMat24 dataset and models enables the research community to build upon our efforts and drive further advancements in AI-assisted materials science.

Brandon M. Wood, Misko Dzamba, Xiang Fu et al. · baidu, cmu

The ability to quickly and accurately compute properties from atomic simulations is critical for advancing a large number of applications in chemistry and materials science including drug discovery, energy storage, and semiconductor manufacturing. To address this need, Meta FAIR presents a family of Universal Models for Atoms (UMA), designed to push the frontier of speed, accuracy, and generalization. UMA models are trained on half a billion unique 3D atomic structures (the largest training runs to date) by compiling data across multiple chemical domains, e.g. molecules, materials, and catalysts. We develop empirical scaling laws to help understand how to increase model capacity alongside dataset size to achieve the best accuracy. The UMA small and medium models utilize a novel architectural design we refer to as mixture of linear experts that enables increasing model capacity without sacrificing speed. For example, UMA-medium has 1.4B parameters but only ~50M active parameters per atomic structure. We evaluate UMA models on a diverse set of applications across multiple domains and find that, remarkably, a single model without any fine-tuning can perform similarly or better than specialized models. We are releasing the UMA code, weights, and associated data to accelerate computational workflows and enable the community to continue to build increasingly capable AI models.

Anuroop Sriram, Logan M. Brabson, Xiaohan Yu et al. · baidu, cmu

Identifying useful sorbent materials for direct air capture (DAC) from humid air remains a challenge. We present the Open DAC 2025 (ODAC25) dataset, a significant expansion and improvement upon ODAC23 (Sriram et al., ACS Central Science, 10 (2024) 923), comprising nearly 60 million DFT single-point calculations for CO$_2$, H$_2$O, N$_2$, and O$_2$ adsorption in 15,000 MOFs. ODAC25 introduces chemical and configurational diversity through functionalized MOFs, high-energy GCMC-derived placements, and synthetically generated frameworks. ODAC25 also significantly improves upon the accuracy of DFT calculations and the treatment of flexible MOFs in ODAC23. Along with the dataset, we release new state-of-the-art machine-learned interatomic potentials trained on ODAC25 and evaluate them on adsorption energy and Henry's law coefficient predictions.

Muhammed Shuaibi, Adeesh Kolluru, Abhishek Das et al.

Progress towards the energy breakthroughs needed to combat climate change can be significantly accelerated through the efficient simulation of atomic systems. Simulation techniques based on first principles, such as Density Functional Theory (DFT), are limited in their practical use due to their high computational expense. Machine learning approaches have the potential to approximate DFT in a computationally efficient manner, which could dramatically increase the impact of computational simulations on real-world problems. Approximating DFT poses several challenges. These include accurately modeling the subtle changes in the relative positions and angles between atoms, and enforcing constraints such as rotation invariance or energy conservation. We introduce a novel approach to modeling angular information between sets of neighboring atoms in a graph neural network. Rotation invariance is achieved for the network's edge messages through the use of a per-edge local coordinate frame and a novel spin convolution over the remaining degree of freedom. Two model variants are proposed for the applications of structure relaxation and molecular dynamics. State-of-the-art results are demonstrated on the large-scale Open Catalyst 2020 dataset. Comparisons are also performed on the MD17 and QM9 datasets.

Weihua Hu, Muhammed Shuaibi, Abhishek Das et al.

With massive amounts of atomic simulation data available, there is a huge opportunity to develop fast and accurate machine learning models to approximate expensive physics-based calculations. The key quantity to estimate is atomic forces, where the state-of-the-art Graph Neural Networks (GNNs) explicitly enforce basic physical constraints such as rotation-covariance. However, to strictly satisfy the physical constraints, existing models have to make tradeoffs between computational efficiency and model expressiveness. Here we explore an alternative approach. By not imposing explicit physical constraints, we can flexibly design expressive models while maintaining their computational efficiency. Physical constraints are implicitly imposed by training the models using physics-based data augmentation. To evaluate the approach, we carefully design a scalable and expressive GNN model, ForceNet, and apply it to OC20 (Chanussot et al., 2020), an unprecedentedly-large dataset of quantum physics calculations. Our proposed ForceNet is able to predict atomic forces more accurately than state-of-the-art physics-based GNNs while being faster both in training and inference. Overall, our promising and counter-intuitive results open up an exciting avenue for future research.

Lowik Chanussot, Abhishek Das, Siddharth Goyal et al.

Catalyst discovery and optimization is key to solving many societal and energy challenges including solar fuels synthesis, long-term energy storage, and renewable fertilizer production. Despite considerable effort by the catalysis community to apply machine learning models to the computational catalyst discovery process, it remains an open challenge to build models that can generalize across both elemental compositions of surfaces and adsorbate identity/configurations, perhaps because datasets have been smaller in catalysis than related fields. To address this we developed the OC20 dataset, consisting of 1,281,040 Density Functional Theory (DFT) relaxations (~264,890,000 single point evaluations) across a wide swath of materials, surfaces, and adsorbates (nitrogen, carbon, and oxygen chemistries). We supplemented this dataset with randomly perturbed structures, short timescale molecular dynamics, and electronic structure analyses. The dataset comprises three central tasks indicative of day-to-day catalyst modeling and comes with pre-defined train/validation/test splits to facilitate direct comparisons with future model development efforts. We applied three state-of-the-art graph neural network models (CGCNN, SchNet, Dimenet++) to each of these tasks as baseline demonstrations for the community to build on. In almost every task, no upper limit on model size was identified, suggesting that even larger models are likely to improve on initial results. The dataset and baseline models are both provided as open resources, as well as a public leader board to encourage community contributions to solve these important tasks.

C. Lawrence Zitnick, Lowik Chanussot, Abhishek Das et al.

Scalable and cost-effective solutions to renewable energy storage are essential to addressing the world's rising energy needs while reducing climate change. As we increase our reliance on renewable energy sources such as wind and solar, which produce intermittent power, storage is needed to transfer power from times of peak generation to peak demand. This may require the storage of power for hours, days, or months. One solution that offers the potential of scaling to nation-sized grids is the conversion of renewable energy to other fuels, such as hydrogen or methane. To be widely adopted, this process requires cost-effective solutions to running electrochemical reactions. An open challenge is finding low-cost electrocatalysts to drive these reactions at high rates. Through the use of quantum mechanical simulations (density functional theory), new catalyst structures can be tested and evaluated. Unfortunately, the high computational cost of these simulations limits the number of structures that may be tested. The use of machine learning may provide a method to efficiently approximate these calculations, leading to new approaches in finding effective electrocatalysts. In this paper, we provide an introduction to the challenges in finding suitable electrocatalysts, how machine learning may be applied to the problem, and the use of the Open Catalyst Project OC20 dataset for model training.

Devi Parikh, C. Lawrence Zitnick

As a first step towards studying the ability of human crowds and machines to effectively co-create, we explore several human-only collaborative co-creation scenarios. The goal in each scenario is to create a digital sketch using a simple web interface. We find that settings in which multiple humans iteratively add strokes and vote on the best additions result in the sketches with highest perceived creativity (value + novelty). Lack of collaboration leads to a higher variance in quality and lower novelty or surprise. Collaboration without voting leads to high novelty but low quality.

Anuroop Sriram, Jure Zbontar, Tullie Murrell et al.

The slow acquisition speed of magnetic resonance imaging (MRI) has led to the development of two complementary methods: acquiring multiple views of the anatomy simultaneously (parallel imaging) and acquiring fewer samples than necessary for traditional signal processing methods (compressed sensing). While the combination of these methods has the potential to allow much faster scan times, reconstruction from such undersampled multi-coil data has remained an open problem. In this paper, we present a new approach to this problem that extends previously proposed variational methods by learning fully end-to-end. Our method obtains new state-of-the-art results on the fastMRI dataset for both brain and knee MRIs.

Florian Knoll, Tullie Murrell, Anuroop Sriram et al.

Purpose: To advance research in the field of machine learning for MR image reconstruction with an open challenge. Methods: We provided participants with a dataset of raw k-space data from 1,594 consecutive clinical exams of the knee. The goal of the challenge was to reconstruct images from these data. In order to strike a balance between realistic data and a shallow learning curve for those not already familiar with MR image reconstruction, we ran multiple tracks for multi-coil and single-coil data. We performed a two-stage evaluation based on quantitative image metrics followed by evaluation by a panel of radiologists. The challenge ran from June to December of 2019. Results: We received a total of 33 challenge submissions. All participants chose to submit results from supervised machine learning approaches. Conclusion: The challenge led to new developments in machine learning for image reconstruction, provided insight into the current state of the art in the field, and highlighted remaining hurdles for clinical adoption.

Anuroop Sriram, Jure Zbontar, Tullie Murrell et al.

Magnetic Resonance Image (MRI) acquisition is an inherently slow process which has spurred the development of two different acceleration methods: acquiring multiple correlated samples simultaneously (parallel imaging) and acquiring fewer samples than necessary for traditional signal processing methods (compressed sensing). Both methods provide complementary approaches to accelerating the speed of MRI acquisition. In this paper, we present a novel method to integrate traditional parallel imaging methods into deep neural networks that is able to generate high quality reconstructions even for high acceleration factors. The proposed method, called GrappaNet, performs progressive reconstruction by first mapping the reconstruction problem to a simpler one that can be solved by a traditional parallel imaging methods using a neural network, followed by an application of a parallel imaging method, and finally fine-tuning the output with another neural network. The entire network can be trained end-to-end. We present experimental results on the recently released fastMRI dataset and show that GrappaNet can generate higher quality reconstructions than competing methods for both $4\times$ and $8\times$ acceleration.

Arthur Szlam, Jonathan Gray, Kavya Srinet et al.

In this document we describe a rationale for a research program aimed at building an open "assistant" in the game Minecraft, in order to make progress on the problems of natural language understanding and learning from dialogue.

Jonathan Gray, Kavya Srinet, Yacine Jernite et al.

This paper describes an implementation of a bot assistant in Minecraft, and the tools and platform allowing players to interact with the bot and to record those interactions. The purpose of building such an assistant is to facilitate the study of agents that can complete tasks specified by dialogue, and eventually, to learn from dialogue interactions.

Jure Zbontar, Florian Knoll, Anuroop Sriram et al.

Accelerating Magnetic Resonance Imaging (MRI) by taking fewer measurements has the potential to reduce medical costs, minimize stress to patients and make MRI possible in applications where it is currently prohibitively slow or expensive. We introduce the fastMRI dataset, a large-scale collection of both raw MR measurements and clinical MR images, that can be used for training and evaluation of machine-learning approaches to MR image reconstruction. By introducing standardized evaluation criteria and a freely-accessible dataset, our goal is to help the community make rapid advances in the state of the art for MR image reconstruction. We also provide a self-contained introduction to MRI for machine learning researchers with no medical imaging background.

Justin Johnson, Bharath Hariharan, Laurens van der Maaten et al.

Existing methods for visual reasoning attempt to directly map inputs to outputs using black-box architectures without explicitly modeling the underlying reasoning processes. As a result, these black-box models often learn to exploit biases in the data rather than learning to perform visual reasoning. Inspired by module networks, this paper proposes a model for visual reasoning that consists of a program generator that constructs an explicit representation of the reasoning process to be performed, and an execution engine that executes the resulting program to produce an answer. Both the program generator and the execution engine are implemented by neural networks, and are trained using a combination of backpropagation and REINFORCE. Using the CLEVR benchmark for visual reasoning, we show that our model significantly outperforms strong baselines and generalizes better in a variety of settings.

Justin Johnson, Bharath Hariharan, Laurens van der Maaten et al.

When building artificial intelligence systems that can reason and answer questions about visual data, we need diagnostic tests to analyze our progress and discover shortcomings. Existing benchmarks for visual question answering can help, but have strong biases that models can exploit to correctly answer questions without reasoning. They also conflate multiple sources of error, making it hard to pinpoint model weaknesses. We present a diagnostic dataset that tests a range of visual reasoning abilities. It contains minimal biases and has detailed annotations describing the kind of reasoning each question requires. We use this dataset to analyze a variety of modern visual reasoning systems, providing novel insights into their abilities and limitations.

C. Lawrence Zitnick, Aishwarya Agrawal, Stanislaw Antol et al.

As machines have become more intelligent, there has been a renewed interest in methods for measuring their intelligence. A common approach is to propose tasks for which a human excels, but one which machines find difficult. However, an ideal task should also be easy to evaluate and not be easily gameable. We begin with a case study exploring the recently popular task of image captioning and its limitations as a task for measuring machine intelligence. An alternative and more promising task is Visual Question Answering that tests a machine's ability to reason about language and vision. We describe a dataset unprecedented in size created for the task that contains over 760,000 human generated questions about images. Using around 10 million human generated answers, machines may be easily evaluated.

Abhishek Das, Harsh Agrawal, C. Lawrence Zitnick et al.

We conduct large-scale studies on `human attention' in Visual Question Answering (VQA) to understand where humans choose to look to answer questions about images. We design and test multiple game-inspired novel attention-annotation interfaces that require the subject to sharpen regions of a blurred image to answer a question. Thus, we introduce the VQA-HAT (Human ATtention) dataset. We evaluate attention maps generated by state-of-the-art VQA models against human attention both qualitatively (via visualizations) and quantitatively (via rank-order correlation). Overall, our experiments show that current attention models in VQA do not seem to be looking at the same regions as humans.

Abhishek Das, Harsh Agrawal, C. Lawrence Zitnick et al.

We conduct large-scale studies on `human attention' in Visual Question Answering (VQA) to understand where humans choose to look to answer questions about images. We design and test multiple game-inspired novel attention-annotation interfaces that require the subject to sharpen regions of a blurred image to answer a question. Thus, we introduce the VQA-HAT (Human ATtention) dataset. We evaluate attention maps generated by state-of-the-art VQA models against human attention both qualitatively (via visualizations) and quantitatively (via rank-order correlation). Overall, our experiments show that current attention models in VQA do not seem to be looking at the same regions as humans.

Ting-Hao, Huang, Francis Ferraro et al.

We introduce the first dataset for sequential vision-to-language, and explore how this data may be used for the task of visual storytelling. The first release of this dataset, SIND v.1, includes 81,743 unique photos in 20,211 sequences, aligned to both descriptive (caption) and story language. We establish several strong baselines for the storytelling task, and motivate an automatic metric to benchmark progress. Modelling concrete description as well as figurative and social language, as provided in this dataset and the storytelling task, has the potential to move artificial intelligence from basic understandings of typical visual scenes towards more and more human-like understanding of grounded event structure and subjective expression.

Ishan Misra, C. Lawrence Zitnick, Martial Hebert

In this paper, we present an approach for learning a visual representation from the raw spatiotemporal signals in videos. Our representation is learned without supervision from semantic labels. We formulate our method as an unsupervised sequential verification task, i.e., we determine whether a sequence of frames from a video is in the correct temporal order. With this simple task and no semantic labels, we learn a powerful visual representation using a Convolutional Neural Network (CNN). The representation contains complementary information to that learned from supervised image datasets like ImageNet. Qualitative results show that our method captures information that is temporally varying, such as human pose. When used as pre-training for action recognition, our method gives significant gains over learning without external data on benchmark datasets like UCF101 and HMDB51. To demonstrate its sensitivity to human pose, we show results for pose estimation on the FLIC and MPII datasets that are competitive, or better than approaches using significantly more supervision. Our method can be combined with supervised representations to provide an additional boost in accuracy.

Ishan Misra, C. Lawrence Zitnick, Margaret Mitchell et al.

When human annotators are given a choice about what to label in an image, they apply their own subjective judgments on what to ignore and what to mention. We refer to these noisy "human-centric" annotations as exhibiting human reporting bias. Examples of such annotations include image tags and keywords found on photo sharing sites, or in datasets containing image captions. In this paper, we use these noisy annotations for learning visually correct image classifiers. Such annotations do not use consistent vocabulary, and miss a significant amount of the information present in an image; however, we demonstrate that the noise in these annotations exhibits structure and can be modeled. We propose an algorithm to decouple the human reporting bias from the correct visually grounded labels. Our results are highly interpretable for reporting "what's in the image" versus "what's worth saying." We demonstrate the algorithm's efficacy along a variety of metrics and datasets, including MS COCO and Yahoo Flickr 100M. We show significant improvements over traditional algorithms for both image classification and image captioning, doubling the performance of existing methods in some cases.

Arjun Chandrasekaran, Ashwin K. Vijayakumar, Stanislaw Antol et al.

Humor is an integral part of human lives. Despite being tremendously impactful, it is perhaps surprising that we do not have a detailed understanding of humor yet. As interactions between humans and AI systems increase, it is imperative that these systems are taught to understand subtleties of human expressions such as humor. In this work, we are interested in the question - what content in a scene causes it to be funny? As a first step towards understanding visual humor, we analyze the humor manifested in abstract scenes and design computational models for them. We collect two datasets of abstract scenes that facilitate the study of humor at both the scene-level and the object-level. We analyze the funny scenes and explore the different types of humor depicted in them via human studies. We model two tasks that we believe demonstrate an understanding of some aspects of visual humor. The tasks involve predicting the funniness of a scene and altering the funniness of a scene. We show that our models perform well quantitatively, and qualitatively through human studies. Our datasets are publicly available.

Sean Bell, C. Lawrence Zitnick, Kavita Bala et al.

It is well known that contextual and multi-scale representations are important for accurate visual recognition. In this paper we present the Inside-Outside Net (ION), an object detector that exploits information both inside and outside the region of interest. Contextual information outside the region of interest is integrated using spatial recurrent neural networks. Inside, we use skip pooling to extract information at multiple scales and levels of abstraction. Through extensive experiments we evaluate the design space and provide readers with an overview of what tricks of the trade are important. ION improves state-of-the-art on PASCAL VOC 2012 object detection from 73.9% to 76.4% mAP. On the new and more challenging MS COCO dataset, we improve state-of-art-the from 19.7% to 33.1% mAP. In the 2015 MS COCO Detection Challenge, our ION model won the Best Student Entry and finished 3rd place overall. As intuition suggests, our detection results provide strong evidence that context and multi-scale representations improve small object detection.

Fereshteh Sadeghi, C. Lawrence Zitnick, Ali Farhadi

In this paper, we study the problem of answering visual analogy questions. These questions take the form of image A is to image B as image C is to what. Answering these questions entails discovering the mapping from image A to image B and then extending the mapping to image C and searching for the image D such that the relation from A to B holds for C to D. We pose this problem as learning an embedding that encourages pairs of analogous images with similar transformations to be close together using convolutional neural networks with a quadruple Siamese architecture. We introduce a dataset of visual analogy questions in natural images, and show first results of its kind on solving analogy questions on natural images.

Jacob Devlin, Saurabh Gupta, Ross Girshick et al.

We explore a variety of nearest neighbor baseline approaches for image captioning. These approaches find a set of nearest neighbor images in the training set from which a caption may be borrowed for the query image. We select a caption for the query image by finding the caption that best represents the "consensus" of the set of candidate captions gathered from the nearest neighbor images. When measured by automatic evaluation metrics on the MS COCO caption evaluation server, these approaches perform as well as many recent approaches that generate novel captions. However, human studies show that a method that generates novel captions is still preferred over the nearest neighbor approach.

Aishwarya Agrawal, Jiasen Lu, Stanislaw Antol et al.

We propose the task of free-form and open-ended Visual Question Answering (VQA). Given an image and a natural language question about the image, the task is to provide an accurate natural language answer. Mirroring real-world scenarios, such as helping the visually impaired, both the questions and answers are open-ended. Visual questions selectively target different areas of an image, including background details and underlying context. As a result, a system that succeeds at VQA typically needs a more detailed understanding of the image and complex reasoning than a system producing generic image captions. Moreover, VQA is amenable to automatic evaluation, since many open-ended answers contain only a few words or a closed set of answers that can be provided in a multiple-choice format. We provide a dataset containing ~0.25M images, ~0.76M questions, and ~10M answers (www.visualqa.org), and discuss the information it provides. Numerous baselines and methods for VQA are provided and compared with human performance. Our VQA demo is available on CloudCV (http://cloudcv.org/vqa).

Xinlei Chen, Hao Fang, Tsung-Yi Lin et al.

In this paper we describe the Microsoft COCO Caption dataset and evaluation server. When completed, the dataset will contain over one and a half million captions describing over 330,000 images. For the training and validation images, five independent human generated captions will be provided. To ensure consistency in evaluation of automatic caption generation algorithms, an evaluation server is used. The evaluation server receives candidate captions and scores them using several popular metrics, including BLEU, METEOR, ROUGE and CIDEr. Instructions for using the evaluation server are provided.

Ramakrishna Vedantam, C. Lawrence Zitnick, Devi Parikh

Automatically describing an image with a sentence is a long-standing challenge in computer vision and natural language processing. Due to recent progress in object detection, attribute classification, action recognition, etc., there is renewed interest in this area. However, evaluating the quality of descriptions has proven to be challenging. We propose a novel paradigm for evaluating image descriptions that uses human consensus. This paradigm consists of three main parts: a new triplet-based method of collecting human annotations to measure consensus, a new automated metric (CIDEr) that captures consensus, and two new datasets: PASCAL-50S and ABSTRACT-50S that contain 50 sentences describing each image. Our simple metric captures human judgment of consensus better than existing metrics across sentences generated by various sources. We also evaluate five state-of-the-art image description approaches using this new protocol and provide a benchmark for future comparisons. A version of CIDEr named CIDEr-D is available as a part of MS COCO evaluation server to enable systematic evaluation and benchmarking.

Xinlei Chen, C. Lawrence Zitnick

In this paper we explore the bi-directional mapping between images and their sentence-based descriptions. We propose learning this mapping using a recurrent neural network. Unlike previous approaches that map both sentences and images to a common embedding, we enable the generation of novel sentences given an image. Using the same model, we can also reconstruct the visual features associated with an image given its visual description. We use a novel recurrent visual memory that automatically learns to remember long-term visual concepts to aid in both sentence generation and visual feature reconstruction. We evaluate our approach on several tasks. These include sentence generation, sentence retrieval and image retrieval. State-of-the-art results are shown for the task of generating novel image descriptions. When compared to human generated captions, our automatically generated captions are preferred by humans over $19.8\%$ of the time. Results are better than or comparable to state-of-the-art results on the image and sentence retrieval tasks for methods using similar visual features.

Hao Fang, Saurabh Gupta, Forrest Iandola et al.

This paper presents a novel approach for automatically generating image descriptions: visual detectors, language models, and multimodal similarity models learnt directly from a dataset of image captions. We use multiple instance learning to train visual detectors for words that commonly occur in captions, including many different parts of speech such as nouns, verbs, and adjectives. The word detector outputs serve as conditional inputs to a maximum-entropy language model. The language model learns from a set of over 400,000 image descriptions to capture the statistics of word usage. We capture global semantics by re-ranking caption candidates using sentence-level features and a deep multimodal similarity model. Our system is state-of-the-art on the official Microsoft COCO benchmark, producing a BLEU-4 score of 29.1%. When human judges compare the system captions to ones written by other people on our held-out test set, the system captions have equal or better quality 34% of the time.

Ramakrishna Vedantam, C. Lawrence Zitnick, Devi Parikh

We describe our two new datasets with images described by humans. Both the datasets were collected using Amazon Mechanical Turk, a crowdsourcing platform. The two datasets contain significantly more descriptions per image than other existing datasets. One is based on a popular image description dataset called the UIUC Pascal Sentence Dataset, whereas the other is based on the Abstract Scenes dataset con- taining images made from clipart objects. In this paper we describe our interfaces, analyze some properties of and show example descriptions from our two datasets.

Piotr Dollár, C. Lawrence Zitnick

Edge detection is a critical component of many vision systems, including object detectors and image segmentation algorithms. Patches of edges exhibit well-known forms of local structure, such as straight lines or T-junctions. In this paper we take advantage of the structure present in local image patches to learn both an accurate and computationally efficient edge detector. We formulate the problem of predicting local edge masks in a structured learning framework applied to random decision forests. Our novel approach to learning decision trees robustly maps the structured labels to a discrete space on which standard information gain measures may be evaluated. The result is an approach that obtains realtime performance that is orders of magnitude faster than many competing state-of-the-art approaches, while also achieving state-of-the-art edge detection results on the BSDS500 Segmentation dataset and NYU Depth dataset. Finally, we show the potential of our approach as a general purpose edge detector by showing our learned edge models generalize well across datasets.

Tsung-Yi Lin, Michael Maire, Serge Belongie et al.

We present a new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of scene understanding. This is achieved by gathering images of complex everyday scenes containing common objects in their natural context. Objects are labeled using per-instance segmentations to aid in precise object localization. Our dataset contains photos of 91 objects types that would be easily recognizable by a 4 year old. With a total of 2.5 million labeled instances in 328k images, the creation of our dataset drew upon extensive crowd worker involvement via novel user interfaces for category detection, instance spotting and instance segmentation. We present a detailed statistical analysis of the dataset in comparison to PASCAL, ImageNet, and SUN. Finally, we provide baseline performance analysis for bounding box and segmentation detection results using a Deformable Parts Model.