Govind Thattai

Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri et al. · allen-ai, berkeley

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

Hangjie Shi, Leslie Ball, Govind Thattai et al. · amazon-science

The Alexa Prize program has empowered numerous university students to explore, experiment, and showcase their talents in building conversational agents through challenges like the SocialBot Grand Challenge and the TaskBot Challenge. As conversational agents increasingly appear in multimodal and embodied contexts, it is important to explore the affordances of conversational interaction augmented with computer vision and physical embodiment. This paper describes the SimBot Challenge, a new challenge in which university teams compete to build robot assistants that complete tasks in a simulated physical environment. This paper provides an overview of the SimBot Challenge, which included both online and offline challenge phases. We describe the infrastructure and support provided to the teams including Alexa Arena, the simulated environment, and the ML toolkit provided to teams to accelerate their building of vision and language models. We summarize the approaches the participating teams took to overcome research challenges and extract key lessons learned. Finally, we provide analysis of the performance of the competing SimBots during the competition.

Yizhou Zhao, Qiaozi Gao, Liang Qiu et al. · amazon-science

We introduce OPEND, a benchmark for learning how to use a hand to open cabinet doors or drawers in a photo-realistic and physics-reliable simulation environment driven by language instruction. To solve the task, we propose a multi-step planner composed of a deep neural network and rule-base controllers. The network is utilized to capture spatial relationships from images and understand semantic meaning from language instructions. Controllers efficiently execute the plan based on the spatial and semantic understanding. We evaluate our system by measuring its zero-shot performance in test data set. Experimental results demonstrate the effectiveness of decision planning by our multi-step planner for different hands, while suggesting that there is significant room for developing better models to address the challenge brought by language understanding, spatial reasoning, and long-term manipulation. We will release OPEND and host challenges to promote future research in this area.

Qiaozi Gao, Govind Thattai, Suhaila Shakiah et al.

We introduce Alexa Arena, a user-centric simulation platform for Embodied AI (EAI) research. Alexa Arena provides a variety of multi-room layouts and interactable objects, for the creation of human-robot interaction (HRI) missions. With user-friendly graphics and control mechanisms, Alexa Arena supports the development of gamified robotic tasks readily accessible to general human users, thus opening a new venue for high-efficiency HRI data collection and EAI system evaluation. Along with the platform, we introduce a dialog-enabled instruction-following benchmark and provide baseline results for it. We make Alexa Arena publicly available to facilitate research in building generalizable and assistive embodied agents.

Ran Gong, Xiaofeng Gao, Qiaozi Gao et al.

Complex manipulation tasks often require robots with complementary capabilities to collaborate. We introduce a benchmark for LanguagE-Conditioned Multi-robot MAnipulation (LEMMA) focused on task allocation and long-horizon object manipulation based on human language instructions in a tabletop setting. LEMMA features 8 types of procedurally generated tasks with varying degree of complexity, some of which require the robots to use tools and pass tools to each other. For each task, we provide 800 expert demonstrations and human instructions for training and evaluations. LEMMA poses greater challenges compared to existing benchmarks, as it requires the system to identify each manipulator's limitations and assign sub-tasks accordingly while also handling strong temporal dependencies in each task. To address these challenges, we propose a modular hierarchical planning approach as a baseline. Our results highlight the potential of LEMMA for developing future language-conditioned multi-robot systems.

Da Yin, Feng Gao, Govind Thattai et al.

A key goal for the advancement of AI is to develop technologies that serve the needs not just of one group but of all communities regardless of their geographical region. In fact, a significant proportion of knowledge is locally shared by people from certain regions but may not apply equally in other regions because of cultural differences. If a model is unaware of regional characteristics, it may lead to performance disparity across regions and result in bias against underrepresented groups. We propose GIVL, a Geographically Inclusive Vision-and-Language Pre-trained model. There are two attributes of geo-diverse visual concepts which can help to learn geo-diverse knowledge: 1) concepts under similar categories have unique knowledge and visual characteristics, 2) concepts with similar visual features may fall in completely different categories. Motivated by the attributes, we design new pre-training objectives Image Knowledge Matching (IKM) and Image Edit Checking (IEC) to pre-train GIVL. Compared with similar-size models pre-trained with similar scale of data, GIVL achieves state-of-the-art (SOTA) and more balanced performance on geo-diverse V&L tasks.

Vasu Sharma, Prasoon Goyal, Kaixiang Lin et al.

We propose a multimodal (vision-and-language) benchmark for cooperative and heterogeneous multi-agent learning. We introduce a benchmark multimodal dataset with tasks involving collaboration between multiple simulated heterogeneous robots in a rich multi-room home environment. We provide an integrated learning framework, multimodal implementations of state-of-the-art multi-agent reinforcement learning techniques, and a consistent evaluation protocol. Our experiments investigate the impact of different modalities on multi-agent learning performance. We also introduce a simple message passing method between agents. The results suggest that multimodality introduces unique challenges for cooperative multi-agent learning and there is significant room for advancing multi-agent reinforcement learning methods in such settings.

Nirbhay Modhe, Qiaozi Gao, Ashwin Kalyan et al.

Offline reinforcement learning (RL) methods strike a balance between exploration and exploitation by conservative value estimation -- penalizing values of unseen states and actions. Model-free methods penalize values at all unseen actions, while model-based methods are able to further exploit unseen states via model rollouts. However, such methods are handicapped in their ability to find unseen states far away from the available offline data due to two factors -- (a) very short rollout horizons in models due to cascading model errors, and (b) model rollouts originating solely from states observed in offline data. We relax the second assumption and present a novel unseen state augmentation strategy to allow exploitation of unseen states where the learned model and value estimates generalize. Our strategy finds unseen states by value-informed perturbations of seen states followed by filtering out states with epistemic uncertainty estimates too high (high error) or too low (too similar to seen data). We observe improved performance in several offline RL tasks and find that our augmentation strategy consistently leads to overall lower average dataset Q-value estimates i.e. more conservative Q-value estimates than a baseline.

Yuqian Jiang, Qiaozi Gao, Govind Thattai et al.

For service robots to become general-purpose in everyday household environments, they need not only a large library of primitive skills, but also the ability to quickly learn novel tasks specified by users. Fine-tuning neural networks on a variety of downstream tasks has been successful in many vision and language domains, but research is still limited on transfer learning between diverse long-horizon tasks. We propose that, compared to reinforcement learning for a new household activity from scratch, home robots can benefit from transferring the value and policy networks trained for similar tasks. We evaluate this idea in the BEHAVIOR simulation benchmark which includes a large number of household activities and a set of action primitives. For easy mapping between state spaces of different tasks, we provide a text-based representation and leverage language models to produce a common embedding space. The results show that the selection of similar source activities can be informed by the semantic similarity of state and goal descriptions with the target task. We further analyze the results and discuss ways to overcome the problem of catastrophic forgetting.

Yuxing Qiu, Feng Gao, Minchen Li et al.

Recent breakthroughs in Vision-Language (V&L) joint research have achieved remarkable results in various text-driven tasks. High-quality Text-to-video (T2V), a task that has been long considered mission-impossible, was proven feasible with reasonably good results in latest works. However, the resulting videos often have undesired artifacts largely because the system is purely data-driven and agnostic to the physical laws. To tackle this issue and further push T2V towards high-level physical realism, we present an autonomous data generation technique and a dataset, which intend to narrow the gap with a large number of multi-modal, 3D Text-to-Video/Simulation (T2V/S) data. In the dataset, we provide high-resolution 3D physical simulations for both solids and fluids, along with textual descriptions of the physical phenomena. We take advantage of state-of-the-art physical simulation methods (i) Incremental Potential Contact (IPC) and (ii) Material Point Method (MPM) to simulate diverse scenarios, including elastic deformations, material fractures, collisions, turbulence, etc. Additionally, high-quality, multi-view rendering videos are supplied for the benefit of T2V, Neural Radiance Fields (NeRF), and other communities. This work is the first step towards fully automated Text-to-Video/Simulation (T2V/S). Live examples and subsequent work are at https://sites.google.com/view/tpa-net.

Rakesh Vaideeswaran, Feng Gao, Abhinav Mathur et al.

The domain of joint vision-language understanding, especially in the context of reasoning in Visual Question Answering (VQA) models, has garnered significant attention in the recent past. While most of the existing VQA models focus on improving the accuracy of VQA, the way models arrive at an answer is oftentimes a black box. As a step towards making the VQA task more explainable and interpretable, our method is built upon the SOTA VQA framework by augmenting it with an end-to-end explanation generation module. In this paper, we investigate two network architectures, including Long Short-Term Memory (LSTM) and Transformer decoder, as the explanation generator. Our method generates human-readable textual explanations while maintaining SOTA VQA accuracy on the GQA-REX (77.49%) and VQA-E (71.48%) datasets. Approximately 65.16% of the generated explanations are approved by humans as valid. Roughly 60.5% of the generated explanations are valid and lead to the correct answers.

Yubo Zhang, Feiyang Niu, Qing Ping et al.

To solve video-and-language grounding tasks, the key is for the network to understand the connection between the two modalities. For a pair of video and language description, their semantic relation is reflected by their encodings' similarity. A good multi-modality encoder should be able to well capture both inputs' semantics and encode them in the shared feature space where embedding distance gets properly translated into their semantic similarity. In this work, we focused on this semantic connection between video and language, and developed a multi-level alignment training scheme to directly shape the encoding process. Global and segment levels of video-language alignment pairs were designed, based on the information similarity ranging from high-level context to fine-grained semantics. The contrastive loss was used to contrast the encodings' similarities between the positive and negative alignment pairs, and to ensure the network is trained in such a way that similar information is encoded closely in the shared feature space while information of different semantics is kept apart. Our multi-level alignment training can be applied to various video-and-language grounding tasks. Together with the task-specific training loss, our framework achieved comparable performance to previous state-of-the-arts on multiple video QA and retrieval datasets.

Yang Fu, Yuliang Zou, Hao Xiang et al.

In autonomous driving, mapping is critical for motion planning but remains an under-utilized resource for perception tasks such as 3D object detection. Maps can provide robust structural priors of the static environment, helping resolve ambiguities and correct for sensor data sparsity or noise, especially for distant objects or under adverse weather conditions. However, conventional High-Definition (HD) maps are resource-intensive to obtain and maintain, which presents a challenge for efficient, large-scale deployment. In this paper, we propose a scalable solution to systematically leverage mapping to improve 3D detection by overcoming two primary challenges. First, we introduce a pipeline to automatically build dense mapping priors from aggregated sensor data, eliminating the need for human labeling. Second, we design a novel Mapping Priors Augmented 3D Detection (MPA3D) framework to effectively integrate mapping priors with different sensor modalities. Extensive experiments on the Waymo Open Dataset demonstrate that our approach achieves new state-of-the-art results, proving the effectiveness of scalable reconstructed scene priors for enhancing 3D detection.

Yingwei Li, Xin Huang, Yang Liu et al.

Model scaling has demonstrated remarkable success through large-scale training on diverse datasets. It remains an open question whether the same paradigm would apply to autonomous driving perception systems due to unique challenges, such as fusing heterogeneous sensor data and the need for sophisticated 3D spatial understanding. To bridge this gap, we present a comprehensive study on systematically analyzing the impact of scale on these systems. We develop our STELLAR model based on Sparse Window Transformer, by extending the input modalities to include LiDAR, radar, camera, and map prior. We train the model on a large-scale dataset of 50 million driving examples with up to 500 million parameters. Our large-scale experiments reveal empirical scaling trends that connect model performance to model size, data, and compute. The resulting model establishes a new state-of-the-art on the Waymo Open Dataset challenge, outperforming prior arts by a large margin. Our work demonstrates that large-scale training is a highly promising path for advancing the capabilities of perception models for autonomous driving.

Neal Lawton, Anoop Kumar, Govind Thattai et al.

Parameter-efficient tuning (PET) methods fit pre-trained language models (PLMs) to downstream tasks by either computing a small compressed update for a subset of model parameters, or appending and fine-tuning a small number of new model parameters to the pre-trained network. Hand-designed PET architectures from the literature perform well in practice, but have the potential to be improved via automated neural architecture search (NAS). We propose an efficient NAS method for learning PET architectures via structured and unstructured pruning. We present experiments on GLUE demonstrating the effectiveness of our algorithm and discuss how PET architectural design choices affect performance in practice.

Xiaofeng Gao, Qiaozi Gao, Ran Gong et al.

Language-guided Embodied AI benchmarks requiring an agent to navigate an environment and manipulate objects typically allow one-way communication: the human user gives a natural language command to the agent, and the agent can only follow the command passively. We present DialFRED, a dialogue-enabled embodied instruction following benchmark based on the ALFRED benchmark. DialFRED allows an agent to actively ask questions to the human user; the additional information in the user's response is used by the agent to better complete its task. We release a human-annotated dataset with 53K task-relevant questions and answers and an oracle to answer questions. To solve DialFRED, we propose a questioner-performer framework wherein the questioner is pre-trained with the human-annotated data and fine-tuned with reinforcement learning. We make DialFRED publicly available and encourage researchers to propose and evaluate their solutions to building dialog-enabled embodied agents.

Cristian-Paul Bara, Qing Ping, Abhinav Mathur et al.

We introduce a novel privacy-preserving methodology for performing Visual Question Answering on the edge. Our method constructs a symbolic representation of the visual scene, using a low-complexity computer vision model that jointly predicts classes, attributes and predicates. This symbolic representation is non-differentiable, which means it cannot be used to recover the original image, thereby keeping the original image private. Our proposed hybrid solution uses a vision model which is more than 25 times smaller than the current state-of-the-art (SOTA) vision models, and 100 times smaller than end-to-end SOTA VQA models. We report detailed error analysis and discuss the trade-offs of using a distilled vision model and a symbolic representation of the visual scene.

Zhiwei Jia, Kaixiang Lin, Yizhou Zhao et al.

Recent years have witnessed an emerging paradigm shift toward embodied artificial intelligence, in which an agent must learn to solve challenging tasks by interacting with its environment. There are several challenges in solving embodied multimodal tasks, including long-horizon planning, vision-and-language grounding, and efficient exploration. We focus on a critical bottleneck, namely the performance of planning and navigation. To tackle this challenge, we propose a Neural SLAM approach that, for the first time, utilizes several modalities for exploration, predicts an affordance-aware semantic map, and plans over it at the same time. This significantly improves exploration efficiency, leads to robust long-horizon planning, and enables effective vision-and-language grounding. With the proposed Affordance-aware Multimodal Neural SLAM (AMSLAM) approach, we obtain more than 40% improvement over prior published work on the ALFRED benchmark and set a new state-of-the-art generalization performance at a success rate of 23.48% on the test unseen scenes.

Tongzhou Mu, Kaixiang Lin, Feiyang Niu et al.

We present a two-step hybrid reinforcement learning (RL) policy that is designed to generate interpretable and robust hierarchical policies on the RL problem with graph-based input. Unlike prior deep reinforcement learning policies parameterized by an end-to-end black-box graph neural network, our approach disentangles the decision-making process into two steps. The first step is a simplified classification problem that maps the graph input to an action group where all actions share a similar semantic meaning. The second step implements a sophisticated rule-miner that conducts explicit one-hop reasoning over the graph and identifies decisive edges in the graph input without the necessity of heavy domain knowledge. This two-step hybrid policy presents human-friendly interpretations and achieves better performance in terms of generalization and robustness. Extensive experimental studies on four levels of complex text-based games have demonstrated the superiority of the proposed method compared to the state-of-the-art.

Feng Gao, Qing Ping, Govind Thattai et al.

Outside-knowledge visual question answering (OK-VQA) requires the agent to comprehend the image, make use of relevant knowledge from the entire web, and digest all the information to answer the question. Most previous works address the problem by first fusing the image and question in the multi-modal space, which is inflexible for further fusion with a vast amount of external knowledge. In this paper, we call for a paradigm shift for the OK-VQA task, which transforms the image into plain text, so that we can enable knowledge passage retrieval, and generative question-answering in the natural language space. This paradigm takes advantage of the sheer volume of gigantic knowledge bases and the richness of pre-trained language models. A Transform-Retrieve-Generate framework (TRiG) framework is proposed, which can be plug-and-played with alternative image-to-text models and textual knowledge bases. Experimental results show that our TRiG framework outperforms all state-of-the-art supervised methods by at least 11.1% absolute margin.

Shane Storks, Qiaozi Gao, Aishwarya Reganti et al.

Language-enabled AI systems can answer complex, multi-hop questions to high accuracy, but supporting answers with evidence is a more challenging task which is important for the transparency and trustworthiness to users. Prior work in this area typically makes a trade-off between efficiency and accuracy; state-of-the-art deep neural network systems are too cumbersome to be useful in large-scale applications, while the fastest systems lack reliability. In this work, we integrate fast syntactic methods with powerful semantic methods for multi-hop explanation generation based on declarative facts. Our best system, which learns a lightweight operation to simulate multi-hop reasoning over pieces of evidence and fine-tunes language models to re-rank generated explanation chains, outperforms a purely syntactic baseline from prior work by up to 7% in gold explanation retrieval rate.

Yizhou Zhao, Kaixiang Lin, Zhiwei Jia et al.

Learning-based methods for training embodied agents typically require a large number of high-quality scenes that contain realistic layouts and support meaningful interactions. However, current simulators for Embodied AI (EAI) challenges only provide simulated indoor scenes with a limited number of layouts. This paper presents Luminous, the first research framework that employs state-of-the-art indoor scene synthesis algorithms to generate large-scale simulated scenes for Embodied AI challenges. Further, we automatically and quantitatively evaluate the quality of generated indoor scenes via their ability to support complex household tasks. Luminous incorporates a novel scene generation algorithm (Constrained Stochastic Scene Generation (CSSG)), which achieves competitive performance with human-designed scenes. Within Luminous, the EAI task executor, task instruction generation module, and video rendering toolkit can collectively generate a massive multimodal dataset of new scenes for the training and evaluation of Embodied AI agents. Extensive experimental results demonstrate the effectiveness of the data generated by Luminous, enabling the comprehensive assessment of embodied agents on generalization and robustness.

Alessandro Suglia, Qiaozi Gao, Jesse Thomason et al.

Language-guided robots performing home and office tasks must navigate in and interact with the world. Grounding language instructions against visual observations and actions to take in an environment is an open challenge. We present Embodied BERT (EmBERT), a transformer-based model which can attend to high-dimensional, multi-modal inputs across long temporal horizons for language-conditioned task completion. Additionally, we bridge the gap between successful object-centric navigation models used for non-interactive agents and the language-guided visual task completion benchmark, ALFRED, by introducing object navigation targets for EmBERT training. We achieve competitive performance on the ALFRED benchmark, and EmBERT marks the first transformer-based model to successfully handle the long-horizon, dense, multi-modal histories of ALFRED, and the first ALFRED model to utilize object-centric navigation targets.

Tao Tu, Qing Ping, Govind Thattai et al.

GuessWhat?! is a two-player visual dialog guessing game where player A asks a sequence of yes/no questions (Questioner) and makes a final guess (Guesser) about a target object in an image, based on answers from player B (Oracle). Based on this dialog history between the Questioner and the Oracle, a Guesser makes a final guess of the target object. Previous baseline Oracle model encodes no visual information in the model, and it cannot fully understand complex questions about color, shape, relationships and so on. Most existing work for Guesser encode the dialog history as a whole and train the Guesser models from scratch on the GuessWhat?! dataset. This is problematic since language encoder tend to forget long-term history and the GuessWhat?! data is sparse in terms of learning visual grounding of objects. Previous work for Questioner introduces state tracking mechanism into the model, but it is learned as a soft intermediates without any prior vision-linguistic insights. To bridge these gaps, in this paper we propose Vilbert-based Oracle, Guesser and Questioner, which are all built on top of pretrained vision-linguistic model, Vilbert. We introduce two-way background/target fusion mechanism into Vilbert-Oracle to account for both intra and inter-object questions. We propose a unified framework for Vilbert-Guesser and Vilbert-Questioner, where state-estimator is introduced to best utilize Vilbert's power on single-turn referring expression comprehension. Experimental results show that our proposed models outperform state-of-the-art models significantly by 7%, 10%, 12% for Oracle, Guesser and End-to-End Questioner respectively.

Shane Storks, Qiaozi Gao, Govind Thattai et al.

Embodied instruction following is a challenging problem requiring an agent to infer a sequence of primitive actions to achieve a goal environment state from complex language and visual inputs. Action Learning From Realistic Environments and Directives (ALFRED) is a recently proposed benchmark for this problem consisting of step-by-step natural language instructions to achieve subgoals which compose to an ultimate high-level goal. Key challenges for this task include localizing target locations and navigating to them through visual inputs, and grounding language instructions to visual appearance of objects. To address these challenges, in this study, we augment the agent's field of view during navigation subgoals with multiple viewing angles, and train the agent to predict its relative spatial relation to the target location at each timestep. We also improve language grounding by introducing a pre-trained object detection module to the model pipeline. Empirical studies show that our approach exceeds the baseline model performance.

Qing Ping, Feiyang Niu, Govind Thattai et al.

Current conversational AI systems aim to understand a set of pre-designed requests and execute related actions, which limits them to evolve naturally and adapt based on human interactions. Motivated by how children learn their first language interacting with adults, this paper describes a new Teachable AI system that is capable of learning new language nuggets called concepts, directly from end users using live interactive teaching sessions. The proposed setup uses three models to: a) Identify gaps in understanding automatically during live conversational interactions, b) Learn the respective interpretations of such unknown concepts from live interactions with users, and c) Manage a classroom sub-dialogue specifically tailored for interactive teaching sessions. We propose state-of-the-art transformer based neural architectures of models, fine-tuned on top of pre-trained models, and show accuracy improvements on the respective components. We demonstrate that this method is very promising in leading way to build more adaptive and personalized language understanding models.

Weixin Liang, Feiyang Niu, Aishwarya Reganti et al.

The predominant approach to visual question answering (VQA) relies on encoding the image and question with a "black-box" neural encoder and decoding a single token as the answer like "yes" or "no". Despite this approach's strong quantitative results, it struggles to come up with intuitive, human-readable forms of justification for the prediction process. To address this insufficiency, we reformulate VQA as a full answer generation task, which requires the model to justify its predictions in natural language. We propose LRTA [Look, Read, Think, Answer], a transparent neural-symbolic reasoning framework for visual question answering that solves the problem step-by-step like humans and provides human-readable form of justification at each step. Specifically, LRTA learns to first convert an image into a scene graph and parse a question into multiple reasoning instructions. It then executes the reasoning instructions one at a time by traversing the scene graph using a recurrent neural-symbolic execution module. Finally, it generates a full answer to the given question with natural language justifications. Our experiments on GQA dataset show that LRTA outperforms the state-of-the-art model by a large margin (43.1% v.s. 28.0%) on the full answer generation task. We also create a perturbed GQA test set by removing linguistic cues (attributes and relations) in the questions for analyzing whether a model is having a smart guess with superficial data correlations. We show that LRTA makes a step towards truly understanding the question while the state-of-the-art model tends to learn superficial correlations from the training data.