Michael Johnston

Anna Gottardi, Osman Ipek, Giuseppe Castellucci et al. · amazon-science

Since its inception in 2016, the Alexa Prize program has enabled hundreds of university students to explore and compete to develop conversational agents through the SocialBot Grand Challenge. The goal of the challenge is to build agents capable of conversing coherently and engagingly with humans on popular topics for 20 minutes, while achieving an average rating of at least 4.0/5.0. However, as conversational agents attempt to assist users with increasingly complex tasks, new conversational AI techniques and evaluation platforms are needed. The Alexa Prize TaskBot challenge, established in 2021, builds on the success of the SocialBot challenge by introducing the requirements of interactively assisting humans with real-world Cooking and Do-It-Yourself tasks, while making use of both voice and visual modalities. This challenge requires the TaskBots to identify and understand the user's need, identify and integrate task and domain knowledge into the interaction, and develop new ways of engaging the user without distracting them from the task at hand, among other challenges. This paper provides an overview of the TaskBot challenge, describes the infrastructure support provided to the teams with the CoBot Toolkit, and summarizes the approaches the participating teams took to overcome the research challenges. Finally, it analyzes the performance of the competing TaskBots during the first year of the competition.

Prasoon Goyal, Sattvik Sahai, Michael Johnston et al. · amazon-science

Post-training Large Language Models requires diverse, high-quality data which is rare and costly to obtain, especially in low resource domains and for multi-turn conversations. Common solutions are crowdsourcing or synthetic generation, but both often yield low-quality or low-diversity data. We introduce Adversarial Arena for building high quality conversational datasets by framing data generation as an adversarial task: attackers create prompts, and defenders generate responses. This interactive competition between multiple teams naturally produces diverse and complex data. We validated this approach by conducting a competition with 10 academic teams from top US and European universities, each building attacker or defender bots. The competition, focused on safety alignment of LLMs in cybersecurity, generated 19,683 multi-turn conversations. Fine-tuning an open-source model on this dataset produced an 18.47% improvement in secure code generation on CyberSecEval-Instruct and 29.42% improvement on CyberSecEval-MITRE.

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.

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.

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.

Cat P. Le, Luke Dai, Michael Johnston et al.

Effective evaluation methods remain a significant challenge for research on open-domain conversational dialogue systems. Explicit satisfaction ratings can be elicited from users, but users often do not provide ratings when asked, and those they give can be highly subjective. Post-hoc ratings by experts are an alternative, but these can be both expensive and complex to collect. Here, we explore the creation of automated methods for predicting both expert and user ratings of open-domain dialogues. We compare four different approaches. First, we train a baseline model using an end-to-end transformer to predict ratings directly from the raw dialogue text. The other three methods are variants of a two-stage approach in which we first extract interpretable features at the turn level that capture, among other aspects, user dialogue behaviors indicating contradiction, repetition, disinterest, compliments, or criticism. We project these features to the dialogue level and train a dialogue-level MLP regression model, a dialogue-level LSTM, and a novel causal inference model called counterfactual-LSTM (CF-LSTM) to predict ratings. The proposed CF-LSTM is a sequential model over turn-level features which predicts ratings using multiple regressors depending on hypotheses derived from the turn-level features. As a causal inference model, CF-LSTM aims to learn the underlying causes of a specific event, such as a low rating. We also bin the user ratings and perform classification experiments with all four models. In evaluation experiments on conversational data from the Alexa Prize SocialBot, we show that the CF-LSTM achieves the best performance for predicting dialogue ratings and classification.

Jiachen Li, Qiaozi Gao, Michael Johnston et al.

Prompt-based learning has been demonstrated as a compelling paradigm contributing to large language models' tremendous success (LLMs). Inspired by their success in language tasks, existing research has leveraged LLMs in embodied instruction following and task planning. In this work, we tackle the problem of training a robot to understand multimodal prompts, interleaving vision signals with text descriptions. This type of task poses a major challenge to robots' capability to understand the interconnection and complementarity between vision and language signals. In this work, we introduce an effective framework that learns a policy to perform robot manipulation with multimodal prompts from multi-task expert trajectories. Our methods consist of a two-stage training pipeline that performs inverse dynamics pretraining and multi-task finetuning. To facilitate multimodal understanding, we design our multimodal prompt encoder by augmenting a pretrained LM with a residual connection to the visual input and model the dependencies among action dimensions. Empirically, we evaluate the efficacy of our method on the VIMA-BENCH and establish a new state-of-the-art (10% improvement in success rate). Moreover, we demonstrate that our model exhibits remarkable in-context learning ability. Project page: \url{https://midas-icml.github.io/}.

James F. Mullen, Prasoon Goyal, Robinson Piramuthu et al.

Home robots intend to make their users lives easier. Our work assists in this goal by enabling robots to inform their users of dangerous or unsanitary anomalies in their home. Some examples of these anomalies include the user leaving their milk out, forgetting to turn off the stove, or leaving poison accessible to children. To move towards enabling home robots with these abilities, we have created a new dataset, which we call SafetyDetect. The SafetyDetect dataset consists of 1000 anomalous home scenes, each of which contains unsafe or unsanitary situations for an agent to detect. Our approach utilizes large language models (LLMs) alongside both a graph representation of the scene and the relationships between the objects in the scene. Our key insight is that this connected scene graph and the object relationships it encodes enables the LLM to better reason about the scene -- especially as it relates to detecting dangerous or unsanitary situations. Our most promising approach utilizes GPT-4 and pursues a categorization technique where object relations from the scene graph are classified as normal, dangerous, unsanitary, or dangerous for children. This method is able to correctly identify over 90% of anomalous scenarios in the SafetyDetect Dataset. Additionally, we conduct real world experiments on a ClearPath TurtleBot where we generate a scene graph from visuals of the real world scene, and run our approach with no modification. This setup resulted in little performance loss. The SafetyDetect Dataset and code will be released to the public upon this papers publication.

Barbara Oakley, Michael Johnston, Ken-Zen Chen et al.

In the age of generative AI and ubiquitous digital tools, human cognition faces a structural paradox: as external aids become more capable, internal memory systems risk atrophy. Drawing on neuroscience and cognitive psychology, this paper examines how heavy reliance on AI systems and discovery-based pedagogies may impair the consolidation of declarative and procedural memory -- systems essential for expertise, critical thinking, and long-term retention. We review how tools like ChatGPT and calculators can short-circuit the retrieval, error correction, and schema-building processes necessary for robust neural encoding. Notably, we highlight striking parallels between deep learning phenomena such as "grokking" and the neuroscience of overlearning and intuition. Empirical studies are discussed showing how premature reliance on AI during learning inhibits proceduralization and intuitive mastery. We argue that effective human-AI interaction depends on strong internal models -- biological "schemata" and neural manifolds -- that enable users to evaluate, refine, and guide AI output. The paper concludes with policy implications for education and workforce training in the age of large language models.

Vishnu Sashank Dorbala, Prasoon Goyal, Robinson Piramuthu et al.

We present and tackle the problem of Embodied Question Answering (EQA) with Situational Queries (S-EQA) in a household environment. Unlike prior EQA work tackling simple queries that directly reference target objects and properties ("What is the color of the car?"), situational queries (such as "Is the house ready for sleeptime?") are challenging as they require the agent to correctly identify multiple object-states (Doors: Closed, Lights: Off, etc.) and reach a consensus on their states for an answer. Towards this objective, we first introduce a novel Prompt-Generate-Evaluate (PGE) scheme that wraps around an LLM's output to generate unique situational queries and corresponding consensus object information. PGE is used to generate 2K datapoints in the VirtualHome simulator, which is then annotated for ground truth answers via a large scale user-study conducted on M-Turk. With a high rate of answerability (97.26%) on this study, we establish that LLMs are good at generating situational data. However, in evaluating the data using an LLM, we observe a low correlation of 46.2% with the ground truth human annotations; indicating that while LLMs are good at generating situational data, they struggle to answer them according to consensus. When asked for reasoning, we observe the LLM often goes against commonsense in justifying its answer. Finally, we utilize PGE to generate situational data in a real-world environment, exposing LLM hallucination in generating reliable object-states when a structured scene graph is unavailable. To the best of our knowledge, this is the first work to introduce EQA in the context of situational queries and also the first to present a generative approach for query creation. We aim to foster research on improving the real-world usability of embodied agents through this work.

Sattvik Sahai, Prasoon Goyal, Michael Johnston et al. · amazon-science

AI systems for software development are rapidly gaining prominence, yet significant challenges remain in ensuring their safety. To address this, Amazon launched the Trusted AI track of the Amazon Nova AI Challenge, a global competition among 10 university teams to drive advances in secure AI. In the challenge, five teams focus on developing automated red teaming bots, while the other five create safe AI assistants. This challenge provides teams with a unique platform to evaluate automated red-teaming and safety alignment methods through head-to-head adversarial tournaments where red teams have multi-turn conversations with the competing AI coding assistants to test their safety alignment. Along with this, the challenge provides teams with a feed of high quality annotated data to fuel iterative improvement. Throughout the challenge, teams developed state-of-the-art techniques, introducing novel approaches in reasoning-based safety alignment, robust model guardrails, multi-turn jail-breaking, and efficient probing of large language models (LLMs). To support these efforts, the Amazon Nova AI Challenge team made substantial scientific and engineering investments, including building a custom baseline coding specialist model for the challenge from scratch, developing a tournament orchestration service, and creating an evaluation harness. This paper outlines the advancements made by university teams and the Amazon Nova AI Challenge team in addressing the safety challenges of AI for software development, highlighting this collaborative effort to raise the bar for AI safety.

Casey Kennington, Malihe Alikhani, Heather Pon-Barry et al. · cmu

The ability to interact with machines using natural human language is becoming not just commonplace, but expected. The next step is not just text interfaces, but speech interfaces and not just with computers, but with all machines including robots. In this paper, we chronicle the recent history of this growing field of spoken dialogue with robots and offer the community three proposals, the first focused on education, the second on benchmarks, and the third on the modeling of language when it comes to spoken interaction with robots. The three proposals should act as white papers for any researcher to take and build upon.

Minhua Chen, Badrinath Jayakumar, Padmasundari Gopalakrishnan et al.

Deep models have improved state-of-the-art for both supervised and unsupervised learning. For example, deep embedded clustering (DEC) has greatly improved the unsupervised clustering performance, by using stacked autoencoders for representation learning. However, one weakness of deep modeling is that the local neighborhood structure in the original space is not necessarily preserved in the latent space. To preserve local geometry, various methods have been proposed in the supervised and semi-supervised learning literature (e.g., spectral clustering and label propagation) using graph Laplacian regularization. In this paper, we combine the strength of deep representation learning with measure propagation (MP), a KL-divergence based graph regularization method originally used in the semi-supervised scenario. The main assumption of MP is that if two data points are close in the original space, they are likely to belong to the same class, measured by KL-divergence of class membership distribution. By taking the same assumption in the unsupervised learning scenario, we propose our Deep Embedded Clustering Aided by Measure Propagation (DECAMP) model. We evaluate DECAMP on short text clustering tasks. On three public datasets, DECAMP performs competitively with other state-of-the-art baselines, including baselines using additional data to generate word embeddings used in the clustering process. As an example, on the Stackoverflow dataset, DECAMP achieved a clustering accuracy of 79%, which is about 5% higher than all existing baselines. These empirical results suggest that DECAMP is a very effective method for unsupervised learning.