Social
Conversational AI

A quick guide to Amazon's 40+ papers at EMNLP 2022

Familiar topics like information extraction and question answering share space with robotics and geolocation learning, and query rewriting emerges as a dynamic area of research.

By Staff writer
December 6, 2022
7 min read
Share

Amazon’s more than 40 papers at this year’s Conference on Empirical Methods in Natural-Language Processing (EMNLP) — including papers accepted to EMNLP’s new industry track — cover some familiar topics, such as natural-language understanding and question answering. But they also wander farther afield, taking in such disparate subjects as robotics and geospatial learning — and two of the papers concern pun generation.

Query rewriting, whose applications include self-learning and reference resolution, has emerged as a dynamic area of research at Amazon, with five related papers at this year’s EMNLP. And several papers explore the burgeoning field of prompt engineering, or priming large language models to produce the desired types of output.

Below is a quick guide to Amazon’s EMNLP papers, both academic track and industry track.

Continual learning

Iterative stratified testing and measurement for automated model updates
Elizabeth Dekeyser, Nicholas Comment, Shermin Pei, Rajat Kumar, Shruti Rai, Fengtao Wu, Lisa Haverty, Kanna Shimizu

Towards need-based spoken language understanding model updates: What have we learned?
Quynh Do, Judith Gaspers, Daniil Sorokin, Patrick Lehnen

Unsupervised training data reweighting for natural language understanding with local distribution approximation
Jose Garrido Ramas, Thu Le, Bei Chen, Manoj Kumar, Kay Rottmann

Dialogue

retrieval-based-response.png
"Deploying a retrieval based response model for task oriented dialogues" proposes a model in which cross-attention layers learn the semantic correlations between history, profile features, and candidate responses, and a score function computes and ranks the candidate responses.

Deploying a retrieval based response model for task oriented dialogues
Lahari Poddar, Gyuri Szarvas, Cheng Wang, Georges Balazs, Pavel Danchenko, Patrick Ernst

Dialogue meaning representation for task-oriented dialogue systems
Xiangkun Hu, Junqi Dai, Hang Yan, Yi Zhang, Qipeng Guo, Xipeng Qiu, Zheng Zhang

Injecting domain knowledge in language models for task-oriented dialogue systems
Denis Emelin, Daniele Bonadiman, Sawsan Alqahtani, Yi Zhang, Saab Mansour

Evaluation

GEMv2: Multilingual NLG benchmarking in a single line of code
Sebastian Gehrmann, Abhik Bhattacharjee, Abinaya Mahendiran, Alex Wang, Alexandros Papangelis, Aman Madaan, Angelina McMillan-Major, Anna Shvets, Ashish Upadhyay, Bernd Bohnet, Bingsheng Yao, Bryan Wilie, Chandra Bhagavatula, Chaobin You, Craig Thomson, Cristina Garbacea, Dakuo Wang, Daniel Deutsch, Deyi Xiong, Di Jin, Dimitra Gkatzia, Dragomir Radev, Elizabeth Clark, Esin Durmus, Faisal Ladhak, Filip Ginter, Genta Indra Winata, Hendrik Strobelt, Jekaterina Novikova, Jenna Kanerva, Jenny Chim, Jiawei Zhou, Jordan Clive, Joshua Maynez, João Sedoc, Juraj Juraska, Kaustubh Dhole, Khyathi Raghavi Chandu, Laura Perez-Beltrachini, Leonardo F. R. Ribeiro, Lewis Tunstall, Li Zhang, Mahima Pushkarna, Mathias Creutz, Michael White, Mihir Sanjay Kale, Moussa Kamal Eddine, Nico Daheim, Nishant Subramani, Ondrej Dusek, Paul Pu Liang, Pawan Sasanka Ammanamanch, Qi Zhu, Ratish Puduppully, Reno Kriz, Rifat Shahriyar, Saad Mahamood, Salomey Osei, Samuel Cahyawijaya, Sanja Štajner, Sebastien Montella, Shailza Jolly, Simon Mille, Tianhao Shen, Tosin Adewumi, Vikas Raunak, Vipul Raheja, Vitaly Nikolaev, Vivian Tsai, Yacine Jernite, Ying Xu, Yisi Sang, Yixin Liu, Yufang Hou

Fact verification

Fact checking machine generated text with dependency trees
Alex Estes, Nikhita Vedula, Marcus D. Collins, Matthew Cecil, Oleg Rokhlenko

Fact checking.png
The method proposed in "Fact checking machine generated text with dependency trees" identifies entity attributes from the dependency parse tree of an input claim whose factuality is to be assessed.

Fairness

MT-GenEval: A counterfactual and contextual dataset for evaluating gender accuracy in machine translation
Anna Currey, Maria Nadejde, Raghavendra Pappagari, Mia Mayer, Stanislas Lauly, Xing Niu, Benjamin Hsu, Georgiana Dinu

Humor

Context-situated pun generation
Jiao Sun, Anjali Narayan-Chen, Shereen Oraby, Shuyang Gao, Tagyoung Chung, Jing Huang, Yang Liu, Nanyun Peng

ExPUNations: Augmenting puns with keywords and explanations
Jiao Sun, Anjali Narayan-Chen, Shereen Oraby, Alessandra Cervone, Tagyoung Chung, Jing Huang, Yang Liu, Nanyun Peng

Information extraction

A hybrid approach to cross-lingual product review summarization
Saleh Soltan, Victor Soto, Ke Tran, Wael Hamza

Ask-and-Verify: Span candidate generation and verification for attribute value extraction
Yifan Ding, Yan Liang, Nasser Zalmout, Xian Li, Christan Grant, Tim Weninger

DORE: Document ordered relation extraction based on generative framework
Qipeng Guo, Yuqing Yang, Hang Yan, Xipeng Qiu, Zheng Zhang

DORE.png
The method proposed in "DORE: Document ordered relation extraction based on generative framework" identifies multiple instances of the same entity in an input document and builds a relation matrix that records relations between entities.

Learning to revise references for faithful summarization
Griffin Adams, Han-Chin Shing, Qing Sun, Christopher Winestock, Kathleen McKeown, Noémie Elhadad

Prototype-representations for training data filtering in weakly-supervised information extraction
Nasser Zalmout, Xian Li

Information retrieval

Accelerating learned sparse indexes via term impact decomposition
Joel Mackenzie, Antonio Mallia, Alistair Moffat, Matthias Petri

Machine translation impact in e-commerce multilingual search
Bryan Zhang, Amita Misra

Knowledge distillation

Distilling multilingual transformers into CNNs for scalable intent classification
Besnik Fetahu, Akash Veeragouni, Oleg Rokhlenko, Shervin Malmasi

Knowledge distillation transfer sets and their impact on downstream NLU tasks
Charith Peris, Lizhen Tan, Thomas Gueudre, Turan Gojayev, Pan Wei, Gokmen Oz

Machine learning

Calibrating imbalanced classifiers with focal loss: An empirical study
Cheng Wang, Georges Balazs, Gyuri Szarvas, Patrick Ernst, Lahari Poddar, Pavel Danchenko

Model adaptation

Meta-learning the difference: Preparing large language models for efficient adaptation
Zejiang Hou, Julian Salazar, George Polovets

Open world.png
In "Open world classification with adaptive negative samples", Amazon researchers propose a new method for discriminating known and open (unknown) categories of data. This figure compares their approach (d) to ordinary supervised learning (a) and an adaptive-decision-boundary method (c).

Open world classification with adaptive negative samples
Ke Bai, Guoyin Wang, Jiwei Li, Sunghyun Park, Sungjin Lee, Puyang Xu, Ricardo Henao, Lawrence Carin

Multimodal interaction

Multimodal context carryover
Prashan Wanigasekara, Nalin Gupta, Fan Yang, Emre Barut, Zeynab Raeesy, Kechen Qin, Stephen Rawls, Xinyue Liu, Chengwei Su, Spurthi Sandiri

Natural-language processing

McPhraSy: Multi context phrase similarity and clustering
Amir DN Cohen, Hila Gonen, Ori Shapira, Ran Levy, Yoav Goldberg

Unsupervised syntactically controlled paraphrase generation with abstract meaning representations
Kuan-Hao Huang, Varun Iyer, Anoop Kumar, Sriram Venkatapathy, Kai-Wei Chang, Aram Galstyan

Natural-language understanding

Improving large-scale conversational assistants using model interpretation based training sample selection
Stefan Schroedl, Manoj Kumar, Kiana Hajebi, Morteza Ziyadi, Sriram Venkatapathy, Anil Ramakrishna, Rahul Gupta, Pradeep Natarajan

Improving text-to-SQL semantic parsing with fine-grained query understanding
Jun Wang, Patrick Ng, Alexander Hanbo Li, Jiarong Jiang, Zhiguo Wang, Ramesh Nallapati, Bing Xiang, Sudipta Sengupta

Learning geolocations for cold-start and hard-to-resolve addresses via deep metric learning
Govind, Saurabh Sohoney

Geolocation.png
"Learning geolocations for cold-start and hard-to-resolve addresses via deep metric learning" proposes a way to use deep metric learning on addresses to capture geospatial distance semantics.

Prompt engineering

DynaMaR: Dynamic prompt with mask token representation
Xiaodi Sun, Sunny Rajagopalan, Priyanka Nigam, Weiyi Lu, Yi Xu, Iman Keivanloo, Belinda Zeng, Trishul Chilimbi

Inducer-tuning: Connecting prefix-tuning and adapter-tuning
Yifan Chen, Devamanyu Hazarika, Mahdi Namazifar, Yang Liu, Di Jin, Dilek Hakkani-Tür

Query rewriting

CGF.png
In "CGF: Constrained generation framework for query rewriting in conversational AI", Amazon researchers use tries — trees in which each node extends a text by one word — to constrain the outputs of a model that generates query rewrites.

CGF: Constrained generation framework for query rewriting in conversational AI
Jie Hao, Yang Liu, Xing Fan, Saurabh Gupta, Saleh Soltan, Rakesh Chada, Pradeep Natarajan, Edward Guo, Gokhan Tur

CycleKQR: Unsupervised bidirectional keyword question rewriting
Andrea Iovine, Anjie Fang, Besnik Fetahu, Jie Zhao, Oleg Rokhlenko, Shervin Malmasi

PAIGE: Personalized adaptive interactions graph encoder for query rewriting in dialogue systems
Daniel Bis, Saurabh Gupta, Jie Hao, Xing Fan, Edward Guo

PENTATRON: PErsonalized coNText-aware transformer for retrieval-based cOnversational uNderstanding
Niranjan Uma Naresh, Ziyan Jiang, Ankit, Sungjin Lee, Jie Hao, Xing Fan, Edward Guo

Reinforced question rewriting for conversational question answering
Zhiyu Chen, Jie Zhao, Anjie Fang, Besnik Fetahu, Oleg Rokhlenko, Shervin Malmasi

Question answering

Ensemble transformer for efficient and accurate ranking tasks: An application to question answering systems
Yoshitomo Matsubara, Luca Soldaini, Eric Lind, Alessandro Moschitti

FocusQA: Open-domain question answering with a context in focus
Gianni Barlacchi, Ivano Lauriola, Alessandro Moschitti, Marco Del Tredici, Xiaoyu Shen, Thuy Vu, Bill Byrne, Adrià de Gispert

Knowledge transfer from answer ranking to answer generation
Matteo Gabburo, Rik Koncel-Kedziorski, Siddhant Garg, Luca Soldaini, Alessandro Moschitti

Pre-training transformer models with sentence-level objectives for answer sentence selection
Luca Di Liello, Siddhant Garg, Luca Soldaini, Alessandro Moschitti

RLET: A reinforcement learning based approach for explainable QA with entailment trees
Tengxiao Liu, Qipeng Guo, Xiangkun Hu, Yue Zhang, Xipeng Qiu, Zheng Zhang

Robotics

ALFRED-L: Investigating the role of language for action learning in interactive visual environments
Arjun R. Akula, Spandana Gella, Aishwarya Padmakumar, Mahdi Namazifar, Mohit Bansal, Jesse Thomason, Dilek Hakkani-Tür

ALFRED-L.png
"ALFRED-L: Investigating the role of language for action learning in interactive visual environments" proposes a new test split to the ALFRED benchmark for embodied-task completion. The test split — ALFRED-L — includes instructions that an agent backtrack to known reference positions along its trajectory, to evaluate whether it can remember their locations.

Research areas
Tags
About the Author
Staff writer

Related content

Work with us

See more jobs See more jobs
Member of Technical Staff, ML Infra, AGI
US, CA, San Francisco
Are you interested in a unique opportunity to advance the accuracy and efficiency of Artificial General Intelligence (AGI) systems? If so, you're at the right place! We are the AGI Autonomy organization, and we are looking for a driven and talented Member of Technical Staff to join us to build state-of-the art agents. As an MTS on our team, you will design, build, and maintain a Spark-based infrastructure to process and manage large datasets critical for machine learning research. You’ll work closely with our researchers to develop data workflows and tools that streamline the preparation and analysis of massive multimodal datasets, ensuring efficiency and scalability. We operate at Amazon's large scale with the energy of a nimble start-up. If you have a learner's mindset, enjoy solving challenging problems and value an inclusive and collaborative team culture, you will thrive in this role, and we hope to hear from you. Key job responsibilities * Develop and maintain reliable infrastructure to enable large-scale data extraction and transformation. * Work closely with researchers to create tooling for emerging data-related needs. * Manage project prioritization, deliverables, timelines, and stakeholder communication. * Illuminate trade-offs, educate the team on best practices, and influence technical strategy. * Operate in a dynamic environment to deliver high quality software.
Applied Scientist II, Last Mile Science
IN, KA, Bangalore
Have you ever ordered a product on Amazon and when that box with the smile arrived you wondered how it got to you so fast? Have you wondered where it came from and how much it cost Amazon to deliver it to you? If so, the WW Amazon Logistics, Business Analytics team is for you. We manage the delivery of tens of millions of products every week to Amazon’s customers, achieving on-time delivery in a cost-effective manner. We are looking for an enthusiastic, customer obsessed, Applied Scientist with good analytical skills to help manage projects and operations, implement scheduling solutions, improve metrics, and develop scalable processes and tools. The primary role of an Operations Research Scientist within Amazon is to address business challenges through building a compelling case, and using data to influence change across the organization. This individual will be given responsibility on their first day to own those business challenges and the autonomy to think strategically and make data driven decisions. Decisions and tools made in this role will have significant impact to the customer experience, as it will have a major impact on how the final phase of delivery is done at Amazon. Candidates will be a high potential, strategic and analytic graduate with a PhD in (Operations Research, Statistics, Engineering, and Supply Chain) ready for challenging opportunities in the core of our world class operations space. Great candidates have a history of operations research, and the ability to use data and research to make changes. This role requires robust program management skills and research science skills in order to act on research outcomes. This individual will need to be able to work with a team, but also be comfortable making decisions independently, in what is often times an ambiguous environment. Responsibilities may include: - Develop input and assumptions based preexisting models to estimate the costs and savings opportunities associated with varying levels of network growth and operations - Creating metrics to measure business performance, identify root causes and trends, and prescribe action plans - Managing multiple projects simultaneously - Working with technology teams and product managers to develop new tools and systems to support the growth of the business - Communicating with and supporting various internal stakeholders and external audiences
Principal Applied Scientist, Amazon
US, NY, New York
Amazon is investing heavily in building a world class advertising business and we are responsible for defining and delivering a collection of self-service performance advertising products that drive discovery and sales. Our products are strategically important to our Retail and Marketplace businesses driving long term growth. We deliver billions of ad impressions and millions of clicks daily and are breaking fresh ground to create world-class products. We are highly motivated, collaborative and fun-loving with an entrepreneurial spirit and bias for action. With a broad mandate to experiment and innovate, we are growing at an unprecedented rate with a seemingly endless range of new opportunities. The Ad Response Prediction team in the Sponsored Products organization builds GenAI-based shopper understanding and audience targeting systems, along with advanced deep-learning models for Click-through Rate (CTR) and Conversion Rate (CVR) predictions. We develop large-scale machine-learning (ML) pipelines and real-time serving infrastructure to match shoppers' intent with relevant ads across all devices, contexts, and marketplaces. Through precise estimation of shoppers' interactions with ads and their long-term value, we aim to drive optimal ad allocation and pricing, helping to deliver a relevant, engaging, and delightful advertising experience to Amazon shoppers. As our business grows and we undertake increasingly complex initiatives, we are looking for entrepreneurial, and self-driven science leaders to join our team. Key job responsibilities As a Principal Applied Scientist in the team, you will: * Seek to understand in depth the Sponsored Products offering at Amazon and identify areas of opportunities to grow our business via principled ML solutions. * Mentor and guide the applied scientists in our organization and hold us to a high standard of technical rigor and excellence in ML. * Design and lead organization wide ML roadmaps to help our Amazon shoppers have a delightful shopping experience while creating long term value for our sellers. * Work with our engineering partners and draw upon your experience to meet latency and other system constraints. * Identify untapped, high-risk technical and scientific directions, and simulate new research directions that you will drive to completion and deliver. * Be responsible for communicating our ML innovations to the broader internal & external scientific community.
Applied Scientist, Workforce Intelligence
CA, BC, Vancouver
Do you want a role with deep meaning and the ability to make a major impact? As part of Intelligent Talent Acquisition (ITA), you'll have the opportunity to reinvent the hiring process and deliver unprecedented scale, sophistication, and accuracy for Amazon Talent Acquisition operations. ITA is an industry-leading people science and technology organization made up of scientists, engineers, analysts, product professionals and more, all with the shared goal of connecting the right people to the right jobs in a way that is fair and precise. Last year we delivered over 6 million online candidate assessments, and helped Amazon deliver billions of packages around the world by making it possible to hire hundreds of thousands of workers in the right quantity, at the right location and at exactly the right time. You’ll work on state-of-the-art research, advanced software tools, new AI systems, and machine learning algorithms, leveraging Amazon's in-house tech stack to bring innovative solutions to life. Join ITA in using technologies to transform the hiring landscape and make a meaningful difference in people's lives. Together, we can solve the world's toughest hiring problems. Global Hiring Science owns and develops products and services using Artificial Intelligence and Machine Learning (ML) that enhance recruitment. We collaborate with scientists to build and maintain machine learning solutions for hiring, offering opportunities to both apply and develop ML engineering skills in a production environment. Key job responsibilities • Design and implement advanced AI models using the latest LLM and GenAI technologies to develop fair and accurate machine learning models for hiring. • Clearly and cogently present your work and ideas, and respond effectively to feedback. • Collaborate with cross-functional teams with Research Scientists and Software Engineers to integrate AI-driven products into Amazon’s hiring process. • Stay at the advance of AI research, continuously exploring and implementing new techniques in NLP, LLMs, and GenAI to drive innovation in hiring. • Implement advanced natural language processing models to extract insights from diverse data sources. • Ensure effective teamwork, communication, collaboration, and commitment across multiple teams with competing priorities. • Contribute to the scientific community through publications, presentations, and collaborations with academic institutions. About the team The mission of Global Hiring Science (GHS) is to improve both the efficiency and effectiveness of hiring across Amazon with assessments and interview improvements. We are a team of experts in machine learning, industrial-organizational psychology, data science, and measuring the knowledge, skills, and abilities that it takes to be successful at Amazon.
Member of Technical Staff, AGI Autonomy
US, CA, San Francisco
Amazon has launched a new research lab in San Francisco to develop foundational capabilities for useful AI agents. We’re enabling practical AI to make our customers more productive, empowered, and fulfilled. In particular, our work combines large language models (LLMs) with reinforcement learning (RL) to solve reasoning, planning, and world modeling in both virtual and physical environments. Our research builds on that of Amazon’s broader AGI organization, which recently introduced Amazon Nova, a new generation of state-of-the-art foundation models (FMs). Our lab is a small, talent-dense team with the resources and scale of Amazon. Each team in the lab has the autonomy to move fast and the long-term commitment to pursue high-risk, high-payoff research. We’re entering an exciting new era where agents can redefine what AI makes possible. We’d love for you to join our lab and build it from the ground up! Key job responsibilities You will contribute directly to AI agent development in an applied research role, including model training, dataset design, and pre- and post-training optimization. You will be hired as a Member of Technical Staff.
Economist, People Experience and Technology Central Science
US, WA, Seattle
PXTCS is looking for an economist who can apply economic methods to address business problems. The ideal candidate will work with engineers and computer scientists to estimate models and algorithms on large scale data, design pilots and measure impact, and transform successful prototypes into improved policies and programs at scale. PXTCS is looking for creative thinkers who can combine a strong technical economic toolbox with a desire to learn from other disciplines, and who know how to execute and deliver on big ideas as part of an interdisciplinary technical team. Ideal candidates will work in a team setting with individuals from diverse disciplines and backgrounds. They will work with teammates to develop scientific models and conduct the data analysis, modeling, and experimentation that is necessary for estimating and validating models. They will work closely with engineering teams to develop scalable data resources to support rapid insights, and take successful models and findings into production as new products and services. They will be customer-centric and will communicate scientific approaches and findings to business leaders, listening to and incorporate their feedback, and delivering successful scientific solutions. A day in the life The Economist will work with teammates to apply economic methods to business problems. This might include identifying the appropriate research questions, writing code to implement a DID analysis or estimate a structural model, or writing and presenting a document with findings to business leaders. Our economists also collaborate with partner teams throughout the process, from understanding their challenges, to developing a research agenda that will address those challenges, to help them implement solutions. About the team The People eXperience and Technology Central Science (PXTCS) team uses economics, behavioral science, statistics, and machine learning to proactively identify mechanisms and process improvements which simultaneously improve Amazon and the lives, wellbeing, and the value of work to Amazonians. PXTCS is an interdisciplinary team that combines the talents of science and engineering to develop and deliver solutions that measurably achieve this goal.
Executive Assistant, AGI
US, CA, San Francisco
The Amazon General Intelligence “AGI” organization is looking for an Executive Assistant to support leaders of our Autonomy Team in our growing AI Lab space located in San Francisco. This role is ideal for exceptionally talented, dependable, customer-obsessed, and self-motivated individuals eager to work in a fast paced, exciting and growing team. This role serves as a strategic business partner, managing complex executive operations across the AGI organization. The position requires superior attention to detail, ability to meet tight deadlines, excellent organizational skills, and juggling multiple critical requests while proactively anticipating needs and driving improvements. High integrity, discretion with confidential information, and professionalism are essential. The successful candidate will complete complex tasks and projects quickly with minimal guidance, react with appropriate urgency, and take effective action while navigating ambiguity. Flexibility to change direction at a moment's notice is critical for success in this role. Key job responsibilities - Serve as strategic partner to senior leadership, identifying opportunities to improve organizational effectiveness and drive operational excellence - Manage complex calendars and scheduling for multiple executives - Drive continuous improvement through process optimization and new mechanisms - Coordinate team activities including staff meetings, offsites, and events - Schedule and manage cost-effective travel - Attend key meetings, track deliverables, and ensure timely follow-up - Create expense reports and manage budget tracking - Serve as liaison between executives and internal/external stakeholders - Build collaborative relationships with Executive Assistants across the company and with critical external partners - Help us build a great team culture in the SF Lab!
Applied Scientist, FAR (Frontier AI & Robotics)
US, CA, San Francisco
Join the next revolution in robotics at Amazon's Frontier AI & Robotics team, where you'll work alongside world-renowned AI pioneers to push the boundaries of what's possible in robotic intelligence. As an Applied Scientist, you'll be at the forefront of developing breakthrough foundation models that enable robots to perceive, understand, and interact with the world in unprecedented ways. You'll drive independent research initiatives in areas such as perception, manipulation, science understanding, locomotion, manipulation, sim2real transfer, multi-modal foundation models and multi-task robot learning, designing novel frameworks that bridge the gap between state-of-the-art research and real-world deployment at Amazon scale. In this role, you'll balance innovative technical exploration with practical implementation, collaborating with platform teams to ensure your models and algorithms perform robustly in dynamic real-world environments. You'll have access to Amazon's vast computational resources, enabling you to tackle ambitious problems in areas like very large multi-modal robotic foundation models and efficient, promptable model architectures that can scale across diverse robotic applications. Key job responsibilities - Drive independent research initiatives across the robotics stack, including robotics foundation models, focusing on breakthrough approaches in perception, and manipulation, for example open-vocabulary panoptic scene understanding, scaling up multi-modal LLMs, sim2real/real2sim techniques, end-to-end vision-language-action models, efficient model inference, video tokenization - Design and implement novel deep learning architectures that push the boundaries of what robots can understand and accomplish - Lead full-stack robotics projects from conceptualization through deployment, taking a system-level approach that integrates hardware considerations with algorithmic development, ensuring robust performance in production environments - Collaborate with platform and hardware teams to ensure seamless integration across the entire robotics stack, optimizing and scaling models for real-world applications - Contribute to the team's technical strategy and help shape our approach to next-generation robotics challenges A day in the life - Design and implement novel foundation model architectures and innovative systems and algorithms, leveraging our extensive infrastructure to prototype and evaluate at scale - Collaborate with our world-class research team to solve complex technical challenges - Lead technical initiatives from conception to deployment, working closely with robotics engineers to integrate your solutions into production systems - Participate in technical discussions and brainstorming sessions with team leaders and fellow scientists - Leverage our massive compute cluster and extensive robotics infrastructure to rapidly prototype and validate new ideas - Transform theoretical insights into practical solutions that can handle the complexities of real-world robotics applications About the team At Frontier AI & Robotics, we're not just advancing robotics – we're reimagining it from the ground up. Our team is building the future of intelligent robotics through innovative foundation models and end-to-end learned systems. We tackle some of the most challenging problems in AI and robotics, from developing sophisticated perception systems to creating adaptive manipulation strategies that work in complex, real-world scenarios. What sets us apart is our unique combination of ambitious research vision and practical impact. We leverage Amazon's massive computational infrastructure and rich real-world datasets to train and deploy state-of-the-art foundation models. Our work spans the full spectrum of robotics intelligence – from multimodal perception using images, videos, and sensor data, to sophisticated manipulation strategies that can handle diverse real-world scenarios. We're building systems that don't just work in the lab, but scale to meet the demands of Amazon's global operations. Join us if you're excited about pushing the boundaries of what's possible in robotics, working with world-class researchers, and seeing your innovations deployed at unprecedented scale.
Sr. Applied Scientist, FAR (Frontier AI & Robotics)
US, CA, San Francisco
Join the next revolution in robotics at Amazon's Frontier AI & Robotics team, where you'll work alongside world-renowned AI pioneers to push the boundaries of what's possible in robotic intelligence. As a Senior Applied Scientist, you'll spearhead the development of breakthrough foundation models and full-stack robotics systems that enable robots to perceive, understand, and interact with the world in unprecedented ways. You'll drive technical excellence in areas such as perception, manipulation, science understanding, locomotion, manipulation, sim2real transfer, multi-modal foundation models and multi-task robot learning, designing novel frameworks that bridge the gap between state-of-the-art research and real-world deployment at Amazon scale. In this role, you'll combine hands-on technical work with scientific leadership, ensuring your team delivers robust solutions for dynamic real-world environments. You'll leverage Amazon's vast computational resources to tackle ambitious problems in areas like very large multi-modal robotic foundation models and efficient, promptable model architectures that can scale across diverse robotic applications. Key job responsibilities - Lead technical initiatives across the robotics stack, driving breakthrough approaches through hands-on research and development in areas including robotics foundation models, focusing on breakthrough approaches in perception, and manipulation, for example open-vocabulary panoptic scene understanding, scaling up multi-modal LLMs, sim2real/real2sim techniques, end-to-end vision-language-action models, efficient model inference, video tokenization - Guide technical direction for full-stack robotics projects from conceptualization through deployment, taking a system-level approach that integrates hardware considerations with algorithmic development, ensuring robust performance in production environments - Mentor fellow scientists while maintaining strong individual technical contributions - Collaborate with platform and hardware teams to ensure seamless integration across the entire robotics stack - Influence technical decisions and implementation strategies within your area of focus A day in the life - Design and implement novel foundation model architectures and innovative systems and algorithms, leveraging our extensive infrastructure to prototype and evaluate at scale - Guide fellow scientists in solving complex technical challenges across the full robotics stack - Lead focused technical initiatives from conception through deployment, ensuring successful integration with production systems - Drive technical discussions within your team and with key stakeholders - Conduct experiments and prototype new ideas using our massive compute cluster and extensive robotics infrastructure - Mentor team members while maintaining significant hands-on contribution to technical solutions About the team At Frontier AI & Robotics, we're not just advancing robotics – we're reimagining it from the ground up. Our team is building the future of intelligent robotics through innovative foundation models and end-to-end learned systems. We tackle some of the most challenging problems in AI and robotics, from developing sophisticated perception systems to creating adaptive manipulation strategies that work in complex, real-world scenarios. What sets us apart is our unique combination of ambitious research vision and practical impact. We leverage Amazon's massive computational infrastructure and rich real-world datasets to train and deploy state-of-the-art foundation models. Our work spans the full spectrum of robotics intelligence – from multimodal perception using images, videos, and sensor data, to sophisticated manipulation strategies that can handle diverse real-world scenarios. We're building systems that don't just work in the lab, but scale to meet the demands of Amazon's global operations. Join us if you're excited about pushing the boundaries of what's possible in robotics, working with world-class researchers, and seeing your innovations deployed at unprecedented scale.
Member of Technical Staff, Data Platform, AGI
US, CA, San Francisco
Amazon AGI Autonomy develops foundational capabilities for useful AI agents. We are the research lab behind Amazon Nova Act, a state-of-the-art computer-use agent. Our work combines Large Language Models (LLMs) with Reinforcement Learning (RL) to solve reasoning, planning, and world modeling in the virtual world. We are a small, talent-dense team with the autonomy to move fast and the long-term commitment to pursue high-risk, high-payoff research. Come be a part of our journey! --- About the team We’re looking for a generalist software engineer to build and evolve our internal data platform. The team builds data-intensive services that ingest, process, store, and distribute multi-modal training data across multiple internal and external sources. This work emphasizes data integrity, reliability, and extensibility in support of large-scale training and experimentation workloads. The team also builds and maintains APIs and SDKs that enable product engineers and researchers to build on top of the platform. As research directions change, so does our data, and today the team is focused on hardening the platform to reliably deliver an evolving set of data schemas, sources, and modalities. By building strong foundations and durable abstractions, we aim to enable new kinds of tooling and workflows over time. The team will play a key role in shaping them as the research evolves. --- Key job responsibilities * Build and operate reliable, performant backend and data platform services that support continuous ingestion and use of multi-modal training data. * Identify and implement opportunities to accelerate data generation, validation, and usage across training and evaluation workflows from multiple internal and external sources. * Partner closely with Human Feedback, Data Generation, Product Engineering, and Research teams to evolve and scale the data platform, APIs, and SDKs. * Own projects end to end, from technical design and implementation through deployment, observability, and long-term maintainability. * Write clear technical documentation and communicate design decisions and tradeoffs to stakeholders across multiple teams. * Raise the team’s technical aptitude through thoughtful code reviews, knowledge sharing, and mentorship.
See more jobs
View from space of a connected network around planet Earth representing the Internet of Things.
Get more from Amazon Science
Subscribe to our monthly newsletter

Amazon.com | Conditions of Use | Privacy | © 1996-2026 Amazon.com, Inc. or its affiliates

Social