Social
Conversational AI

A quick guide to Amazon’s 50-plus papers at EMNLP 2024

Large language models predominate, both as a research subject themselves and as tools for researching topics of particular interest to Amazon, such as speech, recommendations, and information retrieval.

By Staff writer
November 14, 2024
7 min read
Share

Large language models (LLMs) have come to dominate the field of natural-language processing, so it’s no surprise that they also dominate the research that Amazon scientists are presenting at this year’s Conference on Empirical Methods in Natural-Language Processing (EMNLP). LLM training is the topic with the greatest number of Amazon papers, followed closely by strategies for mitigating misinformation in LLMs’ outputs — including but not limited to hallucinations. At the same time, a number of papers apply LLMs to topics of traditional interest at Amazon, such as speech, recommender systems, and information retrieval. (Papers marked with asterisks were accepted to Findings of EMNLP.)

AI agents

MARCO: Multi-agent real-time chat orchestration
Anubhav Shrimal, Shervin Malmasi, Kriti Biswas, Swarnalatha Raghuraman, Anish Nediyanchath, Yi Zhang, Promod Yenigalla

Code generation

CodeFort: Robust training for code generation models
Yuhao Zhang, Shiqi Wang, Haifeng Qian, Zijian Wang, Mingyue Shang, Linbo Liu, Sanjay Krishna Gouda, Baishakhi Ray, Murali Krishna Ramanathan, Xiaofei Ma, Anoop Deoras

Socratic human feedback (SoHF): Expert steering strategies for LLM code generation
Subramanian Chidambaram, Erran Li, Min Bai, Xiaopeng LI, Kaixiang Lin, Xiong Zhou, Alex C. Williams

Structured object language modeling (SoLM): Native structured objects generation conforming to complex schemas with self-supervised denoising
Amir Tavanaei, Kee Kiat Koo, Hayreddin Ceker, Shaobai Jiang, Qi Li, Julien Han, Karim Bouyarmane

Contrastive decoding

Explaining and improving contrastive decoding by extrapolating the probabilities of a huge and hypothetical LM
Haw-Shiuan Chang, Nanyun Peng, Mohit Bansal, Anil Ramakrishna, Tagyoung Chung

Explaining and improving contrastive decoding by extrapolating the probabilities of a huge and hypothetical LM.png
Given a simple question with clues, contrastive decoding could have an “obvious blindness” (e.g., assigning higher probability to an uncommon answer, such as "invertebrate", than to the most obvious answer, "bees"). In contrast, the asymptotic probability decoding proposed in "Explaining and improving contrastive decoding by extrapolating the probabilities of a huge and hypothetical LM" correctly assigns the highest probability to "bees" by leveraging the probabilities from multiple LMs of different sizes.

Data integration

ASTRA: Automatic schema matching using machine translation
Tarang Chugh, Deepak Zambre

Learning from natural language explanations for generalizable entity matching
Somin Wadhwa, Adit Krishnan, Runhui Wang, Byron C. Wallace, Chris (Luyang) Kong

Pretraining and finetuning language models on geospatial networks for accurate address matching
Saket Maheshwary, Arpan Paul, Saurabh Sohoney

Retrieval augmented spelling correction for e-commerce applications
Xuan Guo, Rohit Patki, Dante Everaert, Christopher Potts

Dataset distillation

Textual dataset distillation via language model embedding
Yefan Tao, Chris (Luyang) Kong, Andrey Kan, Laurent Callot

Textual dataset distillation via language model embedding: DaLLME.png
The DaLLME framework proposed in "Textual dataset distillation via language model embedding" begins by using a language model to transform raw textual data into embedding vectors. A set of distilled vectors is then derived in the embedding space, through a process designed to encapsulate maximum informational content. Finally, the vec2text model translates these distilled vectors back into textual form.

Document understanding

DocKD: Knowledge distillation from LLMs for open-world document understanding models
Sungnyun Kim, Haofu Liao, Srikar Appalaraju, Peng Tang, Zhuowen Tu, Ravi Kumar Satzoda, R. Manmatha, Vijay Mahadevan, Stefano Soatto

Information retrieval

Evaluating D-MERIT of partial-annotation on information retrieval
Royi Rassin, Yaron Fairstein, Oren Kalinsky, Guy Kushilevitz, Nachshon Cohen, Alexander Libov, Yoav Goldberg

Identifying high consideration e-commerce search queries
Zhiyu Chen, Jason Choi, Besnik Fetahu, Shervin Malmasi

Learning when to retrieve, what to rewrite, and how to respond in conversational QA*
Nirmal Roy, Leonardo Ribeiro, Rexhina Blloshmi, Kevin Small

Natural-language understanding

Intent detection in the age of LLMs
Gaurav Arora, Shreya Jain, Srujana Merugu

Intent detection in the age of LLMs.png
"Intent detection in the age of LLMs" proposes a methodology for adaptive in-context learning and chain-of-thought-based intent detection using LLMs.

Predicting entity salience in extremely short documents
Ben Bullough, Harrison Lundberg, Chen Hu, Weihang Xiao

LLM evaluation

AXCEL: Automated eXplainable consistency evaluation using LLMs*
P Aditya Sreekar, Sahil Verma, Suransh Chopra, Sarik Ghazarian, Abhishek Persad, Narayanan Sadagopan

Precise model benchmarking with only a few observations
Riccardo Fogliato, Pratik Patil, Nil-Jana Akpinar, Mathew Monfort

LLM fine tuning

AdaZeta: Adaptive zeroth-order tensor-train adaption for memory-efficient large language models fine-tuning
Yifan Yang, Kai Zhen, Ershad Banijamali, Thanasis Mouchtaris, Zheng Zhang

RoseLoRA: Row and column-wise sparse low-rank adaptation of pre-trained language model for knowledge editing and fine-tuning
Haoyu Wang, Tianci Liu, Ruirui Li, Monica Cheng, Tuo Zhao, Jing Gao

RoseLoRA.png
The row- and column-wise sparse low-rank adaptation (RoseLoRA) framework proposed in "RoseLoRA: Row and column-wise sparse low-rank adaptation of pre-trained language model for knowledge editing and fine-tuning".

LLMs for speech

Speechworthy instruction-tuned language models
Hyundong Cho, Nicolaas Jedema, Leonardo Ribeiro, Karishma Sharma, Pedro Szekely, Alessandro Moschitti, Ruben Janssen, Jonathan May

LLM misinformation mitigation

ECON: On the detection and resolution of evidence conflicts
Cheng Jiayang, Chunkit Chan, Qianqian Zhuang, Lin Qiu, Tianhang Zhang, Tengxiao Liu, Yangqiu Song, Yue Zhang, Pengfei Liu, Zheng Zhang

Generative subgraph retrieval for knowledge graph–grounded dialog generation
Jinyoung Park, Minseok Joo, Joo-Kyung Kim, Hyunwoo J. Kim

HalluMeasure: Fine-grained hallucination measurement using chain-of-thought reasoning
Shayan Ali Akbar, Md Mosharaf Hossain, Tess Wood, Si-Chi Chin, Erica Salinas, Victor Alvarez, Erwin Cornejo

Knowledge-centric hallucination detection
Xiangkun Hu, Dongyu Ru, Lin Qiu, Qipeng Guo, Tianhang Zhang, Yang Xu, Yun Luo, Pengfei Liu, Zheng Zhang, Yue Zhang

LLM reasoning

Auto-evolve: Enhancing large language model’s performance via self-reasoning framework*
Krishna Aswani, Alex Lu, Pranav Patankar, Priya Dhalwani, Iris Tan, Jayant Ganeshmohan, Simon Lacasse

LLM self-correction

LLM self-correction with DeCRIM: Decompose, critique, and refine for enhanced following of instructions with multiple constraints
Thomas Palmeira Ferraz, Kartik Mehta, Yu-Hsiang Lin, Haw-Shiuan Chang, Shereen Oraby, Sijia Liu, Vivek Subramanian, Tagyoung Chung, Mohit Bansal, Nanyun Peng

DeCRIM.png
In the DeCRIM pipeline proposed in "LLM self-correction with DeCRIM: Decompose, critique, and refine for enhanced following of instructions with multiple constraints", an LLM first generates a response to a user request. The Decomposer then breaks down the request into granular constraints, and the Critic model gives feedback on whether the response meets those constraints. If it does, the response is output; if not, the LLM uses the feedback to refine the response.

LLM training

Dancing in chains: Reconciling instruction following and faithfulness in language models
Zhengxuan Wu, Yuhao Zhang, Peng Qi, Yumo Xu, Rujun Han, Yian Zhang, Jifan Chen, Bonan Min, Zhiheng Huang

DEM: Distribution edited model for training with mixed data distributions
Dhananjay Ram, Aditya Rawal, Momchil Hardalov, Nikolaos Pappas, Sheng Zha

DEM: Distribution Edited Model for Training with Mixed Data Distributions
The distribution-edited model (ΘD) described in "DEM: Distribution edited model for training with mixed data distributions" results from fine-tuning a pretrained model (Θ) on n individual data distributions (Di) and combining the resulting models with basic element-wise vector operations. Here, the extracted distribution vectors (∆ΘDi ) are multiplied by weight coefficients, and the weighted sum is added to the base model.

Evolutionary contrastive distillation for language model alignment
Julian Katz-Samuels, Zheng Li, Hyokun Yun, Priyanka Nigam, Yi Xu, Vaclav Petricek, Bing Yin, Trishul Chilimbi

Hop, skip, jump to convergence: Dynamics of learning rate transitions for improved training of large language models
Shreyas Subramanian, Vignesh Ganapathiraman, Corey Barrett

Learning from relevant subgoals in successful dialogs using iterative training for task-oriented dialog systems
Magdalena Kaiser, Patrick Ernst, Gyuri Szarvas

Quality matters: Evaluating synthetic data for tool-using LLMs
Shadi Iskander, Nachshon Cohen, Zohar Karnin, Ori Shapira, Sofia Tolmach

Query autocompletion

AmazonQAC: A large-scale, naturalistic query autocomplete dataset
Dante Everaert, Rohit Patki, Tianqi Zheng, Christopher Potts

DiAL: Diversity aware listwise ranking for query auto-complete
Sonali Singh, Sachin Farfade, Prakash Mandayam Comar

Question answering

RAG-QA arena: Evaluating domain robustness for long-form retrieval-augmented question answering
Rujun Han, Yuhao Zhang, Peng Qi, Yumo Xu, Jenyuan Wang, Lan Liu, William Yang Wang, Bonan Min, Vittorio Castelli

Retrieving contextual information for long-form question answering using weak supervision
Philipp Christmann, Svitlana Vakulenko, Ionut Teodor Sorodoc, Bill Byrne, Adrià de Gispert

Recommender systems

Efficient pointwise-pairwise learning-to-rank for news recommendation
Nithish Kannen Senthilkumar, Yao Ma, Gerrit van den Burg, Jean Baptiste Faddoul

Efficient pointwise-pairwise learning-to-rank for news recommendation.png
An illustration of the GLIMPSE framework proposed in "Efficient pointwise-pairwise learning-to-rank for news recommendation". GLIMPSE adopts a multitask approach in which a pretrained language model is fine-tuned on both the relevance prediction task and the pairwise-preference task. During inference, the relevance predictions are used to produce an initial pointwise ranking, which is subsequently improved by one or more right-to-left (RTL) passes using pairwise comparisons.

PEARL: Preference extraction with exemplar augmentation and retrieval with LLM agents
Vijit Malik, Akshay Jagatap, Vinayak Puranik, Anirban Majumder

Sequential LLM framework for fashion recommendation
Han Liu, Xianfeng Tang, Tianlang Chen, Jiapeng Liu, Indu Indu, Henry Peng Zou, Peng Dai, Roberto Fernandez Galan, Mike Porter, Dongmei Jia, Ning Zhang, Lian Xiong

Responsible AI

Attribute controlled fine-tuning for large language models: A case study on detoxification
Tao Meng, Ninareh Mehrabi, Palash Goyal, Anil Ramakrishna, Aram Galstyan, Richard Zemel, Kai-Wei Chang, Rahul Gupta, Charith Peris

FLIRT: Feedback loop in-context red teaming
Ninareh Mehrabi, Palash Goyal, Christophe Dupuy, Qian Hu, Shalini Ghosh, Richard Zemel, Kai-Wei Chang, Aram Galstyan, Rahul Gupta

Order of magnitude speedups for LLM membership inference
Rongting Zhang, Martin Bertran Lopez, Aaron Roth

Synthetic data generation

CorrSynth: A correlated sampling method for diverse dataset generation from LLMs
Suhas Kowshik, Abhishek Divekar, Vijit Malik

A Correlated Sampling Method for Diverse Dataset Generation from LLMs
"CorrSynth: A correlated sampling method for diverse dataset generation from LLMs" introduces a sampling method that uses anti-correlation between examples rather than few-shot generation.

DATA ADVISOR: Dynamic data curation for safety alignment of large language models
Fei Wang, Ninareh Mehrabi, Palash Goyal, Rahul Gupta, Kai-Wei Chang, Aram Galstyan

Evaluating differentially private synthetic data generation in high-stakes domains
Krithika Ramesh, Nupoor Gandhi, Pulkit Madaan, Lisa Bauer, Charith Peris, Anjalie Field

SYNTHESIZRR: Generating diverse datasets with retrieval augmentation
Abhishek Divekar, Greg Durrett

Abstract depiction of the SYNTHESIZRR procedure
Abstract depiction of the procedure proposed in "SYNTHESIZRR: Generating diverse datasets with retrieval augmentation". The content sourcing stage retrieves K unique documents {r1,...,rK} from a large corpus for each in-context covariate xICL. The task-inversion stage uses a parameterized context refinement prompt, Pτ, which takes parameters Rinv (inversion instruction), rk (a retrieved document), and V(yICL) (the verbalized target label). A generalist teacher LLM autoregressively generates a synthetic covariate. Each in-context example thus produces K unique synthetic examples {x̃1,..., x̃K}, which we include in the dataset with target yICL.

Text classification

Distance-aware calibration for pre-trained language models*
Alberto Gasparin, Gianluca Detommaso

Performance-guided LLM knowledge distillation for efficient text classification at scale

Flavio Di Palo, Prateek Singhi, Bilal Fadlallah

Prompt-tuned muti-task taxonomic transformer (PTMTTaxoFormer)
Rajashekar Vasantha, Nhan Nguyen, Yue Zhang

Text summarization

Salient information prompting to steer content in prompt-based abstractive summarization
Lei Xu, Asad Karim, Saket Dingliwal, Aparna Elangovan

Research areas
Tags
About the Author
Staff writer

Related content

Work with us

See more jobs See more jobs
Data Scientist I, Central Learning Solutions Science
US, WA, Bellevue
The Central Learning Solutions (CLS) Science team in Amazon Operations builds state-of-the-art solutions for enhancing leadership and associate development across the organization. We develop technology and mechanisms for building personalized learning experiences based on the profiles of different learners and assess the post-training performance curves for different learner profiles. As a Data Scientist on the team, you will be driving the data science roadmap for the CLS Science team. You will leverage your knowledge in statistics and econometrics, estimate the causal impact of training interventions, recommend the right interventions for a given learner profile, and measure the post-launch success of these interventions through A/B weblabs. These insights will help in dynamically changing the training content of Learning & Development courses and unlock opportunities to improve both training effectiveness and learner experience. You will collaborate effectively with internal stakeholders and cross-functional teams for solving business problems, create operational efficiencies, and deliver successfully against high organizational standards. Key job responsibilities - Establish scalable, efficient, automated processes for large scale data analyses, model development, model validation and implementation. - Use advanced causal inference methodologies to estimate the learning curves for different learner profiles and the effectiveness of training content. - Perform statistical analysis and statistical tests including hypothesis testing and A/B testing. - Implement new statistical, machine learning, or other mathematical methodologies to solve specific business problems. - Present deep dives and analysis to both technical and non-technical stakeholders, ensure clarity, and influence the strategy of business partners. About the team Central Learning Solutions (CLS) is the centralized technical team that builds Learning & Development products for North America Operations employees. CLS Science teams owns the science components that power the different products on Impact Estimation, Self Serve Analytics, Content Understanding, and Personal learning.
2027 Applied Scientist Internship – PhD, Amazon University Talent Acquisition
IT, Turin
We're building systems that turn research into real world impact. We're advancing the intelligence behind how customers converse, connect and interact with Alexa across the globe and we're tackling complex conversational AI challenges with state of the art GenAI systems. We're looking for PhD students across multiple research domains to invent, design, and implement state of the art solutions for never before solved problems. Your work here won't just stay in a notebook, it ships to production and reaches customers worldwide. Check out the details below including the job responsibilities, team details, and basic qualifications before submitting your application. You can find more information about the Amazon Science community as well as interview preparation tips via the links below; - https://www.amazon.science/ - https://amazon.jobs/content/en/career-programs/university/science - https://amazon.jobs/content/en/how-we-hire/university-roles/applied-science Key job responsibilities As an Applied Science Intern, you will own the design and development of end-to-end systems. You'll have the opportunity to write technical white papers, create roadmaps and drive production level projects that will support Amazon Science. You will work closely with Amazon scientists and other science interns to develop solutions and deploy them into production. You will have the opportunity to design new algorithms, models, or other technical solutions whilst experiencing Amazon's customer focused culture. The ideal intern should have the ability to work with diverse groups of people and cross-functional teams to solve complex business problems. A day in the life You'll spend your first weeks scoping your project with your mentor, then own the research and implementation end-to-end. Projects may include developing and implementing real-time AI services, conducting comprehensive analysis of GenAI algorithms and methodologies or contributing to the creation of reliable, efficient code for production systems that serve millions of users — depending on the team you're matched with. Many interns publish at top-tier conferences or see their work deployed to production before the internship ends. Interns may also be considered for a return offer at the end of their internship, subject to performance evaluation and headcount availability. Further benefits of an Amazon Science internship include; - All of our internships offer a competitive salary - Interns are paired with an experienced manager and mentor(s) - Interns get invited to different intern program or office events - Interns can build their professional and personal network with other Amazon Scientists - Interns can potentially publish work at top tier conferences About the team You'll submit a single application and we'll match you with science teams best aligned with your research interests. Applications are reviewed on a rolling basis, and your application stays active until we find a team match or confirm there are no matches available. Start dates are available throughout the year for durations of between 3–6 months. Please note, each team has different start date and duration preferences — your recruiter will confirm the preferences of the team you're matched with prior to interviewing. We offer science internships in multiple locations across the EMEA region and you can indicate your interest in all these locations by applying here (Austria, Estonia, France, Germany, Ireland, Israel, Italy, Jordan, Luxembourg, Netherlands, Poland, Romania, South Africa, Spain, Sweden, UAE, and UK). Please note we do not offer remote internships.
Applied Scientist, Fauna
US, NY, New York
We are seeking an Applied Scientist to lead the development of evaluation frameworks and data collection protocols for robotic capabilities. In this role, you will focus on designing how we measure, stress-test, and improve robot behavior across a wide range of real-world tasks. Your work will play a critical role in shaping how policies are validated and how high-quality datasets are generated to accelerate system performance. You will operate at the intersection of robotics, machine learning, and human-in-the-loop systems, building the infrastructure and methodologies that connect teleoperation, evaluation, and learning. This includes developing evaluation policies, defining task structures, and contributing to operator-facing interfaces that enable scalable and reliable data collection. The ideal candidate is highly experimental, systems-oriented, and comfortable working across software, robotics, and data pipelines, with a strong focus on turning ambiguous capability goals into measurable and actionable evaluation systems. Key job responsibilities - Design and implement evaluation frameworks to measure robot capabilities across structured tasks, edge cases, and real-world scenarios - Develop task definitions, success criteria, and benchmarking methodologies that enable consistent and reproducible evaluation of policies - Create and refine data collection protocols that generate high-quality, task-relevant datasets aligned with model development needs - Build and iterate on teleoperation workflows and operator interfaces to support efficient, reliable, and scalable data collection - Analyze evaluation results and collected data to identify performance gaps, failure modes, and opportunities for targeted data collection - Collaborate with engineering teams to integrate evaluation tooling, logging systems, and data pipelines into the broader robotics stack - Stay current with advances in robotics, evaluation methodologies, and human-in-the-loop learning to continuously improve internal approaches - Lead technical projects from conception through production deployment - Mentor junior scientists and engineers About the team Fauna Robotics, an Amazon company, is building capable, safe, and genuinely delightful robots for everyday life. Our goal is simple: make robots people actually want to live and interact with in everyday human spaces. We believe that future won’t arrive until building for robotics becomes far more accessible. Today, too much effort is spent reinventing the fundamentals. We’re changing that by developing tightly integrated hardware and software systems that make it faster, safer, and more intuitive to create real-world robotic products. Our work spans the full stack: mechanical design, control systems, dynamic modeling, and intelligent software. The focus is not just functionality, but experience. We’re building robots that feel responsive, expressive, and genuinely useful. At Fauna, you’ll work at the frontier of this space, helping define how robots move, manipulate, and interact with people in natural environments. It’s an opportunity to solve hard problems across hardware and software with a team focused on making robotics accessible and joyful to build. If you care about making robotics real for everyone and building systems that are as delightful as they are capable, we’re interested in hearing from you.
Data Scientist, Fleet Planning , Global Fleet Products
US, WA, Bellevue
Amazon Global Fleet and Products (GFP) organization is responsible for fleet programs and capacity. The Fleet Planning team is looking for a Data Scientist to drive the most efficient use of fleet. Amazon Middle and Last Mile fleet planning is a complex resource allocation problem. The goal of fleet allocation planning is to optimize the size and mix of fleet allocated to various programs through. Changes in routes, network, exiting DSPs and new DSP onboarding create continuous need for re-allocation of fleet to maintain an efficient network capacity. This requires allocation to adhere to various operational limits (repair network, EV range, Station Charging capability) and also match route’s cube need to vehicles capacity. As a Data Scientist on the Fleet Planning team (GFP), you will be responsible for building new science models (linear programs, statistical and ML models) and enhancing existing models for changing business needs. You would work with program managers in planning, procurement, redeployment, deployment, remarketing, variable fleet and infrastructure programs to build models that would support the requirements of all programs in a coherent plan. Key job responsibilities • Build models and automation for planners for generating vehicle allocation plans • Partner with program teams to test and measure success of implemented model • Lead reviews with senior leadership, deep dive model outputs and explain implications of model recommendations.
Research Scientist III, Last Mile Science
IN, KA, Bangalore
Are you passionate about solving complex logistics challenges? Our Analytics team is at the forefront of enhancing delivery experiences through data-driven solutions and innovative technology. As a Research Scientist, you will join a team dedicated to optimizing our delivery network, ensuring reliable and efficient service to our customers. We are seeking an enthusiastic, customer-centric professional with strong analytical capabilities to drive impactful projects, implement advanced solutions, and develop scalable processes. In this role, you will have immediate ownership of business-critical challenges and the opportunity to make strategic, data-driven decisions that shape the future of our delivery operations. Your work will directly influence customer experience and operational excellence. The ideal candidate will possess both research science capabilities and program management skills, thriving in an environment that requires independent decision-making and comfort with ambiguity. This role offers the opportunity to make a significant impact on our advanced logistics network while working with pioneering technology and data science applications. Basic qualifications • 3+ years of building machine learning models for business application experience • Knowledge of programming languages such as C/C++, Python, Java or Perl • Experience programming in Java, C++, Python or related language • Experience with neural deep learning methods and machine learning Preferred qualifications: • PhD in engineering, technology, computer science, machine learning, robotics, operations research, statistics, mathematics or equivalent quantitative field • 5+ years of extensive relevant research experience • Deep expertise in Machine Learning • Proficiency in programming • Core competency in mathematics and statistics • Track record of successful projects in algorithm design and product development • Publications at peer-reviewed conferences or journals • Strategic thinker with good execution skills • Exhibits excellent business judgment • Effective verbal and written communication skills • Experience working with real-world data sets and building scalable models from big data • Experience with modern modeling tools and frameworks such as R, scikit-learn, Spark MLLib, MxNet, Tensorflow • Experience with large scale distributed systems
Applied Scientist, Minerva
IN, KA, Bengaluru
We are looking for Applied Scientists to advance how Minerva detects and responds to risk across KDP and Kindle Unlimited, building new models and raising the bar on the ones in production. You will own a problem end to end, from framing and modeling through evaluation and production launch, working with scientists and engineers in Bengaluru and the US. Key job responsibilities - Monitor production risk models and push their quality higher: track performance and drift, improve how performance is measured, and lead the deep-dive science that turns model behavior into better detection. - Design, train, and evaluate improved detection models, and take them to production with engineering partners. - Build datasets, labels, evaluation sets, and metrics, and strengthen them as abuse patterns change. - Write clear readouts with recommendations for scientists, engineers, and policy partners. About the team Kindle Direct Publishing (KDP) has empowered a new wave of self-motivated creators, tearing down barriers that once blocked writers from reaching readers. Minerva is the science and analytics organization for Amazon's Books business. We build systems for content understanding, author marketing and discovery, fraud and abuse prevention, and content risk, and we optimize manufacturing, fulfillment, and quality for Print On Demand (POD). We have mature areas and green-field opportunities, with engineering, business intelligence, and data engineering support across our areas. We are now building science teams in Bengaluru that work closely with our scientists in the US. Risk Science, a team within Minerva, keeps fraud, abuse, copyright infringement, and harmful content out of the Books catalog and programs such as Kindle Unlimited, using text, image, and behavioral data.
Senior Robotics/AI Motor Control Scientist, Fauna
US, NY, New York
We are seeking a Robotics/AI Motor Control Scientist to develop cutting-edge machine learning algorithms for motor control systems in robots. In this role, you will focus on creating and optimizing intelligent motor control strategies to enable robots to perform complex, whole-body tasks. Your contributions will be essential in advancing robotics by enabling fluid, reliable, and safe interactions between robots and their environments. Key job responsibilities - Develop controllers that leverage reinforcement learning, imitation learning, or other advanced AI techniques to achieve natural, robust, and adaptive motor behaviors - Collaborate with multi-disciplinary teams to integrate motor control systems with robotic hardware, ensuring alignment with real-world constraints such as actuator dynamics and energy efficiency - Use simulation and real-world testing to refine and validate control algorithms - Stay updated on advancements in robotics, AI, and control systems to apply advanced techniques to robotic motion challenges - Lead technical projects from conception through production deployment - Mentor junior scientists and engineers - Bridge research initiatives with practical engineering implementation About the team Fauna Robotics, an Amazon company, is building capable, safe, and genuinely delightful robots for everyday life. Our goal is simple: make robots people actually want to live and interact with in everyday human spaces. We believe that future won’t arrive until building for robotics becomes far more accessible. Today, too much effort is spent reinventing the fundamentals. We’re changing that by developing tightly integrated hardware and software systems that make it faster, safer, and more intuitive to create real-world robotic products. Our work spans the full stack: mechanical design, control systems, dynamic modeling, and intelligent software. The focus is not just functionality, but experience. We’re building robots that feel responsive, expressive, and genuinely useful. At Fauna, you’ll work at the frontier of this space, helping define how robots move, manipulate, and interact with people in natural environments. It’s an opportunity to solve hard problems across hardware and software with a team focused on making robotics accessible and joyful to build. If you care about making robotics real for everyone and building systems that are as delightful as they are capable, we’re interested in hearing from you. an opportunity to solve hard problems across hardware and software with a team focused on making robotics accessible and joyful to build. If you care about making robotics real for everyone and building systems that are as delightful as they are capable, we’re interested in hearing from you.
Applied & Research Scientist - Safe Autonomy Frontiers (SAF) Lab, Safe Autonomy Frontiers (SAF) Lab
US, CA, Pasadena
We are seeking Applied and Research Scientists to join the Safe Autonomy Frontiers (SAF) Lab. In this role, you will develop and deploy the science of safe autonomy on highly dynamic robots, advancing the frontier through both foundational research and realization on hardware. You will contribute to achieving safe autonomy across three key areas: control barrier function (CBF) theory for robust and performant safety on hardware; safe reinforcement learning for agile whole-body control; and layered safety filters that interface with learning, perception, and semantic reasoning systems. You will validate your work experimentally on state-of-the-art robotic platforms — removing bottlenecks to deployment and enabling robots to safely operate around humans. You will work with the inventor of CBFs, as well as top scientists and engineers at Amazon developing the next generation of safe autonomy. Key job responsibilities -Advance the science of safe autonomy from formal foundations to the integration with learning and perception to validation on hardware — with particular emphasis on methods that bridge these domains. -Test and validate hardware, with a focus on next generation robotic systems. Including locomotion, manipulation and loco-manipulation. -Leverage the deployments on hardware to validate the underlying science, identifying gaps between theory and practice that drive the next cycle of research. hardware to validate the underlying science, identifying gaps between theory and practice that drive the next cycle of research. -Publish research at top-tier robotics, control, and ML venues, and contribute to Amazon’s scientific reputation in advanced robotics. -Collaborate with SAF Lab and Amazon production teams to move research into robots deployed at Amazon scale. A day in the life Amazon offers a full range of benefits that support you and eligible family members, including domestic partners and their children. Benefits can vary by location, the number of regularly scheduled hours you work, length of employment, and job status such as seasonal or temporary employment. The benefits that generally apply to regular, full-time employees include: -Medical, Dental, and Vision Coverage -Maternity and Parental Leave Options -Paid Time Off (PTO) -401(k) Plan If you are not sure that every qualification on the list above describes you exactly, we’d still love to hear from you! At Amazon, we value people with unique backgrounds, experiences, and skillsets. If you’re passionate about this role and want to make an impact on a global scale, please apply! About the team The Safe Autonomy Frontiers (SAF) Lab is the first industry research lab dedicated to safe autonomy, founded by the inventor of control barrier functions. We are developing a universal safety layer for the next generation of robotic systems: mobile robots, manipulators, quadrupeds, and humanoids. Here you will advance performant safety for highly dynamic robots — CBF theory integrated with perception and learning, evaluated on next-generation platforms — and your work will underpin robots operating alongside people at Amazon’s unprecedented scale.
2027 Applied Scientist Internship – PhD, Amazon University Talent Acquisition
RO, Bucharest
We're building systems that turn research into real-world impact — developing AI-driven detection systems that identify and remediate security access violations across Amazon's global operations, and building ML solutions that leverage product images, text, and tabular data to improve safety and compliance mechanisms across Amazon's entire product catalog. If this sounds interesting to you then we have a range of opportunities for you to explore. We're looking for PhD students across multiple research domains to invent, design, and implement state-of-the-art solutions for never-before-solved problems. Your work here won't just stay in a notebook — it ships to production and reaches customers worldwide. Check out the details below including the job responsibilities, team details, and basic qualifications before submitting your application. You can find more information about the Amazon Science community as well as interview preparation tips via the links below; - https://www.amazon.science/ - https://amazon.jobs/content/en/career-programs/university/science - https://amazon.jobs/content/en/how-we-hire/university-roles/applied-science Key job responsibilities As an Applied Science Intern, you will own the design and development of end-to-end systems. You'll have the opportunity to write technical white papers, create roadmaps and drive production level projects that will support Amazon Science. You will work closely with Amazon scientists and other science interns to develop solutions and deploy them into production. You will have the opportunity to design new algorithms, models, or other technical solutions whilst experiencing Amazon's customer focused culture. The ideal intern should have the ability to work with diverse groups of people and cross-functional teams to solve complex business problems. A day in the life You'll spend your first weeks scoping your project with your mentor, then own the research and implementation end-to-end. Your work could involve building graph-based ML models and knowledge graphs that detect and remediate access violations at global scale, developing behavioral pattern clustering and automated reasoning systems for security operations, or researching multimodal and computer vision approaches on product catalog data to improve safety and compliance mechanisms — depending on the team you're matched with. Many interns publish at top-tier conferences or see their work deployed to production before the internship ends. Interns may also be considered for a return offer at the end of their internship, subject to performance evaluation and headcount availability. Further benefits of an Amazon Science internship include; - All of our internships offer a competitive salary - Interns are paired with an experienced manager and mentor(s) - Interns get invited to different intern program or office events - Interns can build their professional and personal network with other Amazon Scientists - Interns can potentially publish work at top tier conferences About the team We generally hire interns for multiple teams in Romania, including but not limited to; • CoSS — Research and develop ML solutions that improve safety and compliance across Amazon's entire product catalog working with tabular, textual, and product image data using deep learning, computer vision, LLMs, and multimodal models • Monitored Access Detection — Build AI-driven detection and remediation systems that identify global access violations across Amazon — leveraging knowledge graphs, graph modeling, behavioral pattern clustering, generative AI, and reinforcement learning to create a self-improving, closed-loop security system. You'll submit a single application and we'll match you with science teams best aligned with your research interests. Applications are reviewed on a rolling basis, and your application stays active until we find a team match or confirm there are no matches available. Start dates are available throughout the year for durations of between 3–6 months. Please note, each team has different start date and duration preferences — your recruiter will confirm the preferences of the team you're matched with prior to interviewing. We offer science internships in multiple locations across the EMEA region and you can indicate your interest in all these locations by applying here (Austria, Estonia, France, Germany, Ireland, Israel, Italy, Jordan, Luxembourg, Netherlands, Poland, Romania, South Africa, Spain, Sweden, UAE, and UK). Please note we do not offer remote internships.
2027 Applied Scientist Internship – Masters, Amazon University Talent Acquisition
IL, Tel Aviv
We're building systems that turn research into real-world impact — powering sports experiences for millions of Prime Video customers, pioneering multimodal document intelligence, building autonomous AI agents that reason, plan, and act, and transforming how customers discover products they love. If this sounds interesting to you then we have a range of opportunities for you to explore. We're looking for Masters students across multiple research domains to invent, design, and implement state-of-the-art solutions for never-before-solved problems. Your work here won't just stay in a notebook — it ships to production and reaches customers worldwide. Check out the details below including the job responsibilities, team details, and basic qualifications before submitting your application. You can find more information about the Amazon Science community as well as interview preparation tips via the links below; - https://www.amazon.science/ - https://amazon.jobs/content/en/career-programs/university/science - https://amazon.jobs/content/en/how-we-hire/university-roles/applied-science Key job responsibilities As an Applied Science Intern, you will own the design and development of end-to-end systems. You'll have the opportunity to write technical white papers, create roadmaps and drive production level projects that will support Amazon Science. You will work closely with Amazon scientists and other science interns to develop solutions and deploy them into production. You will have the opportunity to design new algorithms, models, or other technical solutions whilst experiencing Amazon's customer focused culture. The ideal intern should have the ability to work with diverse groups of people and cross-functional teams to solve complex business problems. A day in the life You'll spend your first weeks scoping your project with your mentor, then own the research and implementation end-to-end. Your work could involve building computer vision models that power live sports experiences for Prime Video, developing multimodal GenAI solutions for AWS document intelligence, creating agentic AI systems that reason and act autonomously, or advancing recommendation models that transform how customers discover content — depending on the team you're matched with. Many interns publish at top-tier conferences or see their work deployed to production before the internship ends. Interns may also be considered for a return offer at the end of their internship, subject to performance evaluation and headcount availability. Further benefits of an Amazon Science internship include; - All of our internships offer a competitive salary - Interns are paired with an experienced manager and mentor(s) - Interns get invited to different intern program or office events - Interns can build their professional and personal network with other Amazon Scientists About the team We're hiring interns for multiple teams in Israel, including but not limited to; • Prime Video Sports — Build innovative sports experiences for Prime Video, spanning computer vision, 3D simulation, and personalized content recommendations. • Personalization — Leverage LLMs, NLP, and recommender systems to match customers with products that align with their passions and shopping preferences. • Agentic AI — Build next-generation agentic AI systems that automate real-world knowledge work, spanning retrieval-augmented generation, multi-agent workflows, long-term memory and personalization, knowledge-graph construction, and agent evaluation. • DS3 Textract — Develop multimodal generative AI algorithms that pioneer state-of-the-art document understanding solutions impacting millions of customers. You'll submit a single application and we'll match you with science teams best aligned with your research interests. Applications are reviewed on a rolling basis, and your application stays active until we find a team match or confirm there are no matches available. Start dates are available throughout the year. Some teams offer full-time internships (3–6 months) while others offer part-time positions (50–60%, 8–12 months) — your recruiter will confirm the format during team matching. We offer science internships in multiple locations across the EMEA region and you can indicate your interest in all these locations by applying here (Austria, Estonia, France, Germany, Ireland, Israel, Italy, Jordan, Luxembourg, Netherlands, Poland, Romania, South Africa, Spain, Sweden, UAE, and UK). Please note we do not offer remote internships.
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 newsletter

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

Social