Search results

We study offline evaluation of two-stage recommender systems, focusing on the first stage, candidate generation. Traditionally, candidate generators have been evaluated in terms of standard information retrieval metrics, using curated or heuristically labeled data, which does not always reflect their true impact to user experience or business metrics. We instead take a holistic view, measuring their effectiveness

The increasing complexity and fragmentation of financial systems in large organizations have created significant challenges for financial teams, particularly in performing real-time, end-to-end validation, as existing validation methods relying on static rules or batch processing are often inadequate for today's dynamic financial environments. This paper introduces a novel approach using Large Language

We present an autonomous framework that leverages Large Language Models (LLMs) to automate end-to-end business analysis and market report generation. At its core, the system employs specialized agents - Researcher, Reviewer, Writer, and Retriever - that collaborate to analyze data and produce comprehensive reports. These agents learn from real professional consultants' presentation materials at Amazon through

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

Text style transfer in enterprise environments presents unique challenges that extend beyond traditional style transfer approaches, particularly when dealing with complex technical documentation and strict organizational guidelines. This paper introduces Onoma, a novel enterprise-scale style transfer system that addresses the fundamental challenges of maintaining consistent brand voice while preserving

The results of software verification are only as trustworthy as the provided specification. Errors or incompleteness in the specification can result in unwarranted confidence in the implementation. Previous work, particularly in the realm of model checking, has investigated a notion of coverage in verification. Portions of the system that could be replaced arbitrarily without causing verification failure

In this paper we investigate cross-lingual Text-To-Speech (TTS) synthesis through the lens of adapters, in the context of lightweight TTS systems. In particular, we compare the tasks of unseen speaker and language adaptation with the goal of synthesising a target voice in a target language, in which the target voice has no recordings therein. Results from objective evaluations demonstrate the effectiveness

From reimagining storage to serverless computing, Aurora continues to push the boundaries of what's possible in database technology.

Customer Solutions Manager

Generative AI has unlocked new possibilities in content discovery and management. Through collaboration with the National Football League (NFL), we demonstrate how a generative-AI based workflow allows media researchers and analysts to query relevant historical plays using natural language, rather than using traditional filter and click-based interfaces. The agentic workflow takes a user query in natural

Large language models (LLMs) have demonstrated remarkable performance across a wide range of tasks in recent years. While prior work has explored leveraging LLMs to generate synthetic data for self-improvement, repeated iterations often suffer from diminishing returns due to the reliance on homogeneous reasoning patterns and limited exploration of alternative perspectives. In this paper, we introduce a

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

We introduce AlquistCoder, a code-generating system that effectively minimizes the risk of producing malicious content or vulnerable code while maintaining excellent Python coding and question answering standards across a wide range of tasks. The architecture of AlquistCoder employs a sophisticated input guardrail classifier that analyzes whether the user’s intention is benign, potentially harmful, or falls

We present SecureLion, a trustworthy AI assistant designed to securely han-dle cybersecurity queries and generate vulnerability-free code. Compared to the base model, our system achieves a 76.6% relative reduction on insecure messages in an adversarial competition setting at the cost of negligible utility loss. The success of SecureLion stems primarily from: (1) pervasive security-focused reasoning integrated

We introduce PurpCode, a novel post-training method that aligns coding assistants to perform safety reasoning to defend against malicious cyber activities and provide secure and functional code. Our approach trains a reasoning model in two stages:(i) Rule learning, which explicitly teaches the model to reference cyber safety rules to avoid facilitating malicious cyber events and to generate vulnerability-free

This paper presents the vision, scientific contributions, and technical details of RedTWIZ: an adaptive and diverse multi-turn red teaming framework, to audit the robustness of Large Language Models (LLMs) in AI-assisted software development. Our work is driven by three major research streams: (1) robust and systematic assessment of LLM conversational jailbreaks; (2) a diverse generative multi-turn attack

WARNING: Contains harmful content that can be offensive in nature Large language models (LLMs) for code generation have enhanced developer productivity while introducing new misuse vectors, as these models can generate potentially harmful code. Existing evaluation methods fail to assess such misuse scenarios, and structured red teaming pipelines for code generation remain under-developed. This paper presents

Large Language Models (LLMs) for code generation (i.e., Code LLMs) have demonstrated impressive capabilities in AI-assisted software development and testing. However, recent studies have shown that these models are prone to gen-erating vulnerable or even malicious code under adversarial settings. Existing red-teaming approaches rely on extensive human effort, limiting their scalability and practicality,

While large language models (LLMs) achieve strong performance in code generation, persistent security vulnerabilities hinder their safe deployment. Starting from a pretrained CodeGen model without inherent safety mechanisms, we develop a systematic synthetic instruction-tuning workflow to progressively enhance model security. Our pipeline begins with taxonomy-guided synthetic data, capturing diverse attack