Natural and conversational shopping experiences
summary
Ambient Explorer stitches web crawling, transformer-based information extraction, neural document summarization and a multimodal neural retrieval engine to surface curated recommendations and bite-sized summaries from Amazon and publishers; contextual QA uses query rewriting plus answer-extraction and factuality checks; recommendations run slate optimization and multi-armed bandits and train with few-shot and synthetic data (tens of thousands of simulated dialogues). Live demos add items to cart, answer product weight and support wake-word-free interaction via Conversation Mode.
AI-written summary · about
chapters (15)
- 0:00Intro
- 0:22Common shopper user journey
- 1:40How shoppers search and find products
- 2:36Common shopper pain points
- 5:39Conversational shopping
- 9:53Architecture overview
- 11:41Information extraction
- 14:32Shopping recommendations
- 17:48Contextual QA
- 21:35Bite-sized insights
- 24:27Foundations in conversational Al
- 27:10Natural conversations do not always follow a scripted path ...
- 27:32More natural and conversational
- 28:48Alexa conversations dialogue simulator
- 29:27Alexa conversations modeling architecture
related talks
17mEnhancing Chat-Based Customer Recommendations With GraphRAG and Hybrid Search
16mSolve for AI
38mAI-Ready Data Foundation Using Amazon OpenSearch
33mStrategies for Conversational Product Experiences
10mA Day in the Life: An Immersive Data Experience
25mBuilding the shopping journeys of the future - ChatGPT Apps, Adaptive Interfaces and more...
33m"Are You Sure?": blending product comparisons and recommendations with A.I.
11mEnabling richer experiences through multimodal interactions