What is mattwood.fyi?

A live, structured feed of riffs and links from Matt Wood, Chief AI & Technology Officer at AWS — readable by people and queryable by agents as a typed knowledge graph.

Is there an API and is it free?

Yes. A public, read-only query API is available with no authentication and no API keys. See https://mattwood.fyi/openapi.json and https://mattwood.fyi/docs.

How should an AI agent use this site?

Fetch https://mattwood.fyi/feed.json for the full structured feed, or query the API for keyword/semantic search and the connection graph. See https://mattwood.fyi/agents/ for when-to-use guidance.

OneRec-Think: In-Text Reasoning for Generative Recommendation

A framework that enhances LLM-based recommendation systems with explicit reasoning capabilities through item-textual alignment, reasoning scaffolding, and recommendation-specific reward functions, validated on public benchmarks and deployed at scale on Kuaishou.

arxiv.org →

Connections

Related to

PLUM: Adapting Pre-trained Language Models for Industrial-scale Generative Recommendations PLUM and OneRec-Think both adapt pre-trained language models for industrial generative recommendation, though OneRec-Think specifically focuses on reasoning scaffolding via chain-of-thought
Semantic ID-based Generative Retrieval for Podcast Discovery at Spotify Both tackle generative retrieval/recommendation using semantic alignment of items to language model representations, addressing the ID vs. text representation challenge
GenRec: LLM-Native Recommendation System at Netflix Both address LLM-native generative recommendation systems at scale, with OneRec-Think adding explicit reasoning capabilities to the paradigm GenRec pioneered at Netflix

Supports

Contrastive Language Model (CLM) OneRec-Think's item-textual alignment mechanism is closely related to contrastive language modeling approaches, leveraging semantic contrast between items to ground LLM reasoning in recommendation contexts
Thinking Fast and Slow in AI: the Role of Metacognition OneRec-Think's explicit reasoning framework empirically supports the metacognitive 'thinking fast and slow' thesis — deliberate in-text reasoning improves recommendation quality over fast implicit pattern matching
LLM Classification Is Feature Engineering OneRec-Think's recommendation-specific reward functions reinforce the idea that LLM outputs for structured tasks like recommendation require domain-specific evaluation criteria beyond generic language modeling

Type:	link
First published:	Tuesday, 29 September 2026 at 08:10 UTC
Location:	https://arxiv.org/abs/2510.11639
URL:	https://mattwood.fyi/i/1dd58012

I'm Matt Wood, and this is For Your Information. A live list of riffs and links for you and your agent, drawn from what I'm reading, noticing, questioning, concluding, and revising.

Connections