---
type: "raw_signal"
id: "sakana-ai-released-qd-rec-a-quality-diversity-fb76e1"
slug: "sakana-ai-released-qd-rec-a-quality-diversity-fb76e1"
title: "Sakana AI released QD-Rec , a Quality-Diversity recommender that evolves item plus reason pairs instead of ranking."
description: "Sakana AI introduced QD-Rec, a recommendation method that searches over candidate-and-rationale pairs rather than producing only a ranked list."
retrieval_nugget: "Sakana AI introduced QD-Rec, a recommendation method that searches over candidate-and-rationale pairs rather than producing only a ranked list."
observed_at: "2026-10-21"
record_date: "2026-10-21"
date_kind: "observed_at"
why_it_matters: "These are reported experimental results, not evidence of deployment outcomes. The article also notes that generated recommendations can contain unsupported claims, so the outputs should not be treated as decision-ready without validation. Source: https://sakana.ai/qd-rec"
novelty: "notable"
verification_level: "source-linked"
signal_type: "update"
source_platform: "sakana.ai"
topics: []
entities: ["Sakana AI","QD-Rec","Quality-Diversity"]
related_patterns: ["model-answers-become-task-interfaces"]
novelty_label: "new_fact"
what_changed: "Sakana AI’s new QD-Rec research introduces a quality-diversity recommender that iteratively evolves diverse item-and-reason pairs; the prior Sakana stories cover different releases and research."
prior_refs: ["memory:3eaa0ab2-f4e0-86bd-a1e6-07710a2cd84b","memory:f965269c-0fc8-875d-ab12-fce626938566"]
source_url: "https://sakana.ai/qd-rec"
source_urls: ["https://sakana.ai/qd-rec"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:sakana-ai-released-qd-rec-a-quality-diversity-fb76e1"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/signals/sakana-ai-released-qd-rec-a-quality-diversity-fb76e1/","markdown":"https://newruntime.com/signals/sakana-ai-released-qd-rec-a-quality-diversity-fb76e1.md","json":"https://newruntime.com/signals/sakana-ai-released-qd-rec-a-quality-diversity-fb76e1.json"}
---

# Sakana AI released QD-Rec , a Quality-Diversity recommender that evolves item plus reason pairs instead of ranking.

## Retrieval answer

Sakana AI introduced QD-Rec, a recommendation method that searches over candidate-and-rationale pairs rather than producing only a ranked list.

Sakana AI introduced QD-Rec, a recommendation method that searches over candidate-and-rationale pairs rather than producing only a ranked list. The system uses a Quality-Diversity archive: an LLM generates new hypotheses by mutating or combining prior ones, while an AI or human evaluator scores them. Novel directions can occupy new niches; otherwise, a proposal competes with its most similar archived counterpart.

The source reports tests in three settings: M&A candidate discovery, customer targeting, and research-paper recommendation. After 1,000 generation-and-evaluation cycles, QD-Rec built larger archives with higher average evaluations in many settings than the described baselines. Sakana AI frames the approach as supporting comparison across distinct recommendations, not merely selecting a single top-scoring item.

These are reported experimental results, not evidence of deployment outcomes. The article also notes that generated recommendations can contain unsupported claims, so the outputs should not be treated as decision-ready without validation. Source: https://sakana.ai/qd-rec

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)