AI Paper Atlas

Research map / Large language models / Alignment and preferences

TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching and Clap-Ranked Preference Optimization

Chia-Yu Hung, Navonil Majumder, Zhifeng Kong, Ambuj Mehrish, Amir Zadeh, Chuan Li, Rafael Valle, Bryan Catanzaro, Soujanya Poria

ICLR 2026 · Large language models › Alignment and preferences · approach: Robust and calibrated DPO · 84 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use CRPO preference data instead of Audio Alpaca (ablation)TangoFlux-base · AudioCapsCLAP score: 0.448 -> 0.453Improved
use CRPO preference data instead of BATON (ablation)TangoFlux-base · AudioCapsCLAP score: 0.437 -> 0.453Improved
use CRPO preference data instead of BATON (ablation)TangoFlux-base · AudioCapsFD_openl3: 80.5 -> 79.1Improved
apply CRPO alignment to TangoFlux-baseTangoFlux-base · AudioCapsCLAP score: 0.431 -> 0.480Improved
apply CRPO alignment to TangoFlux-baseTangoFlux-base · AudioCapsFD_openl3: 80.2 -> 75.1Improved
apply CRPO alignment to TangoFlux-baseTangoFlux-base · AudioCapsKL_passt: 1.22 -> 1.15Improved
apply CRPO alignment to TangoFlux-baseTangoFlux-base · AudioCaps multi-eventCLAP score: 0.438 -> 0.488Improved
CLAP best-of-N reranking, N=1 to 15TangoFlux · AudioCapsCLAP score: 0.480 -> 0.502Improved
CLAP best-of-N reranking, N=1 to 15TangoFlux · AudioCapsFD_openl3: 75.0 -> 75.1No clear effect
CLAP best-of-N reranking, N=1 to 15Tango 2 · AudioCapsCLAP score: 0.447 -> 0.473Improved
raise CFG scale from 3.0 to 5.0 (ablation)TangoFlux · AudioCapsFD_openl3: 77.7 -> 74.6Improved
raise CFG scale from 3.5 to 5.0 (ablation)TangoFlux · AudioCapsCLAP score: 0.481 -> 0.472Hurt

Later papers that compared with this method

1 later accepted paper reports TangoFlux as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy