AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

TAG: Tangential Amplifying Guidance for Hallucination-Resistant Sampling

Hyunmin Cho, Donghoon Ahn, Susung Hong, Jee Eun Kim, Seungryong Kim, Kyong Hwan Jin

ICML 2026 · Image generation › Fast diffusion sampling · approach: Guidance for conditional diffusion · 0 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
add TAGSD v1.5 · COCO 2014 unconditionalFID: 58.41 -> 46.20Improved
add TAGSDXL · COCO 2014 unconditionalFID: 119.14 -> 90.71Improved
add TAGSD v2.1 · COCO 2014 conditionalFID: 26.12 -> 21.59Improved
stack TAG on SEGSD v1.5 + SEG · COCO 2014 unconditionalFID: 47.69 -> 42.71Improved
stack TAG on PAGSD v1.5 + PAG · COCO 2014 unconditionalFID: 53.72 -> 52.61Improved
add TAG on SDXLSDXL · T2I-CompBench ComplexBLIP-VQA: 0.4443 -> 0.4650Improved
add TAG to DPM++DPM++ · ImageNet (SD v1.4)FID@30K: 55.40 -> 44.08Improved
add TAG to Qwen-ImageQwen-Image · COCO@10KFID: 57.53 -> 59.84Hurt
amplify normal component instead of tangential (ablation)TAG · SD samplingvisual qualityHurt
excessive eta (ablation)TAG · SD samplingfidelityHurt
restrict TAG to timesteps [1000,400] (ablation)TAG · ImageNet val 30K (SD v1.5)FID: 67.805 -> 63.870Improved
use adaptive scaling of eta (ablation)TAG · COCO conditionalFID@10K: 29.63 -> 25.86Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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