AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

FlowCast: Trajectory Forecasting for Scalable Zero-Cost Speculative Flow Matching

Divya Jyoti Bajpai, Shubham Agarwal, Apoorv Saxena, Kuldeep Kulkarni, Subrata Mitra, Manjesh Kumar Hanawal

ICLR 2026 · Image generation › Fast diffusion sampling · approach: Guidance and alignment for flow matching · 3 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 FlowCast speculative generation at 50 steps instead of full samplingBAGEL · GenEvalOverall / speedup: 0.78 -> 0.78 at 2.5x speedupImproved
use FlowCast speculative generation at 50 steps instead of full samplingFLUX · GenEvalOverall / speedup: 0.65 -> 0.65 at 2.4x speedupImproved
FlowCast versus TeaCacheBAGEL · GenEvalOverall / speedup: 0.75 -> 0.78 overall; 1.8x -> 2.5xImproved
apply FlowCast with very few steps (5)BAGEL/FLUX · GenEvalOverall: 0.57 -> 0.57 (BAGEL); 0.44 -> 0.43 (FLUX)No clear effect
add FlowCast on top of PerFlowPerFlow · GenEvalOverall / speedup: 0.57 -> 0.57; 1.0x -> 1.9xImproved
add FlowCast to TeaCache (ablation)Step-1X-Edit · GEditG_O: 6.91 -> 6.90; speedup 1.8x -> 2.4xNo clear effect
FlowCast versus reducing steps staticallyStep-1X-Edit · GEditG_O: 6.54 -> 6.50 at 10 steps (3.0x -> 3.6x)Improved
increase MSE threshold epsilon (ablation)FlowCast · Multi-turn editing (BAGEL)speedup vs PSNRMixed
speculative drafting with parallel verificationFlowCast · FLUX latencyMemory (GB): 48 -> 79 GB at 50 stepsHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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