# ATHENA: Adaptive Test-Time Steering for Improving Count Fidelity in Diffusion Models

Source: [ATHENA: Adaptive Test-Time Steering for Improving Count Fidelity in Diffusion Models](<https://www.sota2.com/research/paper/69c0c8da0636e5bf4afc3973>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-24T05:00:03.763Z

Authors: Mohammad Shahab Sepehri, Asal Mehradfar, Berk Tinaz, Salman Avestimehr, Mahdi Soltanolkotabi

Published: 2026-03-20T06:20:44.000Z

## Abstract

Text-to-image diffusion models achieve high visual fidelity but surprisingly exhibit systematic failures in numerical control when prompts specify explicit object counts. To address this limitation, we introduce ATHENA, a model-agnostic, test-time adaptive steering framework that improves object count fidelity without modifying model architectures or requiring retraining. ATHENA leverages intermediate representations during sampling to estimate object counts and applies count-aware noise corrections early in the denoising process, steering the generation trajectory before structural errors become difficult to revise. We present three progressively more advanced variants of ATHENA that trade additional computation for improved numerical accuracy, ranging from static prompt-based steering to dynamically adjusted count-aware control. Experiments on established benchmarks and a new visually and semantically complex dataset show that ATHENA consistently improves count fidelity, particularly at higher target counts, while maintaining favorable accuracy-runtime trade-offs across multiple diffusion backbones.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Object Counting on CoCoCount](<https://www.sota2.com/research/sota/object-counting-on-cococount.md>): Object Counting; CoCoCount; Accuracy: 78.3.

- [Object Counting on CoCoCount-E](<https://www.sota2.com/research/sota/object-counting-on-cococount-e.md>): Object Counting; CoCoCount-E; Accuracy: 65.6.

- [Object Counting on ATHENA Dataset](<https://www.sota2.com/research/sota/object-counting-on-athena-dataset.md>): Object Counting; ATHENA Dataset; Accuracy: 56.1.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.19676>)

[PDF](<https://arxiv.org/pdf/2603.19676.pdf>)