# Calibrate What You SHIP: Post-Selection Risk Control for Verifier-Guided Text-to-Image Generation

arXiv: https://arxiv.org/abs/2608.21748 · HTML: https://mlsanity.com/paper/2608.21748/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 44 references that are in the map.
- First posted on arXiv 22 Aug 2026.

## Builds on (most cited first, 22 of 22)

- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [Scaling Laws for Reward Model Overoptimization](https://mlsanity.com/paper/2210.10760/) (2022)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space](https://mlsanity.com/paper/2506.15742/) (2025)
- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment](https://mlsanity.com/paper/2403.05135/) (2024)
- [Evaluating Text-to-Visual Generation with Image-to-Text Generation](https://mlsanity.com/paper/2404.01291/) (2024)
- [TIFA: Accurate and Interpretable Text-to-Image Faithfulness Evaluation with Question Answering](https://mlsanity.com/paper/2303.11897/) (2023)
- [Distribution-Free, Risk-Controlling Prediction Sets](https://mlsanity.com/paper/2101.02703/) (2021)
- [Test-time Alignment of Diffusion Models without Reward Over-optimization](https://mlsanity.com/paper/2501.05803/) (2025)
- [From Reflection to Perfection: Scaling Inference-Time Optimization for Text-to-Image Diffusion Models via Reflection Tuning](https://mlsanity.com/paper/2504.16080/) (2025)
- [GenEval 2: Addressing Benchmark Drift in Text-to-Image Evaluation](https://mlsanity.com/paper/2512.16853/) (2025)
- [Reflect-DiT: Inference-Time Scaling for Text-to-Image Diffusion Transformers via In-Context Reflection](https://mlsanity.com/paper/2503.12271/) (2025)
- [T2I-Copilot: A Training-Free Multi-Agent Text-to-Image System for Enhanced Prompt Interpretation and Interactive Generation](https://mlsanity.com/paper/2507.20536/) (2025)
- [Cycle Consistency as Reward: Learning Image-Text Alignment without Human Preferences](https://mlsanity.com/paper/2506.02095/) (2025)
- [Conformal Selective Prediction with General Risk Control](https://mlsanity.com/paper/2603.24704/) (2026)
- [RAISE: Requirement-Adaptive Evolutionary Refinement for Training-Free Text-to-Image Alignment](https://mlsanity.com/paper/2603.00483/) (2026)
- [A Joint Finite-Sample Certificate for Adaptive Selective Conformal Risk Control](https://mlsanity.com/paper/2606.08517/) (2026)
- [Flash-BoN: Instant Drafts for Inference-Time Scaling in Diffusion Models](https://mlsanity.com/paper/2607.04461/) (2026)

## Built on by (most cited first, 1 of 1)

- [Background Gradients Shape Memorization in Flow Matching](https://mlsanity.com/paper/2609.33210/) (2026)

## Abstract (arXiv)

Verifier-guided text-to-image systems increasingly use test-time search to select, refine, or stop among multiple candidates, yet release thresholds are often calibrated on individual images. This creates a candidate-to-policy calibration mismatch: search changes both which prompts receive an output and which candidate is released, so candidate-level risk control need not imply control of released-output risk. We formalize this estimand shift through prompt reweighting and within-prompt selection, and introduce SHIP, Selection-aware Held-out calibration of Inference Policies. SHIP runs or replays the complete deployed policy on held-out prompts, evaluates the image it actually releases using an independent target judge, and selects the most permissive threshold whose risk upper bound satisfies a prescribed budget. For replayable policies with a prespecified threshold grid, simultaneous confidence control provides finite-sample validity. Experiments across fixed, sequential, and adaptive T2I inference procedures show that policy-level calibration recovers lower-risk operating points while exposing policy-dependent tradeoffs among risk, coverage, and compute. On GenEval2 with FLUX at N=16, a pooled-candidate threshold yields released risk 0.310, whereas SHIP reduces it to 0.162. Across 200 cached-stream splits, the fixed-grid certificate has no target crossing. Reliable inference-time scaling therefore requires calibrating the output distribution induced by the complete deployed policy.