Think Bright, Diffuse Nice: Enhancing T2I-ICL via Inductive-Bias Hint Instruction and Query Contrastive Decoding

TL;DR

This paper introduces TBDN, a training-free framework that enhances text-to-image in-context learning by mitigating compliance failure and hallucination through inductive bias hints and contrastive decoding, achieving state-of-the-art results.

Contribution

The paper proposes a novel training-free approach combining Hint Instruction and Query Contrastive Decoding to improve T2I-ICL performance and robustness without additional training.

Findings

Achieves state-of-the-art on CoBSAT and Text-to-Image Fast Mini-ImageNet.

Robust generalization across models, prompts, and hyperparameters.

Maintains promising performance on concept preservation and prompt following.

Abstract

Text-to-Image In-Context Learning (T2I-ICL) enables customized image synthesis via interleaved text-image examples but faces two mutually reinforcing bottlenecks, compliance failure and prior-dominated hallucination, that form a vicious cycle degrading generation quality. Existing methods rely on tailored training, which limits flexibility and raises deployment costs. To address these challenges effectively, we propose TBDN, a training-free framework integrating two complementary closed-loop mechanisms: Hint Instruction (HI) and Query Contrastive Decoding (QCD). HI injects task-aware inductive bias via lightweight prompt engineering to anchor models on contextual mapping rules, thereby mitigating compliance failure. QCD adjusts the decoding distributions of language models by contrasting full-input and query-omitted distributions, suppressing prior-dominated hallucination. TBDN achieves State-of-the-Art performance on CoBSAT and Text-to-Image Fast Mini-ImageNet, with robust generalization across model backbones, prompt designs, and hyperparameters. It also maintains promising performance in concept preservation and prompt following on Dreambench++. By breaking the two bottlenecks, TBDN establishes a simple yet effective framework for efficient and reliable T2I-ICL.