Confidence-aware Reward Optimization for Fine-tuning Text-to-Image Models

• URL: http://arxiv.org/abs/2404.01863v1
• Date: Tue, 2 Apr 2024 11:40:38 GMT
• Title: Confidence-aware Reward Optimization for Fine-tuning Text-to-Image Models
• Authors: Kyuyoung Kim, Jongheon Jeong, Minyong An, Mohammad Ghavamzadeh, Krishnamurthy Dvijotham, Jinwoo Shin, Kimin Lee,
• Abstract summary: Fine-tuning text-to-image models with reward functions trained on human feedback data has proven effective for aligning model behavior with human intent. However, excessive optimization with such reward models, which serve as mere proxy objectives, can compromise the performance of fine-tuned models. We propose TextNorm, a method that enhances alignment based on a measure of reward model confidence estimated across a set of semantically contrastive text prompts.
• Score: 85.96013373385057
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Fine-tuning text-to-image models with reward functions trained on human feedback data has proven effective for aligning model behavior with human intent. However, excessive optimization with such reward models, which serve as mere proxy objectives, can compromise the performance of fine-tuned models, a phenomenon known as reward overoptimization. To investigate this issue in depth, we introduce the Text-Image Alignment Assessment (TIA2) benchmark, which comprises a diverse collection of text prompts, images, and human annotations. Our evaluation of several state-of-the-art reward models on this benchmark reveals their frequent misalignment with human assessment. We empirically demonstrate that overoptimization occurs notably when a poorly aligned reward model is used as the fine-tuning objective. To address this, we propose TextNorm, a simple method that enhances alignment based on a measure of reward model confidence estimated across a set of semantically contrastive text prompts. We demonstrate that incorporating the confidence-calibrated rewards in fine-tuning effectively reduces overoptimization, resulting in twice as many wins in human evaluation for text-image alignment compared against the baseline reward models.

Related papers

• PrefPaint: Aligning Image Inpainting Diffusion Model with Human Preference [62.72779589895124]
  We make the first attempt to align diffusion models for image inpainting with human aesthetic standards via a reinforcement learning framework. We train a reward model with a dataset we construct, consisting of nearly 51,000 images annotated with human preferences. Experiments on inpainting comparison and downstream tasks, such as image extension and 3D reconstruction, demonstrate the effectiveness of our approach.
  arXiv  Detail & Related papers  (2024-10-29T11:49:39Z)
• Evaluating Robustness of Reward Models for Mathematical Reasoning [14.97819343313859]
  We introduce a new design for reliable evaluation of reward models, and to validate this, we construct RewardMATH. We demonstrate that the scores on RewardMATH strongly correlate with the results of optimized policy and effectively estimate reward overoptimization.
  arXiv  Detail & Related papers  (2024-10-02T16:39:58Z)
• Elephant in the Room: Unveiling the Impact of Reward Model Quality in Alignment [50.21842377409232]
  Despite vital role reward models play in alignment, previous works have consistently overlooked their performance. This work first investigates the quality of the widely-used preference dataset, HH-RLHF, and curates a clean version, CHH-RLHF. Based on CHH-RLHF, we benchmark the accuracy of a broad range of reward models used in previous alignment works, unveiling the unreliability of using them both for optimization and evaluation.
  arXiv  Detail & Related papers  (2024-09-26T04:28:35Z)
• Beyond Thumbs Up/Down: Untangling Challenges of Fine-Grained Feedback for Text-to-Image Generation [67.88747330066049]
  Fine-grained feedback captures nuanced distinctions in image quality and prompt-alignment. We show that demonstrating its superiority to coarse-grained feedback is not automatic. We identify key challenges in eliciting and utilizing fine-grained feedback.
  arXiv  Detail & Related papers  (2024-06-24T17:19:34Z)
• EVALALIGN: Supervised Fine-Tuning Multimodal LLMs with Human-Aligned Data for Evaluating Text-to-Image Models [16.18275805302776]
  We propose EvalAlign, a metric characterized by its accuracy, stability, and fine granularity. We develop evaluation protocols that focus on two key dimensions: image faithfulness and text-image alignment. EvalAlign aligns more closely with human preferences than existing metrics, confirming its effectiveness and utility in model assessment.
  arXiv  Detail & Related papers  (2024-06-24T11:56:15Z)
• Calibrated Self-Rewarding Vision Language Models [27.686545023186852]
  Large Vision-Language Models (LVLMs) have made substantial progress by integrating pre-trained large language models (LLMs) and vision models through instruction tuning. LVLMs often exhibit the hallucination phenomenon, where generated text responses appear linguistically plausible but contradict the input image. We propose the Calibrated Self-Rewarding (CSR) approach, which enables the model to self-improve by iteratively generating candidate responses, evaluating the reward for each response, and curating preference data for fine-tuning.
  arXiv  Detail & Related papers  (2024-05-23T14:30:33Z)
• Multimodal Large Language Model is a Human-Aligned Annotator for Text-to-Image Generation [87.50120181861362]
  VisionPrefer is a high-quality and fine-grained preference dataset that captures multiple preference aspects. We train a reward model VP-Score over VisionPrefer to guide the training of text-to-image generative models and the preference prediction accuracy of VP-Score is comparable to human annotators.
  arXiv  Detail & Related papers  (2024-04-23T14:53:15Z)
• Aligning Text-to-Image Models using Human Feedback [104.76638092169604]
  Current text-to-image models often generate images that are inadequately aligned with text prompts. We propose a fine-tuning method for aligning such models using human feedback. Our results demonstrate the potential for learning from human feedback to significantly improve text-to-image models.
  arXiv  Detail & Related papers  (2023-02-23T17:34:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.