# Reward-Guided Semantic Evolution for Test-time Adaptive Object Detection

arXiv: https://arxiv.org/abs/2605.04531 · HTML: https://mlsanity.com/paper/2605.04531/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Prompt learning and adaptation of CLIP, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 7 of its 60 references that are in the map.
- First posted on arXiv 6 May 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Prompt learning and adaptation of CLIP](https://mlsanity.com/mllm/prompt-learning-and-adaptation-of-clip/)

## Builds on (most cited first, 7 of 7)

- [The CMA Evolution Strategy: A Tutorial](https://mlsanity.com/paper/1604.00772/) (2016)
- [Population Based Training of Neural Networks](https://mlsanity.com/paper/1711.09846/) (2017)
- [Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Coming](https://mlsanity.com/paper/1907.07484/) (2019)
- [STFAR: Improving Object Detection Robustness at Test-Time by Self-Training with Feature Alignment Regularization](https://mlsanity.com/paper/2303.17937/) (2023)
- [Generalizing vision-language models to novel domains: A comprehensive survey](https://mlsanity.com/paper/2506.18504/) (2025)
- [VLOD-TTA: Test-Time Adaptation of Vision-Language Object Detectors](https://mlsanity.com/paper/2510.00458/) (2025)
- [Bayesian Test-time Adaptation for Object Recognition and Detection with Vision-language Models](https://mlsanity.com/paper/2510.02750/) (2025)

## Abstract (arXiv)

Open-vocabulary object detection with vision-language models (VLMs) such as Grounding DINO suffers from performance degradation under test-time distribution shifts, primarily due to semantic misalignment between text embeddings and shifted visual embeddings of region proposals. While recent test-time adaptive object detection methods for VLM-based either rely on costly backpropagation or bypass semantic misalignment via external memory, none directly and efficiently align text and vision in a training-free manner. To address this, we propose Reward-Guided Semantic Evolution (RGSE), a training-free framework that directly refines the text embeddings at test time. Inspired by evolutionary search, RGSE treats text embedding adaptation as a semantic search process: it perturbs text embeddings as candidate variants, evaluates them via cosine similarity with current and historical high-confidence visual proposals as a reward signal, and fuses them into a refined embedding through reward-weighted averaging. Without any backpropagation, RGSE achieves state-of-the-art performance across multiple detection benchmarks while adding minimal computational overhead. Our code will be open source upon publication.