<!-- Release notes generated using configuration in .github/release.yml at main -->
New Features
- Added support for the Azure OpenAI model. ([cbd4c8d](https://github.com/RUC-NLPIR/FlashRAG/commit/cbd4c8d))
- Added a new method: `trace`. This feature was implemented twice, suggesting it might have been enhanced or corrected in the second commit. ([b9d6fa1](https://github.com/RUC-NLPIR/FlashRAG/commit/b9d6fa1), [7a728a2](https://github.com/RUC-NLPIR/FlashRAG/commit/7a728a2))
- Integrated Sentence Transformers into the retriever module. ([12e330e](https://github.com/RUC-NLPIR/FlashRAG/commit/12e330e))
New Models
- Optimized generator settings specifically for the Baichuan model. ([7e0f124](https://github.com/RUC-NLPIR/FlashRAG/commit/7e0f124))
Bug Fixes
- Fixed a bug in the retrieval cache saving mechanism. ([18a727e](https://github.com/RUC-NLPIR/FlashRAG/commit/18a727e), [45662e3](https://github.com/RUC-NLPIR/FlashRAG/commit/45662e3))
- Fixed an error in the evaluation metrics. ([d0f403e](https://github.com/RUC-NLPIR/FlashRAG/commit/d0f403e))
- Resolved issues with the cache manager. ([8277011](https://github.com/RUC-NLPIR/FlashRAG/commit/8277011))
- Fixed a bug in the sentence transformer encoder. ([b0fdfbb](https://github.com/RUC-NLPIR/FlashRAG/commit/b0fdfbb))
- Resolved a bug in single GPU inference. ([982d433](https://github.com/RUC-NLPIR/FlashRAG/commit/982d433))
- Fixed a bug in BM25 functionality. ([5009089](https://github.com/RUC-NLPIR/FlashRAG/commit/5009089))
- Resolved a bug in the loading of T5-based models in DataParallel. ([6666b4a](https://github.com/RUC-NLPIR/FlashRAG/commit/6666b4a))
- Fixed a bug related to the max token setting in the EncoderDecoder generator. ([96b20c8](https://github.com/RUC-NLPIR/FlashRAG/commit/96b20c8))
Enhancements
- Optimized the implementation of the generator. ([d7a7136](https://github.com/RUC-NLPIR/FlashRAG/commit/d7a7136))
- Optimized the detection of generator types. ([ed7ef8e](https://github.com/RUC-NLPIR/FlashRAG/commit/ed7ef8e), [72c831d](https://github.com/RUC-NLPIR/FlashRAG/commit/72c831d))
- Optimized the refiner's memory usage. ([1984ca8](https://github.com/RUC-NLPIR/FlashRAG/commit/1984ca8))
- Optimized generator settings for better performance. ([7e0f124](https://github.com/RUC-NLPIR/FlashRAG/commit/7e0f124))
- Modified the LoRA loading of the generator under VLLM setting. ([c55ea39](https://github.com/RUC-NLPIR/FlashRAG/commit/c55ea39))
- Increased support for multi-choice datasets in metrics. ([95d9fcc](https://github.com/RUC-NLPIR/FlashRAG/commit/95d9fcc))
- Modified data path settings in the evaluator. ([70e739e](https://github.com/RUC-NLPIR/FlashRAG/commit/70e739e))
- Changed the metric name from `sub_em` to `acc`. ([15de563](https://github.com/RUC-NLPIR/FlashRAG/commit/15de563))
- Changed sample data to general knowledge. ([3d20a75](https://github.com/RUC-NLPIR/FlashRAG/commit/3d20a75))
- Updated default configuration settings. ([1acce56](https://github.com/RUC-NLPIR/FlashRAG/commit/1acce56))
Documentation & Code Formatting
- Added guidance for setting up configurations. ([37fc012](https://github.com/RUC-NLPIR/FlashRAG/commit/37fc012))
- Updated the README file multiple times for various improvements, including adding installation instructions, fixing typos, and providing more detailed documentation.
- Formatted code files for better readability. ([785a901](https://github.com/RUC-NLPIR/FlashRAG/commit/785a901))
- Added an introduction for beginners in Korean. ([c25f16e](https://github.com/RUC-NLPIR/FlashRAG/commit/c25f16e))
New Contributors
* PeterGriffinJin made their first contribution in https://github.com/RUC-NLPIR/FlashRAG/pull/23
* yifeikong made their first contribution in https://github.com/RUC-NLPIR/FlashRAG/pull/24
* kerrychu made their first contribution in https://github.com/RUC-NLPIR/FlashRAG/pull/39
* VSteinborn made their first contribution in https://github.com/RUC-NLPIR/FlashRAG/pull/52
* oneonlee made their first contribution in https://github.com/RUC-NLPIR/FlashRAG/pull/57
**Full Changelog**: https://github.com/RUC-NLPIR/FlashRAG/compare/v0.1.0...v0.1.1