Evaluator of LLM
OtherTextClassificationText Generation
I currently work on an RLHF project where I create prompts and assess the responses that the model generates. The prompts should be natural, without contrived constraints, and belong to specific fields according to client requirements, such as health, science, fitness, education, travel, engineering, and math. Responses are evaluated according to several categories, such as localization, instruction following, truthfulness, verbosity, and writing quality.
Also, I work on a second project where I compare model responses, evaluate them, and improve them. In the past I have worked on recognizing images and classifying them into different categories
I currently work on an RLHF project where I create prompts and assess the responses that the model generates. The prompts should be natural, without contrived constraints, and belong to specific fields according to client requirements, such as health, science, fitness, education, travel, engineering, and math. Responses are evaluated according to several categories, such as localization, instruction following, truthfulness, verbosity, and writing quality. Also, I work on a second project where I compare model responses, evaluate them, and improve them. In the past I have worked on recognizing images and classifying them into different categories
2024