We strive to build an interdisciplinary team working at the area of natural language processing, medical AI and brain computer interfaces. Our core values are:

  • Experimental:  We value scientific rigor, focusing on researching under strong scientific grounds and conducting sound experiments that provide definitive and repeatable findings.  
  • Computational:  Our scientific nature is to use algorithms, mathematical models, strong theoretical background, and strong coding skills.

Lab Entry

We welcomed students from all disciplines but those with a huge passion for publishing to top-tier conferences/journals and/or making startup-related product.  Make sure you read our graduation criteria here first:  Each year, we TRY not to take more than 10 students.


We welcomed internship students to work on some of the applications/products (not Theory) we are working at.  There is no criteria for finishing internship - the primary function is to give you some real job and let you get it done.  Please contact me if you are interested.

Current Projects

We recommend everyone joining our lab to pursue these continuing projects;  we TRY NOT to take any students new topics.


  1. Foundational models for Thai language:  develop embedding, instruct, speech-text, image-text, style transfer, and emotion-speech models for Thai language, using pruning, distillation, grafting, quantization, prefix-tuning for efficient models.
  2. Summarization + knowledge transfer:   apply knowledge transfer such as shared prefix tuning, multi-task training, self-distillation to summarization tasks
  3. Summarization + masking:  apply selective masking techniques for better summarization
  4. Summarization + contrastive learning - apply contrastive learning for better summarization models
  5. Summarization + reinforcement learning: apply RL for better summarization models
  6. Medical Visual QA:  design models for better medical visual QA


  1. Glucose monitoring: contributes to the use of Raman spectroscopy and the development of Raman wearables to monitor blood glucose in real-time
  2. BCI speller: contributes to the development of BCI speller using EEG paradigms such as P300, SSVEP, Hybrid P300-SSVEP and motor imagery for locked-in patients.
  3. ChatRad:  contributes to the development of automatic annotation, segmentation, search and reporting for fMRI/MRI/CT scan images.   It should support both textual and menu commands input.
  4. ChatData:   contributes to the development of a platform we called ChatData that can be the core platform to create other products such as legalGPT, medicalGPT, insuranceGPT, financeGPT, etc.   There will many components:  (1) instruction (parameter efficient) tuning with dialogue datasets, (2) reinforcement learning, (3) retrieval augmentation, (4) quantization for efficiency, (5) support images using OCR and image-text models, (6) support English and Thai, (7) agents - using tools of the internet.
  5. ThaiGovAI:  contributes to the development of models that can help turn informal text into formal text used in government, as well as generating common forms used in government operations.
  6. ElephantTalking:  contributes to the development for voice-based LLM used for insurance and banking industry.
  7. Scholarly: contributes to the development of product we called "Scholarly" which help researchers to write papers using style transfer and generative models.