Skip to content
View xiaojing29's full-sized avatar

Highlights

  • Pro

Block or report xiaojing29

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse

Pinned Loading

  1. PromptContrastiveDecoding PromptContrastiveDecoding Public

    Forked from ZurichNLP/ContraDecode

    Implementation of prompt-based contrastive decoding to reduce omission errors in low-resource machine translation with LLMs. Based on thesis work extending contrastive decoding to the prompt level.

    Jupyter Notebook

  2. HallucinationDetection HallucinationDetection Public

    Detecting hallucinations in LLM-generated answers using cross-checking consistency across models. Implements and extends the SAC3 method to smaller open-source models.

    Jupyter Notebook

  3. FinancialSummarizationLLMs FinancialSummarizationLLMs Public

    Fine-tuning small LLMs (Phi-3.5, LLaMA-3.2) with QLoRA for summarizing earnings call transcripts. Evaluated with ROUGE and BERTScore. Part of NLP for Finance coursework.

    Jupyter Notebook

  4. ProfilingCreditRepairConsumerComplaints ProfilingCreditRepairConsumerComplaints Public

    Few-shot learning project classifying credit repair complaints by consumer knowledge level. Combines transformer embeddings, manual features, and multiple classifiers.

    Jupyter Notebook

  5. IBMWatsonxGenAIChallenge IBMWatsonxGenAIChallenge Public

    IBM GenAI Challenge 2024 submission: Automated processing of customer-service emails using LLMs on IBM Watsonx. Delivered a real-world solution in a collaborative team setting.

    Jupyter Notebook

  6. WordWizard WordWizard Public

    A multilingual crossword game for language learners. Supports 5 languages (A1–C1 CEFR levels). Built with Flask, JavaScript, and Three.js. Includes user auth, clue generation, and gamification.

    Python