BACK TOProject Archive

EP.04 / PROJECT CASE FILE / 01

Chatbot Arena Preference Prediction

Fine-tuned and ensembled open-source LLMs for human preference prediction, placing 67th out of 1,849 teams.

QLoRAGemma 2Llama 3.1Multi-GPU

A closer look at the problem, engineering decisions, system design, and measurable outcome.

PRIVATE REPOSITORY · DETAILS AVAILABLE ON REQUEST

01 / Context

Predict which of two chatbot responses a human evaluator would prefer in the LMSYS Chatbot Arena competition.

02 / Approach

Fine-tuned Gemma-2-9B and Llama-3.1-8B with 4-bit QLoRA, inserting LoRA adapters into attention layers while working within limited GPU memory.

03 / Engineering

Tuned learning rate, prompt length, and frozen layers for stable training, then sorted test samples by length to reduce padding during multi-GPU inference.

04 / Model Strategy

Combined complementary model outputs in a weighted ensemble to improve generalization and preference-prediction accuracy.

05 / Result

Earned a Kaggle Silver Medal, finishing 67th of 1,849 teams—top 4% overall.

Next projectLive-Stream Danmaku SystemProject Archive