lab21-qwen3.5-4b-triage-lora API & Inference Endpoint

Lab 21 — LoRA adapter for Vietnamese CSKH ticket triage

LoRA adapter (bf16/fp16, PEFT) fine-tuned on unsloth/Qwen3.5-4B for a 4-field JSON triage task (intent, urgency, product, sentiment) on Vietnamese customer-support tickets. Trained as part of the AICB-P2T3 Day 21 fine-tuning lab.

Placement: all linear (text-linear), r=16, lora_alpha=32, LR=1e-4
2 epochs, 30 steps, effective batch 16, T4 (fp16)
target accuracy: 0.970 (vs. 0.765 for the best-prompted un-tuned base model)
Known limitation: the accompanying report found a regression metric drop of 0.147 (catastrophic forgetting on general-purpose questions) that fails the lab's regression gate — see the source repo's submission/REPORT.md for the full writeup, including a systematic bias where the model predicts urgency=trung_binh instead of the correct thap on several held-out examples.

Source code + full evaluation: https://github.com/daitrong94/K4-DAY21-2A202601242-HoangTrongDai

Lab 21 — LoRA adapter for Vietnamese CSKH ticket triage

Placement: all linear (text-linear), r=16, lora_alpha=32, LR=1e-4
2 epochs, 30 steps, effective batch 16, T4 (fp16)
target accuracy: 0.970 (vs. 0.765 for the best-prompted un-tuned base model)
Known limitation: the accompanying report found a regression metric drop of 0.147 (catastrophic forgetting on general-purpose questions) that fails the lab's regression gate — see the source repo's submission/REPORT.md for the full writeup, including a systematic bias where the model predicts urgency=trung_binh instead of the correct thap on several held-out examples.

Source code + full evaluation: https://github.com/daitrong94/K4-DAY21-2A202601242-HoangTrongDai

lab21-qwen3.5-4b-triage-lora

README

Lab 21 — LoRA adapter for Vietnamese CSKH ticket triage

Explore FriendliAI today

README

Lab 21 — LoRA adapter for Vietnamese CSKH ticket triage