Update README.md

README.md

@@ -4,7 +4,7 @@ datasets:
 - u-10bei/sft_alfworld_trajectory_dataset_v5
 language:
 - en
-license: apache-2.0
+license: mit
 library_name: peft
 pipeline_tag: text-generation
 tags:
@@ -12,35 +12,35 @@ tags:
 - agent
 - tool-use
 - alfworld
-- dbbench
 ---
 
-(2026-02-26) README practice edit by KOUJI039
-
-# ＜【課題】ここは自分で記入して下さい＞
+# LLM Lecture 2025 Advanced Competition (AgentBench: DBBench + ALFWorld)
 
 This repository provides a **LoRA adapter** fine-tuned from
-**Qwen/Qwen3-4B-Instruct-2507** using **LoRA + Unsloth**.
+**Qwen/Qwen3-4B-Instruct-2507** using LoRA + Unsloth.
 
 This repository contains **LoRA adapter weights only**.
 The base model must be loaded separately.
 
 ## Training Objective
 
-This adapter is trained to improve **multi-turn agent task performance**
-on ALFWorld (household tasks) and DBBench (database operations).
+This adapter is trained to improve multi-turn agent task performance
+on ALFWorld (household tasks).
 
-Loss is applied to **all assistant turns** in the multi-turn trajectory,
-enabling the model to learn environment observation, action selection,
+Loss is applied to all assistant turns in the multi-turn trajectory,
+enabling the model to learn observation grounding, action selection,
 tool use, and recovery from errors.
 
+Training data used in this run is ALFWorld only (see `datasets` in YAML).
+Evaluation in the competition includes AgentBench tasks (**DBBench + ALFWorld**) by the organizers.
+
 ## Training Configuration
 
 - Base model: Qwen/Qwen3-4B-Instruct-2507
-- Method: LoRA (full precision base)
+- Method: LoRA (PEFT)
 - Max sequence length: 2048
 - Epochs: 2
-- Learning rate: 2e-06
+- Learning rate: 1.5e-6
 - LoRA: r=64, alpha=128
 
 ## Usage
@@ -51,7 +51,7 @@ from peft import PeftModel
 import torch
 
 base = "Qwen/Qwen3-4B-Instruct-2507"
-adapter = "your_id/your-repo"
+adapter = "KOUJI039/structeval-qwen3-4b-sft-try20"
 
 tokenizer = AutoTokenizer.from_pretrained(base)
 model = AutoModelForCausalLM.from_pretrained(
@@ -64,7 +64,8 @@ model = PeftModel.from_pretrained(model, adapter)
 
 ## Sources & Terms (IMPORTANT)
 
-Training data: u-10bei/sft_alfworld_trajectory_dataset_v5
+Training data:
+- u-10bei/sft_alfworld_trajectory_dataset_v5
 
-Dataset License: MIT License. This dataset is used and distributed under the terms of the MIT License.
-Compliance: Users must comply with the MIT license (including copyright notice) and the base model's original terms of use.
+This repository does NOT redistribute the dataset.
+Users must comply with the dataset license and base model terms.