yale-cadmy/qwen3-14B-bargaining-sft

Qwen3-14B Bargaining SFT

A Qwen3-14B model fine-tuned via supervised fine-tuning (SFT) on high-quality bilateral bargaining negotiation transcripts.

Overview

This model was trained as part of the LLM Bilateral Bargaining project, which studies how LLM agents negotiate in structured buyer-seller bargaining games.

Training method: Supervised fine-tuning on negotiation demonstrations generated by GPT-4.1 vs GPT-4.1 in a bilateral bargaining environment. The model learns basic negotiation format, tool usage, and constraint-compliant behavior.

Role: This SFT checkpoint serves as the initialization for subsequent reinforcement learning (RL) training.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model = AutoModelForCausalLM.from_pretrained(
    "yale-cadmy/qwen3-14B-bargaining-sft",
    torch_dtype=torch.bfloat16,
    device_map="auto",
)
tokenizer = AutoTokenizer.from_pretrained("yale-cadmy/qwen3-14B-bargaining-sft")

License

CC-BY-NC-4.0. See the LLM Bilateral Bargaining repository for details.