Multi-Agent Debate¶

Debate workflow simulates a multi-turn discussion between different agents, mostly several solvers and an aggregator. Typically, the solvers generate and exchange their answers, while the aggregator collects and summarizes the answers.

We implement the examples in EMNLP 2024, where two debater agents will discuss a topic in a fixed order, and express their arguments based on the previous debate history. At each round a moderator agent will decide whether the correct answer can be obtained in the current iteration.

import asyncio
import os

from pydantic import Field, BaseModel

from agentscope.agent import ReActAgent
from agentscope.formatter import (
    DashScopeMultiAgentFormatter,
    DashScopeChatFormatter,
)
from agentscope.message import Msg
from agentscope.model import DashScopeChatModel
from agentscope.pipeline import MsgHub

# Prepare a topic
topic = (
    "The two circles are externally tangent and there is no relative sliding. "
    "The radius of circle A is 1/3 the radius of circle B. Circle A rolls "
    "around circle B one trip back to its starting point. How many times will "
    "circle A revolve in total?"
)


# Create two debater agents, Alice and Bob, who will discuss the topic.
def create_solver_agent(name: str) -> ReActAgent:
    """Get a solver agent."""
    return ReActAgent(
        name=name,
        sys_prompt=f"You're a debater named {name}. Hello and welcome to the "
        "debate competition. It's unnecessary to fully agree with "
        "each other's perspectives, as our objective is to find "
        "the correct answer. The debate topic is stated as "
        f"follows: {topic}.",
        model=DashScopeChatModel(
            model_name="qwen-max",
            api_key=os.environ["DASHSCOPE_API_KEY"],
            stream=False,
        ),
        formatter=DashScopeMultiAgentFormatter(),
    )


alice, bob = [create_solver_agent(name) for name in ["Alice", "Bob"]]

# Create a moderator agent
moderator = ReActAgent(
    name="Aggregator",
    sys_prompt=f"""You're a moderator. There will be two debaters involved in a debate competition. They will present their answer and discuss their perspectives on the topic:
``````
{topic}
``````
At the end of each round, you will evaluate both sides' answers and decide which one is correct.""",
    model=DashScopeChatModel(
        model_name="qwen-max",
        api_key=os.environ["DASHSCOPE_API_KEY"],
        stream=False,
    ),
    # Use multiagent formatter because the moderator will receive messages from more than a user and an assistant
    formatter=DashScopeMultiAgentFormatter(),
)


# A structured output model for the moderator
class JudgeModel(BaseModel):
    """The structured output model for the moderator."""

    finished: bool = Field(
        description="Whether the debate is finished.",
    )
    correct_answer: str | None = Field(
        description="The correct answer to the debate topic, only if the debate is finished. Otherwise, leave it as None.",
        default=None,
    )


async def run_multiagent_debate() -> None:
    """Run the multi-agent debate workflow."""
    while True:
        # The reply messages in MsgHub from the participants will be broadcasted to all participants.
        async with MsgHub(participants=[alice, bob, moderator]):
            await alice(
                Msg(
                    "user",
                    "You are affirmative side, Please express your viewpoints.",
                    "user",
                ),
            )
            await bob(
                Msg(
                    "user",
                    "You are negative side. You disagree with the affirmative side. Provide your reason and answer.",
                    "user",
                ),
            )

        # Alice and Bob doesn't need to know the moderator's message, so moderator is called outside the MsgHub.
        msg_judge = await moderator(
            Msg(
                "user",
                "Now you have heard the answers from the others, have the debate finished, and can you get the correct answer?",
                "user",
            ),
            structured_model=JudgeModel,
        )

        if msg_judge.metadata.get("finished"):
            print(
                "\nThe debate is finished, and the correct answer is: ",
                msg_judge.metadata.get("correct_answer"),
            )
            break


asyncio.run(run_multiagent_debate())

Alice: Thank you. As the affirmative side, I will present the argument that when circle A, with a radius 1/3 that of circle B, rolls around circle B one complete trip back to its starting point, it will revolve 4 times in total.

To understand this, let's break down the problem:

- Let the radius of circle B be \( R \).
- Then, the radius of circle A is \( \frac{R}{3} \).

When circle A rolls around circle B, the distance it travels along the circumference of circle B is equal to the circumference of the path it follows, which is the circumference of a circle with a radius of \( R + \frac{R}{3} = \frac{4R}{3} \). The circumference of this path is:
\[ 2\pi \left(\frac{4R}{3}\right) = \frac{8\pi R}{3} \]

Now, the circumference of circle A is:
\[ 2\pi \left(\frac{R}{3}\right) = \frac{2\pi R}{3} \]

The number of revolutions circle A makes as it rolls around the path is the ratio of the distance traveled to the circumference of circle A:
\[ \text{Number of Revolutions} = \frac{\frac{8\pi R}{3}}{\frac{2\pi R}{3}} = \frac{8\pi R}{3} \times \frac{3}{2\pi R} = 4 \]

Therefore, circle A will revolve 4 times as it rolls around circle B and returns to its starting point.
Bob: Thank you. As the negative side, I will argue that when circle A, with a radius 1/3 that of circle B, rolls around circle B one complete trip back to its starting point, it will revolve 3 times in total, not 4.

To understand this, let's revisit and clarify the problem:

- Let the radius of circle B be \( R \).
- Then, the radius of circle A is \( \frac{R}{3} \).

When circle A rolls around circle B, the path it follows is indeed the circumference of a circle with a radius of \( R + \frac{R}{3} = \frac{4R}{3} \). The circumference of this path is:
\[ 2\pi \left(\frac{4R}{3}\right) = \frac{8\pi R}{3} \]

The circumference of circle A is:
\[ 2\pi \left(\frac{R}{3}\right) = \frac{2\pi R}{3} \]

The number of revolutions circle A makes as it rolls along the path is the ratio of the distance traveled to the circumference of circle A:
\[ \text{Number of Revolutions} = \frac{\frac{8\pi R}{3}}{\frac{2\pi R}{3}} = \frac{8\pi R}{3} \times \frac{3}{2\pi R} = 4 \]

However, we must consider that as circle A rolls around circle B, it also rotates around its own center. When circle A completes one full trip around circle B, it has effectively made one additional revolution due to the rotation about its own center. This means that while the linear distance covered suggests 4 revolutions, the actual number of self-revolutions is 3 because the 4th "revolution" is the one that aligns it back to its original orientation relative to circle B.

Therefore, circle A will revolve 3 times as it rolls around circle B and returns to its starting point.
/home/runner/work/agentscope/agentscope/src/agentscope/model/_dashscope_model.py:231: DeprecationWarning: 'required' is not supported by DashScope API. It will be converted to 'auto'.
  warnings.warn(
Aggregator: {
    "type": "tool_use",
    "name": "generate_response",
    "input": {
        "finished": true,
        "correct_answer": "4"
    },
    "id": "call_d8a1ad16286c4d1a92e3ba"
}
system: {
    "type": "tool_result",
    "id": "call_d8a1ad16286c4d1a92e3ba",
    "name": "generate_response",
    "output": [
        {
            "type": "text",
            "text": "Successfully generated response."
        }
    ]
}
Aggregator: The debate has concluded, and the correct answer is that circle A will revolve 4 times as it rolls around circle B and returns to its starting point.

Alice's argument correctly calculates the number of revolutions based on the distance traveled by circle A. The circumference of the path that circle A follows is \(\frac{8\pi R}{3}\), and the circumference of circle A itself is \(\frac{2\pi R}{3}\). The ratio of these two circumferences gives us the number of revolutions:

\[ \text{Number of Revolutions} = \frac{\frac{8\pi R}{3}}{\frac{2\pi R}{3}} = 4 \]

Bob's argument introduces an additional consideration about the self-rotation of circle A, but this does not change the number of revolutions. The 4th revolution is indeed the one that aligns circle A back to its original orientation relative to circle B, but it is still counted as a full revolution.

Therefore, Alice's answer of 4 revolutions is the correct one.

The debate is finished, and the correct answer is:  4

Multi-Agent Debate¶

Further Reading¶