MuSR: Testing the Limits of Chain-of-thought with Multistep Soft Reasoning

Evaluating models on multistep soft reasoning tasks in the form of free text narratives.

Contributed By

@farrelmahaztra

Code

src/inspect_evals/musr

Paper

https://arxiv.org/abs/2310.16049

Overview

MuSR is a dataset designed for evaluating models on multistep soft reasoning tasks in the form of free text narratives. It is composed of 3 domains: murder_mysteries, object_placements, and team_allocation, which have 250, 256, and 250 instances respectively in the dataset.

Usage

Installation

Install with pip install inspect-evals, or uv sync from a checkout of this repository.

Running evaluations

uv run inspect eval inspect_evals/musr --model openai/gpt-5-nano

You can also import tasks as normal Python objects and run them from python:

from inspect_ai import eval
from inspect_evals.musr import musr
eval(musr)

Drop uv run if you manage dependencies yourself. Log viewing (inspect view) and default-model setup are documented in the Inspect Evals README.

Options

You can control a variety of options from the command line. For example:

uv run inspect eval inspect_evals/musr --limit 10 --sample-shuffle
uv run inspect eval inspect_evals/musr --max-connections 10
uv run inspect eval inspect_evals/musr --temperature 0.5

See uv run inspect eval --help for all available options.

Parameters

musr

  • domain (Literal['murder_mysteries', 'object_placements', 'team_allocation']): Which domain in the dataset to evaluate. (default: 'murder_mysteries')
  • prompt_technique (Literal['regular', 'cot', 'cot+']): The prompt technique to use. “regular” includes only the narrative and the question. “cot” uses chain-of-thought prompting. “cot+” includes a hint. (default: 'regular')
  • example_count (int): Number of solved examples to include at the beginning of each prompt. Defaults to 0. Currently only supports 1 example. (default: 0)
  • shuffle (bool): Whether we shuffle the dataset for evaluation (default: True)

Dataset

Here is a truncated example from the dataset:

The tension between them was palpable. Alice had been awarded a major journalist award that Gabrielle had desired. This only deepened their rivalry, with Gabrielle feeling overlooked for this recognition in the Jazz scene.

Winston cast his gaze over the club once more—a hub of pulsating rhythms now eerily silent.

A significant part of the evening was Gabrielle's recorded interview with Alice. It played on the local radio, their professional rivalry subtly echoing under their professional demeanor.

With a deep breath, Winston knew he had a tall task ahead. The jazz club, where Alice was last seen alive was now shrouded in an eerie silence, the vibrant rhythms of what used to be a lively night echoing in the abandoned stage. It was up to him to piece together the missing notes and bring the symphony of this unsolved case to a satisfying finale.

Who is the most likely murderer?

Pick one of the following choices:
A - Eugene
B - Gabrielle

The model is tasked to answer the question and choose the appropriate option.

Scoring

The prompts are based on the official MuSR repository. The in-built multiple_choice scorer is used for evaluation.

Changelog

[2-A] - 2026-02-16

  • Migrate version to new scheme. See #907.

[1.0.1] - 2025-12-18

  • Adds backoff policy for functions that connect to huggingface servers.