# Structured, flexible, and robust: benchmarking and improving large language models towards more human-like behavior in out-of-distribution reasoning tasks

arXiv: https://arxiv.org/abs/2205.05718 · HTML: https://mlsanity.com/paper/2205.05718/

## Key facts

- Cited by 56 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 10 of its 18 references that are in the map.
- First posted on arXiv 11 May 2022.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 10 of 10)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [Scaling Language Models: Methods, Analysis & Insights from Training Gopher](https://mlsanity.com/paper/2112.11446/) (2021)
- [Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents](https://mlsanity.com/paper/2201.07207/) (2022)
- [DreamCoder: Growing generalizable, interpretable knowledge with wake-sleep Bayesian program learning](https://mlsanity.com/paper/2006.08381/) (2020)
- [Skill Induction and Planning with Latent Language](https://mlsanity.com/paper/2110.01517/) (2021)
- [Improving Coherence and Consistency in Neural Sequence Models with Dual-System, Neuro-Symbolic Reasoning](https://mlsanity.com/paper/2107.02794/) (2021)
- [Leveraging Language to Learn Program Abstractions and Search Heuristics](https://mlsanity.com/paper/2106.11053/) (2021)

## Built on by (most cited first, 30 of 56)

- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://mlsanity.com/paper/2303.12712/) (2023)
- [Using cognitive psychology to understand GPT-3](https://mlsanity.com/paper/2206.14576/) (2022)
- [Generalized Planning in PDDL Domains with Pretrained Large Language Models](https://mlsanity.com/paper/2305.11014/) (2023)
- [Harms from Increasingly Agentic Algorithmic Systems](https://mlsanity.com/paper/2302.10329/) (2023)
- [Translating Natural Language to Planning Goals with Large-Language Models](https://mlsanity.com/paper/2302.05128/) (2023)
- [From Word Models to World Models: Translating from Natural Language to the Probabilistic Language of Thought](https://mlsanity.com/paper/2306.12672/) (2023)
- [Thor: Wielding Hammers to Integrate Language Models and Automated Theorem Provers](https://mlsanity.com/paper/2205.10893/) (2022)
- [Beyond Accuracy: Evaluating the Reasoning Behavior of Large Language Models -- A Survey](https://mlsanity.com/paper/2404.01869/) (2024)
- [A Survey of AI Agent Protocols](https://mlsanity.com/paper/2504.16736/) (2025)
- [Eight Things to Know about Large Language Models](https://mlsanity.com/paper/2304.00612/) (2023)
- [CogBench: a large language model walks into a psychology lab](https://mlsanity.com/paper/2402.18225/) (2024)
- [Can Language Models Learn to Skip Steps?](https://mlsanity.com/paper/2411.01855/) (2024)
- [Large Language Models and Cognitive Science: A Comprehensive Review of Similarities, Differences, and Challenges](https://mlsanity.com/paper/2409.02387/) (2024)
- [Distilling Script Knowledge from Large Language Models for Constrained Language Planning](https://mlsanity.com/paper/2305.05252/) (2023)
- [A Scalable Communication Protocol for Networks of Large Language Models](https://mlsanity.com/paper/2410.11905/) (2024)
- [LLMs as Planning Formalizers: A Survey for Leveraging Large Language Models to Construct Automated Planning Models](https://mlsanity.com/paper/2503.18971/) (2025)
- [Graph-enhanced Large Language Models in Asynchronous Plan Reasoning](https://mlsanity.com/paper/2402.02805/) (2024)
- [MacGyver: Are Large Language Models Creative Problem Solvers?](https://mlsanity.com/paper/2311.09682/) (2023)
- [Dissociating language and thought in large language models](https://mlsanity.com/paper/2301.06627/) (2023)
- [AI for social science and social science of AI: A Survey](https://mlsanity.com/paper/2401.11839/) (2024)
- [On Benchmarking Human-Like Intelligence in Machines](https://mlsanity.com/paper/2502.20502/) (2025)
- [A Survey of Robotic Language Grounding: Tradeoffs between Symbols and Embeddings](https://mlsanity.com/paper/2405.13245/) (2024)
- [The Neuro-Symbolic Inverse Planning Engine (NIPE): Modeling Probabilistic Social Inferences from Linguistic Inputs](https://mlsanity.com/paper/2306.14325/) (2023)
- [NL2Plan: Robust LLM-Driven Planning from Minimal Text Descriptions](https://mlsanity.com/paper/2405.04215/) (2024)
- [Large Language Models: The Need for Nuance in Current Debates and a Pragmatic Perspective on Understanding](https://mlsanity.com/paper/2310.19671/) (2023)
- [Dark Patterns Meet GUI Agents: LLM Agent Susceptibility to Manipulative Interfaces and the Role of Human Oversight](https://mlsanity.com/paper/2509.10723/) (2025)
- [Evaluating statistical language models as pragmatic reasoners](https://mlsanity.com/paper/2305.01020/) (2023)
- [PROC2PDDL: Open-Domain Planning Representations from Texts](https://mlsanity.com/paper/2403.00092/) (2024)
- [PDDLEGO: Iterative Planning in Textual Environments](https://mlsanity.com/paper/2405.19793/) (2024)
- [Modeling Open-World Cognition as On-Demand Synthesis of Probabilistic Models](https://mlsanity.com/paper/2507.12547/) (2025)

## Abstract (arXiv)

Human language offers a powerful window into our thoughts -- we tell stories, give explanations, and express our beliefs and goals through words. Abundant evidence also suggests that language plays a developmental role in structuring our learning. Here, we ask: how much of human-like thinking can be captured by learning statistical patterns in language alone? We first contribute a new challenge benchmark for comparing humans and distributional large language models (LLMs). Our benchmark contains two problem-solving domains (planning and explanation generation) and is designed to require generalization to new, out-of-distribution problems expressed in language. We find that humans are far more robust than LLMs on this benchmark. Next, we propose a hybrid Parse-and-Solve model, which augments distributional LLMs with a structured symbolic reasoning module. We find that this model shows more robust adaptation to out-of-distribution planning problems, demonstrating the promise of hybrid AI models for more human-like reasoning.