TurtleBench: A Visual Programming Benchmark in Turtle Geometry

• URL: http://arxiv.org/abs/2411.00264v2
• Date: Fri, 11 Apr 2025 21:25:02 GMT
• Title: TurtleBench: A Visual Programming Benchmark in Turtle Geometry
• Abstract summary: We introduce Turtle, a benchmark designed to evaluate LMMs' capacity to interpret geometric patterns.<n>TurtleBench features tasks with patterned shapes that have underlying algorithmic logic.<n>Our evaluation reveals that leading LMMs struggle significantly with these tasks.
• Score: 14.856377809214747
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Humans have the ability to reason about geometric patterns in images and scenes from a young age. However, developing large multimodal models (LMMs) capable of similar reasoning remains a challenge, highlighting the need for robust evaluation methods to assess these capabilities. We introduce \Turtle, a benchmark designed to evaluate LMMs' capacity to interpret geometric patterns -- given visual examples, textual instructions, or both -- and generate precise code outputs. Inspired by turtle geometry, a notion used to teach children foundational coding and geometric concepts, TurtleBench features tasks with patterned shapes that have underlying algorithmic logic. Our evaluation reveals that leading LMMs struggle significantly with these tasks, with GPT-4o achieving only 19\% accuracy on the simplest tasks and few-shot prompting only marginally improves their performance ($<2\%$). \Turtle highlights the gap between human and AI performance in intuitive and visual geometrical understanding, setting the stage for future research in this area. \Turtle stands as one of the few benchmarks to evaluate the integration of visual understanding and code generation capabilities in LMMs, setting the stage for future research. Code and Dataset for this paper is provided here: \href{https://github.com/sinaris76/TurtleBench}{https://github.com/sinaris76/TurtleBench}

Related papers

• TurtleAI: Benchmarking Multimodal Models for Visual Programming in Turtle Graphics [24.2553229691479]
  Vision-language models (VLMs) have been explored for visual programming, where they generate code to solve visual tasks.<n>We introduce TurtleAI, a benchmark containing 823 tasks curated based on real-world visual programming tasks in the Turtle Graphics domain.<n>We evaluate 20+ VLMs, including GPT-5, GPT-4o, and Qwen2-VL-72B, and find that they struggle significantly, with most achieving success rates below 30%.<n>We propose a data generation technique that requires only a small set of seed samples. Fine-tuning Qwen2-VL-72B on the resulting synthetic data yields an improvement of
  arXiv  Detail & Related papers  (2026-06-02T13:25:05Z)
• Benchmarking PhD-Level Coding in 3D Geometric Computer Vision [43.42170296855238]
  We introduce GeoCodeBench, a benchmark that evaluates coding for 3D vision.<n>We evaluate eight representative open- and closed-source models to reflect the current ecosystem.<n>The best model, GPT-5, attains only 36.6% pass rate.
  arXiv  Detail & Related papers  (2026-03-31T17:50:55Z)
• EyeLayer: Integrating Human Attention Patterns into LLM-Based Code Summarization [15.602166020026045]
  EyeLayer is a lightweight attention-augmentation module that incorporates human eye-gaze patterns.<n>We show that EyeLayer consistently outperforms strong fine-tuning baselines across standard metrics.
  arXiv  Detail & Related papers  (2026-02-25T20:04:39Z)
• TimeLens: Rethinking Video Temporal Grounding with Multimodal LLMs [81.78017865436816]
  We present TimeLens, a systematic investigation into building MLLMs with strong video temporal grounding ability.<n>We first expose critical quality issues in existing VTG benchmarks and introduce TimeLens-Bench, comprising meticulously re-annotated versions of three popular benchmarks.<n>We also address noisy training data through an automated re-annotation pipeline, yielding TimeLens-100K, a large-scale, high-quality training dataset.
  arXiv  Detail & Related papers  (2025-12-16T18:59:58Z)
• CodePlot-CoT: Mathematical Visual Reasoning by Thinking with Code-Driven Images [69.93976232543066]
  We propose CodePlot-CoT, a code-driven Chain-of-Thought paradigm for "thinking with images" in mathematics.<n>To achieve this, we first construct Math-VR, the first large-scale, bilingual dataset and benchmark for Mathematics problems with Visual Reasoning.<n>Our model achieves up to 21% increase over base model on our new benchmark, fully validating the efficacy of our proposed code-driven reasoning paradigm.
  arXiv  Detail & Related papers  (2025-10-13T17:59:55Z)
• ArtifactsBench: Bridging the Visual-Interactive Gap in LLM Code Generation Evaluation [51.297873393639456]
  ArtifactsBench is a framework for automated visual code generation evaluation.<n>Our framework renders each generated artifact and captures its dynamic behavior through temporal screenshots.<n>We construct a new benchmark of 1,825 diverse tasks and evaluate over 30 leading Large Language Models.
  arXiv  Detail & Related papers  (2025-07-07T12:53:00Z)
• MMGeoLM: Hard Negative Contrastive Learning for Fine-Grained Geometric Understanding in Large Multimodal Models [60.20220180316705]
  Large Multimodal Models (LMMs) typically build on ViTs (e.g., CLIP), yet their training with simple random in-batch negatives limits the ability to capture fine-grained visual differences.<n>We propose a novel hard negative contrastive learning framework for the vision encoder, which combines image-based contrastive learning using generation-based hard negatives.<n>We train a vision encoder (CLIP) using our hard negative training method, namely MMCLIP (Multimodal Math CLIP), and subsequently train an LMM for geometric problem-solving.
  arXiv  Detail & Related papers  (2025-05-26T15:55:28Z)
• OmniGenBench: A Benchmark for Omnipotent Multimodal Generation across 50+ Tasks [77.19223035769248]
  Recent breakthroughs in large multimodal models (LMMs) have demonstrated remarkable proficiency in following general-purpose instructions for image generation.<n>We introduce OmniGenBench, a novel benchmark meticulously designed to assess the instruction-following abilities of state-of-the-art LMMs.<n>Our OmniGenBench includes 57 diverse sub-tasks grounded in real-world scenarios, systematically categorized according to the specific model capabilities they demand.
  arXiv  Detail & Related papers  (2025-05-24T16:29:34Z)
• SnipGen: A Mining Repository Framework for Evaluating LLMs for Code [51.07471575337676]
  Language Models (LLMs) are trained on extensive datasets that include code repositories. evaluating their effectiveness poses significant challenges due to the potential overlap between the datasets used for training and those employed for evaluation. We introduce SnipGen, a comprehensive repository mining framework designed to leverage prompt engineering across various downstream tasks for code generation.
  arXiv  Detail & Related papers  (2025-02-10T21:28:15Z)
• Open Eyes, Then Reason: Fine-grained Visual Mathematical Understanding in MLLMs [62.875934732547435]
  Current large language models (MLLMs) often underperform on mathematical problem-solving tasks that require fine-grained visual understanding. In this paper, we evaluate the visual grounding capabilities of state-of-the-art MLLMs and reveal a significant negative correlation between visual grounding accuracy and problem-solving performance. We propose a novel approach, SVE-Math, featuring a geometric-grounded vision encoder and a feature router that dynamically adjusts the contribution of hierarchical visual feature maps.
  arXiv  Detail & Related papers  (2025-01-11T04:08:44Z)
• Euclid: Supercharging Multimodal LLMs with Synthetic High-Fidelity Visual Descriptions [23.294711275107606]
  This paper introduces Geoperception, a benchmark to evaluate an MLLM's ability to accurately transcribe 2D geometric information from an image. We then conduct a comprehensive empirical study to explore strategies for improving their performance on geometric tasks. We develop Euclid, a family of models specifically optimized for strong low-level geometric perception.
  arXiv  Detail & Related papers  (2024-12-11T19:12:13Z)
• ScratchEval: Are GPT-4o Smarter than My Child? Evaluating Large Multimodal Models with Visual Programming Challenges [20.316852491762788]
  We propose ScratchEval, a novel benchmark designed to evaluate the visual programming reasoning ability of LMMs. ScratchEval is based on Scratch, a block-based visual programming language widely used in children's programming education.
  arXiv  Detail & Related papers  (2024-11-28T05:51:45Z)
• GeoCoder: Solving Geometry Problems by Generating Modular Code through Vision-Language Models [10.443672399225983]
  Vision-parametric models (VLMs) have made significant progress in various multimodal tasks. They still struggle with geometry problems and are significantly limited by their inability to perform mathematical operations not seen during pre-training. We present GeoCoder, which leverages modular code-finetuning to generate and execute code using a predefined geometry function library.
  arXiv  Detail & Related papers  (2024-10-17T12:56:52Z)
• HumanEval-V: Evaluating Visual Understanding and Reasoning Abilities of Large Multimodal Models Through Coding Tasks [25.959032350818795]
  HumanEval-V is a benchmark designed to evaluate Large Language Models' visual understanding and reasoning capabilities through code generation. HumanEval-V includes 108 carefully crafted, entry-level Python coding tasks derived from platforms like CodeForces and Stack Overflow. We evaluate 19 state-of-the-art LMMs using HumanEval-V, uncovering significant challenges.
  arXiv  Detail & Related papers  (2024-10-16T09:04:57Z)
• MC-Bench: A Benchmark for Multi-Context Visual Grounding in the Era of MLLMs [61.56904387052982]
  This paper proposes a new visual grounding task called multi-context visual grounding.<n>It aims to localize instances of interest across multiple images based on open-ended text prompts.<n>We benchmark over 20 state-of-the-art MLLMs and foundation models with potential multi-context visual grounding capabilities.
  arXiv  Detail & Related papers  (2024-10-16T07:52:57Z)
• Tangram: A Challenging Benchmark for Geometric Element Recognizing [4.97503948320156]
  Tangram is a novel benchmark designed to evaluate the performance of LMMs on geometric recognition element. Tangram includes 1,080 diverse geometric diagrams sourced from primary and secondary school exams, competitions, and textbooks. It requires models to perform a "simple but interesting" counting task. The overall accuracy of the top performer across all tested models is only 56.8%, marking a significant gap when compared to human performance.
  arXiv  Detail & Related papers  (2024-08-25T14:47:25Z)
• Can Large Language Models Understand Symbolic Graphics Programs? [136.5639211254501]
  Symbolic graphics programs are popular in computer graphics. We create a benchmark for the semantic visual understanding of symbolic graphics programs. We find that LLMs considered stronger at reasoning generally perform better.
  arXiv  Detail & Related papers  (2024-08-15T17:59:57Z)
• VisualAgentBench: Towards Large Multimodal Models as Visual Foundation Agents [50.12414817737912]
  Large Multimodal Models (LMMs) have ushered in a new era in artificial intelligence, merging capabilities in both language and vision to form highly capable Visual Foundation Agents. Existing benchmarks fail to sufficiently challenge or showcase the full potential of LMMs in complex, real-world environments. VisualAgentBench (VAB) is a pioneering benchmark specifically designed to train and evaluate LMMs as visual foundation agents.
  arXiv  Detail & Related papers  (2024-08-12T17:44:17Z)
• Lumen: Unleashing Versatile Vision-Centric Capabilities of Large Multimodal Models [87.47400128150032]
  We propose a novel LMM architecture named Lumen, a Large multimodal model with versatile vision-centric capability enhancement. Lumen first promotes fine-grained vision-language concept alignment. Then the task-specific decoding is carried out by flexibly routing the shared representation to lightweight task decoders.
  arXiv  Detail & Related papers  (2024-03-12T04:13:45Z)
• MARIO: MAth Reasoning with code Interpreter Output -- A Reproducible Pipeline [12.186691561822256]
  We postulate that the inherent nature of large language models (LLMs) presents challenges in modeling mathematical reasoning. This paper introduces a novel math dataset, enhanced with a capability to utilize a Python code interpreter. We propose a tentative, easily replicable protocol for the fine-tuning of math-specific LLMs.
  arXiv  Detail & Related papers  (2024-01-16T08:08:01Z)
• PythonSaga: Redefining the Benchmark to Evaluate Code Generating LLMs [1.9207412600219353]
  We evaluate two popular benchmarks for Python code generation, analyzing their diversity and difficulty. Our findings unveil a critical bias towards a limited set of programming concepts, neglecting most of the other concepts entirely. We propose a novel benchmark, PythonSaga, featuring 185 hand-crafted prompts on a balanced representation of 38 programming concepts.
  arXiv  Detail & Related papers  (2024-01-08T12:36:43Z)
• The Languini Kitchen: Enabling Language Modelling Research at Different Scales of Compute [66.84421705029624]
  We introduce an experimental protocol that enables model comparisons based on equivalent compute, measured in accelerator hours. We pre-process an existing large, diverse, and high-quality dataset of books that surpasses existing academic benchmarks in quality, diversity, and document length. This work also provides two baseline models: a feed-forward model derived from the GPT-2 architecture and a recurrent model in the form of a novel LSTM with ten-fold throughput.
  arXiv  Detail & Related papers  (2023-09-20T10:31:17Z)
• SimTeG: A Frustratingly Simple Approach Improves Textual Graph Learning [131.04781590452308]
  We present SimTeG, a frustratingly Simple approach for Textual Graph learning. We first perform supervised parameter-efficient fine-tuning (PEFT) on a pre-trained LM on the downstream task. We then generate node embeddings using the last hidden states of finetuned LM.
  arXiv  Detail & Related papers  (2023-08-03T07:00:04Z)
• Great Truths are Always Simple: A Rather Simple Knowledge Encoder for Enhancing the Commonsense Reasoning Capacity of Pre-Trained Models [89.98762327725112]
  Commonsense reasoning in natural language is a desired ability of artificial intelligent systems. For solving complex commonsense reasoning tasks, a typical solution is to enhance pre-trained language models(PTMs) with a knowledge-aware graph neural network(GNN) encoder. Despite the effectiveness, these approaches are built on heavy architectures, and can't clearly explain how external knowledge resources improve the reasoning capacity of PTMs.
  arXiv  Detail & Related papers  (2022-05-04T01:27:36Z)
• Self-supervised Geometric Perception [96.89966337518854]
  Self-supervised geometric perception is a framework to learn a feature descriptor for correspondence matching without any ground-truth geometric model labels. We show that SGP achieves state-of-the-art performance that is on-par or superior to the supervised oracles trained using ground-truth labels.
  arXiv  Detail & Related papers  (2021-03-04T15:34:43Z)
• Multi-Task Learning for Dense Prediction Tasks: A Survey [87.66280582034838]
  Multi-task learning (MTL) techniques have shown promising results w.r.t. performance, computations and/or memory footprint. We provide a well-rounded view on state-of-the-art deep learning approaches for MTL in computer vision.
  arXiv  Detail & Related papers  (2020-04-28T09:15:50Z)
• Spatial Pyramid Based Graph Reasoning for Semantic Segmentation [67.47159595239798]
  We apply graph convolution into the semantic segmentation task and propose an improved Laplacian. The graph reasoning is directly performed in the original feature space organized as a spatial pyramid. We achieve comparable performance with advantages in computational and memory overhead.
  arXiv  Detail & Related papers  (2020-03-23T12:28:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.