NL2Scratch: An Executable Benchmark and Evaluation for Block-Based Programming

• URL: http://arxiv.org/abs/2606.22061v1
• Date: Sat, 20 Jun 2026 14:22:59 GMT
• Title: NL2Scratch: An Executable Benchmark and Evaluation for Block-Based Programming
• Abstract summary: We introduce NL2Scratch, an executable benchmark for natural-language-to-Scratch generation.<n>We construct a semantically validated pool of 23,594 examples, and a slot-balanced 800 diagnostic benchmark.<n>Experiments across instruction-tuned and fine-tuned LLMs reveal a notable gap between lexical similarity and semantic alignment.
• Score: 62.89531126732269
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Block-based programming environments such as Scratch are widely used in early programming education, yet natural-language-to-code (NL2Code) research has focused primarily on text-based languages. Scratch programs are event-driven, visually compositional, and distributed across concurrent scripts, making conventional NL2Code assumptions and evaluation insufficient. We introduce NL2Scratch, an executable benchmark for natural-language-to-Scratch generation comprising 311,648 parser-valid NL--program pairs, whose program side is extracted from real Scratch projects and paired with semantically aligned NL descriptions. For reliable evaluation beyond surface overlap, we propose Semantic Alignment Consistency (SAC), an interpretable slot-level metric for measuring semantic agreement between descriptions and programs. With SAC, we construct a semantically validated pool of 23,594 examples, and a slot-balanced 800 diagnostic benchmark. Experiments across instruction-tuned and fine-tuned LLMs reveal a notable gap between lexical similarity and semantic alignment: models achieving token-level F1 above 0.93 often fail to attain perfect SAC, particularly on longer examples. Errors concentrate on operational slots like actions, conditions, and numeric arguments, exposing failure modes largely invisible under conventional metrics.

Related papers

• Evaluating Language Models on Cross-Language Code Functional Equivalence [6.76025470599609]
  We introduce PolyHuman, a dataset of human-written programs in CPP, Java, and Python.<n>Using this dataset, we evaluate intra- and inter-language equivalence detection across open-weight and proprietary LLMs.<n>We analyze 81 cases of systematic disagreement in which models incorrectly judge functional equivalence.
  arXiv  Detail & Related papers  (2026-08-25T01:41:00Z)
• Weave of Formal Thought [51.56484100374058]
  We introduce Weave of Formal Thought (WoFT), a paradigm uniting rigorous syntactic validation with learned structural representations.<n>We present a latent-variable fine-tuning method training the language model to interleave non-terminal grammar symbols directly into generation.<n>For Python, fine-tuning StarCoder2-3B with our RWS objective reduces per-token cross-entropy by 14.3% relative to a text-only SFT baseline.
  arXiv  Detail & Related papers  (2026-06-24T15:58:11Z)
• What Really Improves Mathematical Reasoning: Structured Reasoning Signals Beyond Pure Code [72.9921566968371]
  We revisit the claim that code improves reasoning through controlled pretraining experiments on a 10T-token corpus with fine-grained domain separation.<n>Code substantially improves programming ability but does not act as a general reasoning enhancer.<n>Cross-domain structured reasoning traces, such as code-text and math-text mixtures, are better explained.
  arXiv  Detail & Related papers  (2026-05-19T12:37:01Z)
• PLSemanticsBench: Large Language Models As Programming Language Interpreters [31.611330217819713]
  As large language models (LLMs) excel at code reasoning, a natural question arises: can an LLM execute programs (i.e., act as an interpreter) purely based on a programming language's formal semantics?<n>We study this question using the imperative language IMP, formalized via small-step operational semantics (SOS) and rewriting-based operational semantics (K-semantics)<n>We introduce three evaluation sets-Human-Written, LLM-Translated, and Fuzzer- Generated-whose difficulty is controlled by code-complexity metrics spanning the size, control-flow, and data-flow axes.
  arXiv  Detail & Related papers  (2025-10-03T18:23:26Z)
• EquiBench: Benchmarking Large Language Models' Reasoning about Program Semantics via Equivalence Checking [58.15568681219339]
  We introduce EquiBench, a new benchmark for evaluating large language models (LLMs)<n>This task directly tests a model's ability to reason about program semantics.<n>We evaluate 19 state-of-the-art LLMs and find that in the most challenging categories, the best accuracies are 63.8% and 76.2%, only modestly above the 50% random baseline.
  arXiv  Detail & Related papers  (2025-02-18T02:54:25Z)
• Synthetic Programming Elicitation for Text-to-Code in Very Low-Resource Programming and Formal Languages [21.18996339478024]
  We introduce emphsynthetic programming elicitation and compilation (SPEAC) SPEAC produces syntactically correct programs more frequently and without sacrificing semantic correctness. We empirically evaluate the performance of SPEAC in a case study for the UCLID5 formal verification language.
  arXiv  Detail & Related papers  (2024-06-05T22:16:19Z)
• LEVER: Learning to Verify Language-to-Code Generation with Execution [64.36459105535]
  We propose LEVER, a simple approach to improve language-to-code generation by learning to verify the generated programs with their execution results. Specifically, we train verifiers to determine whether a program sampled from the LLMs is correct or not based on the natural language input, the program itself and its execution results. LEVER consistently improves over the base code LLMs(4.6% to 10.9% with code-davinci) and achieves new state-of-the-art results on all of them.
  arXiv  Detail & Related papers  (2023-02-16T18:23:22Z)
• Synchromesh: Reliable code generation from pre-trained language models [38.15391794443022]
  We propose Synchromesh: a framework for substantially improving the reliability of pre-trained models for code generation. First, it retrieves few-shot examples from a training bank using Target Similarity Tuning (TST), a novel method for semantic example selection. Then, Synchromesh feeds the examples to a pre-trained language model and samples programs using Constrained Semantic Decoding (CSD), a general framework for constraining the output to a set of valid programs in the target language.
  arXiv  Detail & Related papers  (2022-01-26T22:57:44Z)
• CodeBLEU: a Method for Automatic Evaluation of Code Synthesis [57.87741831987889]
  In the area of code synthesis, the commonly used evaluation metric is BLEU or perfect accuracy. We introduce a new automatic evaluation metric, dubbed CodeBLEU. It absorbs the strength of BLEU in the n-gram match and further injects code syntax via abstract syntax trees (AST) and code semantics via data-flow.
  arXiv  Detail & Related papers  (2020-09-22T03:10:49Z)

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.