arXiv:2610.19238v1 Announce Type: new Abstract: When writing code, the same computation can be expressed in different programming languages and through different implementation strategies. We show th...
The paper introduces Program Executability Prediction (PrEx), a task that asks large language models (LLMs) to determine whether a program is semantically valid or invalid and, if invalid, to identify the violated formal rule. To evaluate this, the authors create a dataset of systematically generated invalid programs derived from valid ones and test open‑source coding LLMs across different semantic formalisms, semantic shifts, and program splits (human‑written, LLM‑translated, fuzzer‑generated). Results show that LLMs rely more on pre‑training priors than on the provided semantics, performing poorly on modified semantics and with increasing program complexity.
arXiv:2607. 13921v1 Announce Type: cross Abstract: Languages with rich static semantics, such as Rust, provide stronger guarantees for AI-generated code, but their strictness makes generation more difficult.
arXiv:2606.12764v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used to generate code at scale. Meanwhile, prior work has investigated whether training data may be r...
arXiv:2505. 03818v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) can achieve strong performance on everyday coding tasks, but they can fail on complex tasks that require non-trivial reasoning about program semantics.
Large Language Models (LLMs) treat all context tokens uniformly, which creates prompt injection vulnerabilities and unreliable instruction execution. The paper introduces LLMON (LLM Object Notation), a post‑training framework that separates executable instructions from passive data and requires explicit identifier‑based execution. Using a 4.6 B‑token structured corpus and post‑training on Granite‑4.1 models, LLMON reduces indirect prompt‑injection success rates from 53–100 % to 0.02–5.6 % while maintaining general capabilities, and provides attention‑based interpretability metrics to diagnose failure modes.
arXiv:2610.15290v1 Announce Type: cross Abstract: We present Co-DSL, a framework in which users and LLMs co-design reusable spatial concepts for layout generation. In the first stage, our system allo...
The paper introduces "compile by training," a method that converts natural-language specifications into reusable neural functions. By generating task-specific examples with teacher models at compile time, a small adapter is trained for a compact interpreter, eliminating the need for remote model calls during execution. The approach achieves 83.6% semantic accuracy on FuzzyBench-Hard, though it incurs a higher compile-time cost of about a minute, and is demonstrated in applications such as a multi-site website helper, a language-controlled 3D avatar, and a bidirectional English‑Claudish translator.
arXiv:2602. 08857v2 Announce Type: replace-cross Abstract: Recent work has shown that the computations of Transformers can be simulated in the RASP family of programming languages.
The paper introduces LLMON, an LLM-native markup language designed to encode the structure and semantic metadata of textual inputs for large language models. By distinguishing components such as instructions and data within prompts, LLMON aims to reduce model confusion and mitigate security risks like prompt injection. The authors outline design requirements, present the language, and provide preliminary evidence that incorporating LLMON can improve model accuracy, safety, and security during training and inference.
SLMFix is a code‑generation pipeline that uses a small language model fine‑tuned with reinforcement learning to correct syntactic errors in programs produced by large language models for domain‑specific languages. The approach relies on interpreter feedback to guide the error‑fixing process. Experiments show that SLMFix improves validator pass rates by 40% on low‑resource programming languages and removes over 50% of syntactic errors on high‑resource DSLs, outperforming supervised fine‑tuning even for 7B models.
arXiv:2608.22638v1 Announce Type: cross Abstract: Copying a function from a chat window into an editor takes less than a second. For many uses of AI coding tools, that speed is the point; in settings...