Tootfinder

Mix-of-Language-Experts Architecture for Multilingual Programming
Yifan Zong, Yuntian Deng, Pengyu Nie
https://arxiv.org/abs/2506.18923 https://

Mix-of-Language-Experts Architecture for Multilingual Programming
Large language models (LLMs) have demonstrated impressive capabilities in aiding developers with tasks like code comprehension, generation, and translation. Supporting multilingual programming -- i.e., coding tasks across multiple programming languages -- typically requires either (1) finetuning a single LLM across all programming languages, which is cost-efficient but sacrifices language-specific specialization and performance, or (2) finetuning separate LLMs for each programming language, whi…

Steering Conceptual Bias via Transformer Latent-Subspace Activation
Vansh Sharma, Venkat Raman
https://arxiv.org/abs/2506.18887 https://

Steering Conceptual Bias via Transformer Latent-Subspace Activation
This work examines whether activating latent subspaces in language models (LLMs) can steer scientific code generation toward a specific programming language. Five causal LLMs were first evaluated on scientific coding prompts to quantify their baseline bias among four programming languages. A static neuron-attribution method, perturbing the highest activated MLP weight for a C++ or CPP token, proved brittle and exhibited limited generalization across prompt styles and model scales. To address th…

Text-Driven 3D Hand Motion Generation from Sign Language Data
L\'eore Bensabath, Mathis Petrovich, G\"ul Varol
https://arxiv.org/abs/2508.15902 https://

Text-Driven 3D Hand Motion Generation from Sign Language Data
Our goal is to train a generative model of 3D hand motions, conditioned on natural language descriptions specifying motion characteristics such as handshapes, locations, finger/hand/arm movements. To this end, we automatically build pairs of 3D hand motions and their associated textual labels with unprecedented scale. Specifically, we leverage a large-scale sign language video dataset, along with noisy pseudo-annotated sign categories, which we translate into hand motion descriptions via an LLM…

NoCode-bench: A Benchmark for Evaluating Natural Language-Driven Feature Addition
Le Deng, Zhonghao Jiang, Jialun Cao, Michael Pradel, Zhongxin Liu
https://arxiv.org/abs/2507.18130

NoCode-bench: A Benchmark for Evaluating Natural Language-Driven Feature Addition
Natural language-driven no-code development allows users to specify software functionality using natural language (NL) instead of editing source code, promising increased productivity and democratized development. Large language models (LLMs) show potential in enabling this paradigm. In this context, software documentation acts as an NL specification for functionality. This work introduces NoCode-bench, a benchmark designed to evaluate LLMs on real-world NL-driven feature addition tasks, consis…

Universal properties of the many-body Lanczos algorithm at finite size
Luca Capizzi, Leonardo Mazza, Sara Murciano
https://arxiv.org/abs/2507.17424 https://

Universal properties of the many-body Lanczos algorithm at finite size
We study the universal properties of the Lanczos algorithm applied to finite-size many-body quantum systems. Focusing on autocorrelation functions of local operators and on their infinite-time behaviour at finite size, we conjecture that in the large $n$ limit, the ratios between consecutive Lanczos coefficients should have specific scalings with the size of the lattice that we make precise and that depend on the hydrodynamic tail of the autocorrelation function. The scaling associated with str…

Google launches Web Guide, an AI-powered Search Labs experiment that organizes Google Search results by grouping pages related to specific aspects of the query (Sarah Perez/TechCrunch)
https://techcrunch.com/2025/07/24/googles-new-web-view-…

Google's new Web View search experiment organizes results with AI | TechCrunch
Google's new Search Labs experiment will organize the search results using AI smarts.

TimelyHLS: LLM-Based Timing-Aware and Architecture-Specific FPGA HLS Optimization
Nowfel Mashnoor, Mohammad Akyash, Hadi Kamali, Kimia Azar
https://arxiv.org/abs/2507.17962 http…

TimelyHLS: LLM-Based Timing-Aware and Architecture-Specific FPGA HLS Optimization
Achieving timing closure and design-specific optimizations in FPGA-targeted High-Level Synthesis (HLS) remains a significant challenge due to the complex interaction between architectural constraints, resource utilization, and the absence of automated support for platform-specific pragmas. In this work, we propose TimelyHLS, a novel framework integrating Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) to automatically generate and iteratively refine HLS code optimized for…

Lost in Translation? Converting RegExes for Log Parsing into Dynatrace Pattern Language
Julian Fragner, Christian Macho, Bernhard Dieber, Martin Pinzger
https://arxiv.org/abs/2506.19539

Lost in Translation? Converting RegExes for Log Parsing into Dynatrace Pattern Language
Log files provide valuable information for detecting and diagnosing problems in enterprise software applications and data centers. Several log analytics tools and platforms were developed to help filter and extract information from logs, typically using regular expressions (RegExes). Recent commercial log analytics platforms provide domain-specific languages specifically designed for log parsing, such as Grok or the Dynatrace Pattern Language (DPL). However, users who want to migrate to these p…

Synthesizing DSLs for Few-Shot Learning
Paul Krogmeier, P. Madhusudan
https://arxiv.org/abs/2508.16063 https://arxiv.org/pdf/2508.16063

Synthesizing DSLs for Few-Shot Learning
We study the problem of synthesizing domain-specific languages (DSLs) for few-shot learning in symbolic domains. Given a base language and instances of few-shot learning problems, where each instance is split into training and testing samples, the DSL synthesis problem asks for a grammar over the base language that guarantees that small expressions solving training samples also solve corresponding testing samples. We prove that the problem is decidable for a class of languages whose semantics o…

Google launches its virtual clothes try-on feature, letting users upload photos of themselves, in the US and updates price alerts to let users specify an amount (Aisha Malik/TechCrunch)
https://techcrunch.com/2025/07/24/googles-new-ai-feature-lets-…

Google's new AI feature lets you virtually try on clothes | TechCrunch
The feature is launching in the United States today, letting users try on apparel items in Google's Shopping Graph across Search, Google Shopping and product results on Google Images.