Tootfinder

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…

Combining this with a text to speech reader would be an interesting thing for sight-challenged folk. https://github.com/ngxson/smolvlm-realtime-webcam

"What Stroustrup wanted was apparently what a lot of people wanted: a language that provided OOP features, such as those offered by Smalltalk, but compiled into binaries that would execute as fast as an executable written in C.
The solution Stroustrup adopted was not to adopt the Smalltalk model, but instead to take inspiration from fellow Scandinavians and to bolt ideas borrowed from Simula into C."

End-to-End Framework for Robot Lawnmower Coverage Path Planning using Cellular Decomposition
Nikunj Shah, Utsav Dey, Kenji Nishimiya
https://arxiv.org/abs/2506.06028

End-to-End Framework for Robot Lawnmower Coverage Path Planning using Cellular Decomposition
Efficient Coverage Path Planning (CPP) is necessary for autonomous robotic lawnmowers to effectively navigate and maintain lawns with diverse and irregular shapes. This paper introduces a comprehensive end-to-end pipeline for CPP, designed to convert user-defined boundaries on an aerial map into optimized coverage paths seamlessly. The pipeline includes user input extraction, coordinate transformation, area decomposition and path generation using our novel AdaptiveDecompositionCPP algorithm, pr…

This https://arxiv.org/abs/2505.23786 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Mind the Gap: A Practical Attack on GGUF Quantization
With the increasing size of frontier LLMs, post-training quantization has become the standard for memory-efficient deployment. Recent work has shown that basic rounding-based quantization schemes pose security risks, as they can be exploited to inject malicious behaviors into quantized models that remain hidden in full precision. However, existing attacks cannot be applied to more complex quantization methods, such as the GGUF family used in the popular ollama and llama.cpp frameworks. In this …