Tootfinder

What’s brewing?
#StudioVibes
#WorkInProgress

Advances in LLMs with Focus on Reasoning, Adaptability, Efficiency and Ethics
Asifullah khan, Muhammad Zaeem Khan, Saleha Jamshed, Sadia Ahmad, Aleesha Zainab, Kaynat Khatib, Faria Bibi, Abdul Rehman
https://arxiv.org/abs/2506.12365

Advances in LLMs with Focus on Reasoning, Adaptability, Efficiency and Ethics
This survey paper outlines the key developments in the field of Large Language Models (LLMs), such as enhancing their reasoning skills, adaptability to various tasks, increased computational efficiency, and ability to make ethical decisions. The techniques that have been most effective in bridging the gap between human and machine communications include the Chain-of-Thought prompting, Instruction Tuning, and Reinforcement Learning from Human Feedback. The improvements in multimodal learning and…

I know as a developer just how hard it is to actually make it work, but perhaps the lesson of the Liquid Glass public reaction fiasco is for Apple to just give users •much• more control over the UI theme.
Between the retina transition, iOS 7, accessibility improvements like variable text size, dark mode support, and the long migration away from hard-coded layouts, we developers have grown used to making more adaptable UIs. Lots of the building blocks for OS-level theming are there, both in SwiftUI and in UIKit.
If Apple said “Themes are coming!” at WWDC 26, developers would adapt.
Assuming we still have a functioning society at that point.
From @…:
https://mastodon.social/@realmacdan/113792374380828124

D-LiFT: Improving LLM-based Decompiler Backend via Code Quality-driven Fine-tuning
Muqi Zou (Jing), Hongyu Cai (Jing), Hongwei Wu (Jing), Zion Leonahenahe Basque (Jing), Arslan Khan (Jing), Berkay Celik (Jing), Dave (Jing), Tian (Fish), Antonio Bianchi (Fish), Ruoyu (Fish), Wang, Dongyan Xu
https://arxiv.org/abs/2506.10125…

D-LiFT: Improving LLM-based Decompiler Backend via Code Quality-driven Fine-tuning
Decompilers, which reconstruct human-readable source code from binary executables, are vital to many security tasks. Yet, despite recent advances, their output often suffers from syntactic and semantic errors and remains difficult to read. Recently, with the advent of large language models (LLMs), researchers began to explore the potential of LLMs to refine decompiler output. Nevertheless, our study of these approaches reveals significant limitations, such as introducing new errors and relying …

Ny demonstration i #Paris nu för att kräva frigivandet av de övriga i besättningen på #Madleen. Oklart om Greta Thunberg kan/tillåts komma hit. Enligt uppgift flyger hon till Sverige ikväll.

Can LLM-based Financial Investing Strategies Outperform the Market in Long Run?
Weixian Waylon Li, Hyeonjun Kim, Mihai Cucuringu, Tiejun Ma
https://arxiv.org/abs/2505.07078

Can LLM-based Financial Investing Strategies Outperform the Market in Long Run?
Large Language Models (LLMs) have recently been leveraged for asset pricing tasks and stock trading applications, enabling AI agents to generate investment decisions from unstructured financial data. However, most evaluations of LLM timing-based investing strategies are conducted on narrow timeframes and limited stock universes, overstating effectiveness due to survivorship and data-snooping biases. We critically assess their generalizability and robustness by proposing FINSABER, a backtesting …

När det gäller energipolitik så pratas det mest om kärn- vs vindkraft. Det pratas mindre om att biobränslen står för en stor del av energitillförseln i Sverige. Det räknas som en förnybar energikälla men är den hållbar? Ja, under vissa förutsättningar säger Naturvårdsverket. https://www.

Energitillförsel
Energi kan inte förstöras utan enbart omvandlas. Använd mängd energi måste därför alltid motsvaras av tillförd mängd energi.

Quantum magnetotransport in monolayer $\mathrm{Pt_{2}HgSe_{3}}$
Muzamil Shah, Imtiaz Khan, Kashif Sabeeh, Muhammad Sabieh Anwar, Reza Asgari
https://arxiv.org/abs/2506.10697

Quantum magnetotransport in monolayer $\mathrm{Pt_{2}HgSe_{3}}$
We present a theoretical framework to investigate quantum magnetotransport in monolayer jacutingaite, focusing on its response to external electric fields and off-resonant circularly polarized laser irradiation. Our analysis reveals a sequence of topological phase transitions triggered by tuning these external parameters. We find that the zeroth LL exhibits spin- and valley-polarized splitting, leading to four distinct peaks in the DOSs for the $K$ and $K'$ valleys. Using the Kubo formalism, we…

In-situ tuning of cavity dissipation and a topological transition in an atom-nanotip-cavity system
Taegyu Ha, Kiyanoush Goudarzi, Dowon Lee, Donggeon Kim, Eunchul Jeong, Uijin Kim, Myunghun Kim, Jinuk Kim, Moonjoo Lee
https://arxiv.org/abs/2506.07800

In-situ tuning of cavity dissipation and a topological transition in an atom-nanotip-cavity system
We theoretically investigate a method for tuning a dissipation rate in an atom-cavity system. By inserting a nanotip into the cavity mode, we estimate that the cavity dissipation rate increases by a factor of approximately 20, due to the scattering loss at the tip. As applications of our in-situ technique, we demonstrate that an exceptional line can be obtained when a single atom or ion is coupled to the resonator. Moreover, the position of an exceptional point is tuned by adjusting the decay r…

SOFT: Selective Data Obfuscation for Protecting LLM Fine-tuning against Membership Inference Attacks
Kaiyuan Zhang, Siyuan Cheng, Hanxi Guo, Yuetian Chen, Zian Su, Shengwei An, Yuntao Du, Charles Fleming, Ashish Kundu, Xiangyu Zhang, Ninghui Li
https://arxiv.org/abs/2506.10424

SOFT: Selective Data Obfuscation for Protecting LLM Fine-tuning against Membership Inference Attacks
Large language models (LLMs) have achieved remarkable success and are widely adopted for diverse applications. However, fine-tuning these models often involves private or sensitive information, raising critical privacy concerns. In this work, we conduct the first comprehensive study evaluating the vulnerability of fine-tuned LLMs to membership inference attacks (MIAs). Our empirical analysis demonstrates that MIAs exploit the loss reduction during fine-tuning, making them highly effective in re…