Tootfinder

Genuinely torn about this. Keeping people safe from a roving right-wing murderer does seem prudent. But I don’t want to send out the message that this is a way to get a protest cancelled.
❝The No Kings coalition that organized protests across the US today is cancelling all demonstrations in Minnesota that have not already started after two lawmakers were shot, Reuters reports.
Indivisible said on Bluesky: “Governor Walz has recommended that we cancel No Kings events across Minnesota because the individual who assassinated a Democratic lawmaker is still at large.”❞
https://www.theguardian.com/us-news/live/2025/jun/14/trump-birthday-military-parade-no-kings-protest-live-updates

»Brot und Spiele – Tech-Konzerne sponsern heimlich Trumps umstrittene Militärparade:
Donald Trump droht Demonstranten mit "sehr großer Gewalt", Datenschützer warnen vor Überwachung und Repressalien durch die Behörden«
Kein weiteren Kommentar dazu, denn es ist so wie schon lange vorhergesehen und wie seit je-und-eh!
😠

Tech-Konzerne sponsern heimlich Trumps umstrittene Militärparade
Donald Trump droht Demonstranten mit "sehr großer Gewalt", Datenschützer warnen vor Überwachung und Repressalien durch die Behörden

Millions in US expected to protest against Trump in ‘No Kings’ demonstrations | US news | The Guardian
https://www.theguardian.com/us-news/2025/jun/13/no-kings-protests

Mitigating Spurious Correlations in LLMs via Causality-Aware Post-Training
Shurui Gui, Shuiwang Ji
https://arxiv.org/abs/2506.09433 https://

Mitigating Spurious Correlations in LLMs via Causality-Aware Post-Training
While large language models (LLMs) have demonstrated remarkable capabilities in language modeling, recent studies reveal that they often fail on out-of-distribution (OOD) samples due to spurious correlations acquired during pre-training. Here, we aim to mitigate such spurious correlations through causality-aware post-training (CAPT). By decomposing a biased prediction into two unbiased steps, known as \textit{event estimation} and \textit{event intervention}, we reduce LLMs' pre-training biases…

Quantum secret sharing in a triangular superconducting quantum network
Haoxiong Yan, Allen Zang, Joel Grebel, Xuntao Wu, Ming-Han Chou, Gustav Andersson, Christopher R. Conner, Yash J. Joshi, Shiheng Li, Jacob M. Miller, Rhys G. Povey, Hong Qiao, Eric Chitambar, Andrew N. Cleland
https://arxiv.org/abs/2506.10878

Quantum secret sharing in a triangular superconducting quantum network
We present a three-node quantum communication testbed with a triangular topology, each side of the triangle formed by a 1.3-meter-long transmission line. We demonstrate state transfer and entanglement generation between any two nodes, generate genuine multipartite entangled GHZ states, and implement quantum secret sharing (QSS) of classical information. Our experiments show that the QSS protocol can effectively detect eavesdropping, ensuring the secure sharing of secrets. This device forms a te…

EXPEREPAIR: Dual-Memory Enhanced LLM-based Repository-Level Program Repair
Fangwen Mu, Junjie Wang, Lin Shi, Song Wang, Shoubin Li, Qing Wang
https://arxiv.org/abs/2506.10484

EXPEREPAIR: Dual-Memory Enhanced LLM-based Repository-Level Program Repair
Automatically repairing software issues remains a fundamental challenge at the intersection of software engineering and AI. Although recent advancements in Large Language Models (LLMs) have demonstrated potential for repository-level repair tasks, current methodologies exhibit two notable limitations: (1) they often address issues in isolation, neglecting to incorporate insights from previously resolved issues, and (2) they rely on static and rigid prompting strategies, which constrain their ab…

FedRAG: A Framework for Fine-Tuning Retrieval-Augmented Generation Systems
Val Andrei Fajardo, David B. Emerson, Amandeep Singh, Veronica Chatrath, Marcelo Lotif, Ravi Theja, Alex Cheung, Izuki Matsubi
https://arxiv.org/abs/2506.09200

FedRAG: A Framework for Fine-Tuning Retrieval-Augmented Generation Systems
Retrieval-augmented generation (RAG) systems have been shown to be effective in addressing many of the drawbacks of relying solely on the parametric memory of large language models. Recent work has demonstrated that RAG systems can be improved via fine-tuning of their retriever and generator models. In this work, we introduce FedRAG, a framework for fine-tuning RAG systems across centralized and federated architectures. FedRAG supports state-of-the-art fine-tuning methods, offering a simple and…

Beyond Overconfidence: Foundation Models Redefine Calibration in Deep Neural Networks
Achim Hekler, Lukas Kuhn, Florian Buettner
https://arxiv.org/abs/2506.09593

Beyond Overconfidence: Foundation Models Redefine Calibration in Deep Neural Networks
Reliable uncertainty calibration is essential for safely deploying deep neural networks in high-stakes applications. Deep neural networks are known to exhibit systematic overconfidence, especially under distribution shifts. Although foundation models such as ConvNeXt, EVA and BEiT have demonstrated significant improvements in predictive performance, their calibration properties remain underexplored. This paper presents a comprehensive investigation into the calibration behavior of foundation mo…

Robust Noise Attenuation via Adaptive Pooling of Transformer Outputs
Greyson Brothers
https://arxiv.org/abs/2506.09215 https://arxiv.…

Robust Noise Attenuation via Adaptive Pooling of Transformer Outputs
We investigate the design of pooling methods used to summarize the outputs of transformer embedding models, primarily motivated by reinforcement learning and vision applications. This work considers problems where a subset of the input vectors contains requisite information for a downstream task (signal) while the rest are distractors (noise). By framing pooling as vector quantization with the goal of minimizing signal loss, we demonstrate that the standard methods used to aggregate transformer…

This https://arxiv.org/abs/2505.18344 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Sample Complexity of Diffusion Model Training Without Empirical Risk Minimizer Access
Diffusion models have demonstrated state-of-the-art performance across vision, language, and scientific domains. Despite their empirical success, prior theoretical analyses of the sample complexity suffer from poor scaling with input data dimension or rely on unrealistic assumptions such as access to exact empirical risk minimizers. In this work, we provide a principled analysis of score estimation, establishing a sample complexity bound of $\widetilde{\mathcal{O}}(ε^{-6})$. Our approach lever…