Tootfinder

Memory in Large Language Models: Mechanisms, Evaluation and Evolution
Dianxing Zhang, Wendong Li, Kani Song, Jiaye Lu, Gang Li, Liuchun Yang, Sheng Li
https://arxiv.org/abs/2509.18868

Memory in Large Language Models: Mechanisms, Evaluation and Evolution
Under a unified operational definition, we define LLM memory as a persistent state written during pretraining, finetuning, or inference that can later be addressed and that stably influences outputs. We propose a four-part taxonomy (parametric, contextual, external, procedural/episodic) and a memory quadruple (location, persistence, write/access path, controllability). We link mechanism, evaluation, and governance via the chain write -> read -> inhibit/update. To avoid distorted comparisons acr…

»Zenit erreicht — Social-Media-Konsum nimmt weltweit ab:
Immer weniger Personen geben an, soziale Medien für den Austausch mit anderen zu nutzen«
Dies teile ich euch nun im Social-Media mit aber in der kann Mensch noch real Reagieren auf chronologische so wie menschliche Posts. Abgesehen davon, so wichtig sind die nicht ausser eben für Propaganda.
🤷

Zenit erreicht: Social Media-Konsum nimmt weltweit ab
Immer weniger Personen geben an, soziale Medien für den Austausch mit anderen Menschen zu nutzen

M4SER: Multimodal, Multirepresentation, Multitask, and Multistrategy Learning for Speech Emotion Recognition
Jiajun He, Xiaohan Shi, Cheng-Hung Hu, Jinyi Mi, Xingfeng Li, Tomoki Toda
https://arxiv.org/abs/2509.18706

Time for authors to get off Audible (and ideally Amazon as well) before you'll find yourself trapped in a death spiral.
And yes, I know how hard that would hit most. But if Amazon is able to destroy competitors, nothing will stop them from exploiting you even more.
https://www.instagram.com/r…

Danish Minister of Justice and chief architect of the current Chat Control proposal, Peter Hummelgaard:
"We must break with the totally erroneous perception that it is everyone's civil liberty to communicate on encrypted messaging services."
Ok, I agree but on one condition: "everyone" means EVERYONE. Police, military, government, politicians, lobbyists, 100% 7x24 transparency. No closed door meetings, no off the record, no sly meets in a pub. Everyone.
Anything else is sinister.

AudioGenie-Reasoner: A Training-Free Multi-Agent Framework for Coarse-to-Fine Audio Deep Reasoning
Yan Rong, Chenxing Li, Dong Yu, Li Liu
https://arxiv.org/abs/2509.16971 https:…

AudioGenie-Reasoner: A Training-Free Multi-Agent Framework for Coarse-to-Fine Audio Deep Reasoning
Audio deep reasoning is a challenging task that requires expert-level perception, multi-step logical inference, and the integration of contextual knowledge. However, existing models suffer from a gap between audio perception and reasoning abilities due to the lack of training data with explicit reasoning chains and the absence of mechanisms for active exploration and iterative refinement. To address these challenges, we propose AudioGenie-Reasoner (AGR), the first unified training-free multi-ag…

Teaching RDM in a smart advanced inorganic lab course and its provision in the DALIA platform
Alexander Hoffmann, Jochen Ortmeyer, Fabian Fink, Charles Tapley Hoyt, Jonathan D. Geiger, Paul Kehrein, Torsten Schrade, Sonja Herres-Pawlis
https://arxiv.org/abs/2509.18902

Enviei para os alunos as notas da turma de Lógica, agora só falta corrigir as provas de uma turma. Mas essa eu vou deixar pra amanhã, vou alimentar os gatinhos da rua e deitar.

Sparse-Autoencoder-Guided Internal Representation Unlearning for Large Language Models
Tomoya Yamashita, Akira Ito, Yuuki Yamanaka, Masanori Yamada, Takayuki Miura, Toshiki Shibahara
https://arxiv.org/abs/2509.15631

Sparse-Autoencoder-Guided Internal Representation Unlearning for Large Language Models
As large language models (LLMs) are increasingly deployed across various applications, privacy and copyright concerns have heightened the need for more effective LLM unlearning techniques. Many existing unlearning methods aim to suppress undesirable outputs through additional training (e.g., gradient ascent), which reduces the probability of generating such outputs. While such suppression-based approaches can control model outputs, they may not eliminate the underlying knowledge embedded in the…

Implicit Kinodynamic Motion Retargeting for Human-to-humanoid Imitation Learning
Xingyu Chen, Hanyu Wu, Sikai Wu, Mingliang Zhou, Diyun Xiang, Haodong Zhang
https://arxiv.org/abs/2509.15443

Implicit Kinodynamic Motion Retargeting for Human-to-humanoid Imitation Learning
Human-to-humanoid imitation learning aims to learn a humanoid whole-body controller from human motion. Motion retargeting is a crucial step in enabling robots to acquire reference trajectories when exploring locomotion skills. However, current methods focus on motion retargeting frame by frame, which lacks scalability. Could we directly convert large-scale human motion into robot-executable motion through a more efficient approach? To address this issue, we propose Implicit Kinodynamic Motion R…