Tootfinder

"Critically endangered chameleon discovered outside its known habitat in Madagascar"
#Madagascar #Animals

Critically endangered chameleon discovered outside its known habitat in Madagascar
ANTANANARIVO — At the end of April, a team of researchers scouring a scarred area of spiny forest in southwestern Madagascar came across three individuals of a critically endangered chameleon species. Furcifer belalandaensis had not previously been recorded outside of a tiny, 4-square-kilometer (1.5-square-mile) area near the village of Belalanda after which it’s named, so […]

CrowdStrike released its annual threat hunting report showing that we're in a new era of cyberattacks from sophisticated attackers who rely on clever social engineering, AI, and devices outside of IT's purview.
Check out my latest CSO piece for more details and tips on how defenders can better guard against these threats.
CrowdStrike: A new era of cyberthreats from sophisticated threat actors is here

CrowdStrike: A new era of cyberthreats from sophisticated threat actors is here
The company’s latest threat hunting report highlights the speed and AI sophistication of threat groups today, offering defenders strategies for keeping up.

Meta-Policy Reflexion: Reusable Reflective Memory and Rule Admissibility for Resource-Efficient LLM Agent
Chunlong Wu, Zhibo Qu
https://arxiv.org/abs/2509.03990 https://

Meta-Policy Reflexion: Reusable Reflective Memory and Rule Admissibility for Resource-Efficient LLM Agent
Large language model (LLM) agents achieve impressive single-task performance but commonly exhibit repeated failures, inefficient exploration, and limited cross-task adaptability. Existing reflective strategies (e.g., Reflexion, ReAct) improve per-episode behavior but typically produce ephemeral, task-specific traces that are not reused across tasks. Reinforcement-learning based alternatives can produce transferable policies but require substantial parameter updates and compute. In this work we …

Replaced article(s) found for cs.CV. https://arxiv.org/list/cs.CV/new
[1/4]:
- CPCL: Cross-Modal Prototypical Contrastive Learning for Weakly Supervised Text-based Person Retri...
Xinpeng Zhao, Yanwei Zheng, Chuanlin Lan, Xiaowei Zhang, Bowen Huang, Jibin Yang, Dongxiao Yu

Finite $N$ precursors of the free cumulants
Sylvain Lacroix, Jean-Bernard Zuber
https://arxiv.org/abs/2508.21483 https://arxiv.org/pdf/2508.21483

Finite $N$ precursors of the free cumulants
We introduce $\mathrm{U}(N)$ invariant polynomials on the space of $N\times N$ matrices that are precursors of free cumulants in various respects. First, they are polynomials of deterministic matrices, that are not yet evaluated over some probability law, contrary to what is usually meant by cumulants. Secondly, they converge towards the algebraic expression of free cumulants in terms of moments as $N\to \infty$, with $1/N^2$ corrections expressed in terms of monotone Hurwitz numbers. Their mos…

Check out my latest CSO piece, which examines how Trump’s nominees for national cyber director and CISA director will face shared challenges but different prospects in coping with unprecedented reductions in cyber personnel across the federal government.
Many thanks to Michael Daniel of the Cyber Threat Alliance and Mark Montgomery of FDD for their help and insight.
Daniel made a point about Trump's cutbacks on cyber spending, a situation that most cyber folks haven't see…

Top US cyber officials face divergent paths after Senate confirmation
Following their Senate confirmations, Sean Cairncross and Sean Plankey — Trump’s nominees for national cyber director and CISA director — will face shared challenges but different prospects in coping with unprecedented reductions in cyber personnel across the federal government.

E2ATST: A Temporal-Spatial Optimized Energy-Efficient Architecture for Training Spiking Transformer
Yunhao Ma, Yanyu Lin, Mingjing Li, Puli Quan, Chenlin Zhou, Wenyue Zhang, Zhiwei Zhong, Wanyi Jia, Xueke Zhu, Qingyan Meng, Huihui Zhou, Fengwei An
https://arxiv.org/abs/2508.00475

E2ATST: A Temporal-Spatial Optimized Energy-Efficient Architecture for Training Spiking Transformer
(1) Pengcheng Laboratory, (2) Southern University of Science and Technology, (3) Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, (4) University of Chinese Academy of Sciences

Na das üben wir aber nochmal, lieber {$randomVendor}. Ich kaufe ein "f". XD
"Lieber {$channelName}, Ich hoffe, diese E-Mail bringt dir gute Wünsche.[…]Deshalb möchte ich Sie einladen, ein YouTube-Shopping-Video für XXXXX-Produkte zu erstellen."

Monochromatic components with many edges in random graphs
Hannah Fox, Sammy Luo
https://arxiv.org/abs/2509.01766 https://arxiv.org/pdf/2509.01766

Monochromatic components with many edges in random graphs
In an $r$-coloring of edges of the complete graph on $n$ vertices, how many edges are there in the largest monochromatic connected component? A construction of Gyárfás shows that for infinitely many values of $r$, there exist colorings where all monochromatic components have at most $\left(\frac{1}{r^2-r}+o(1)\right)\binom{n}{2}$ edges. Conlon, Luo, and Tyomkyn conjectured that components with at least this many edges are attainable for all $r \ge 3$. Conlon, Luo, and Tyomkyn proved this conj…

Detecting, intercepting, and ejecting adversaries is getting harder as cybercriminals move faster and smarter.
Check out my latest CSO piece on how CISOs need to rethink their digital playbooks in light of these trends.
Thanks to Matt Immler of Okta, Tom Etheridge of CrowdStrike and Pierre Cadieux of Cisco Talos for their insight.

CISOs must rethink defense playbooks as cybercriminals move faster, smarter
Facing faster, stealthier intruders, CISOs are under pressure to modernize their cybersecurity strategies, toolsets, and tactics. From detection to post-mortem, here are key points of renewed emphasis.