sbulaev's submissions

1.		Dense Contexts Are Hard: Lexical Density Limits LLM Context Windows (arxiv.org)
		2 points by sbulaev 22 hours ago \| past \| discuss
2.		OpenAI's Codex chained decade-old DoS attacks to crash web servers (theregister.com)
		1 point by sbulaev 1 day ago \| past \| discuss
3.		Same Weights, Different Robot: A Deployment Safety View of VLA Policies (arxiv.org)
		1 point by sbulaev 2 days ago \| past \| discuss
4.		Forge: Multi-Agent Graduated Exploitation and Detection Engineering (arxiv.org)
		1 point by sbulaev 2 days ago \| past \| discuss
5.		Spikes in LLMs Are Bias Vectors: Spike-Free Quantization (arxiv.org)
		4 points by sbulaev 3 days ago \| past \| discuss
6.		Forget Attention: Importance-Aware Attention Is All You Need (arxiv.org)
		3 points by sbulaev 3 days ago \| past \| discuss
7.		AI Investments Are 'Circular Bet' as ROI Disappoints, Bain Survey Shows (bloomberg.com)
		5 points by sbulaev 4 days ago \| past \| discuss
8.		LLMs Are Closer to Religion Than They Appear (theregister.com)
		84 points by sbulaev 4 days ago \| past \| 92 comments
9.		Too dangerous to release: is Mythos the start of the restricted-AI era? (nature.com)
		3 points by sbulaev 10 days ago \| past \| 1 comment
10.		Cox Media fined after bragging it spied on users through their phones (theverge.com)
		23 points by sbulaev 11 days ago \| past \| 1 comment
11.		Mec builds first High-NA EUV-fabricated quantum dot qubit (tomshardware.com)
		1 point by sbulaev 11 days ago \| past \| discuss
12.		Google has seriously leaned into AI enshittification lately (theregister.com)
		4 points by sbulaev 11 days ago \| past \| discuss
13.		Ghost CMS SQL injection flaw exploited in large-scale ClickFix campaign (bleepingcomputer.com)
		3 points by sbulaev 12 days ago \| past \| discuss
14.		Advanced Shader Delivery on RX 9070 XT: up to 95% faster load times (tomshardware.com)
		1 point by sbulaev 12 days ago \| past \| discuss
15.		To A.I. Executives, We're All Just 'Meat Computers' (nytimes.com)
		3 points by sbulaev 12 days ago \| past \| discuss
16.		Laravel Lang packages hijacked to deploy credential-stealing malware (bleepingcomputer.com)
		3 points by sbulaev 13 days ago \| past \| discuss
17.		Megalodon chums the waters in 5.5K+ GitHub repo poisonings (theregister.com)
		6 points by sbulaev 14 days ago \| past \| 2 comments
18.		Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems (arxiv.org)
		38 points by sbulaev 14 days ago \| past \| 4 comments
19.		A hacker group is poisoning open source code at an unprecedented scale (wired.com)
		7 points by sbulaev 14 days ago \| past
20.		Is Capability a Liability? More Capable Language Models Make Worse Forecasts (arxiv.org)
		1 point by sbulaev 14 days ago \| past
21.		GitHub faces a fight for its survival at Microsoft (theverge.com)
		62 points by sbulaev 15 days ago \| past \| 13 comments
22.		Imaging hidden objects with consumer Lidar via motion-induced sampling (nature.com)
		2 points by sbulaev 15 days ago \| past
23.		Auditing Apple's DifferentialPrivacy.framework: Bugs, Misconfig, Practical Risks (arxiv.org)
		2 points by sbulaev 15 days ago \| past
24.		Even Claude agrees: hole in its sandbox was real and dangerous (theregister.com)
		5 points by sbulaev 16 days ago \| past
25.		SpaceX IPO Filing Reveals Anthropic Is Paying $15B/Year to Access Data Centers (wired.com)
		7 points by sbulaev 16 days ago \| past
26.		Agent Security Is a Systems Problem (arxiv.org)
		5 points by sbulaev 16 days ago \| past
27.		Max-severity flaw in ChromaDB for AI apps allows server hijacking (bleepingcomputer.com)
		1 point by sbulaev 16 days ago \| past
28.		N display of stupid, secret CISA credentials found in public GitHub rep (arstechnica.com)
		2 points by sbulaev 17 days ago \| past \| 1 comment
29.		'It is incredible': How AI is transforming mathematics (nature.com)
		2 points by sbulaev 17 days ago \| past \| 1 comment
30.		Polynomial Quantum Attack on Module-LWE over Power-of-2 Cyclotomics (arxiv.org)
		1 point by sbulaev 17 days ago \| past
		More