ZeroSlop — October 20, 2026
Today: AM-Bench: A Unified Taxonomy and Evaluation Suite for…; Joint, Decoupled, or Agent: When Each Multimodal…; Wine Ported to OpenBSD
12 stories worth knowing about today — AI breakthroughs, launches, and innovations making a difference.
1. AM-Bench: A Unified Taxonomy and Evaluation Suite for Agentic Misalignment
arXiv CS.AI
arXiv:2610.24264v1 Announce Type: new Abstract: As LLM-based agents are deployed with increasing autonomy in real-world environments, a critical failure mode emerges: agentic misalignment, where agents choose harmful actions over failure. However, existing safety benchmarks primarily evaluate wheth…
2. Joint, Decoupled, or Agent: When Each Multimodal Alignment Paradigm Wins
arXiv CS.AI
arXiv:2610.24108v1 Announce Type: new Abstract: Jointly pretrained multimodal encoders are a common way to combine modalities, while alternatives integrate their information during downstream training or inference. When does each approach win? We compare three paradigms, instantiated with public ch…
3. Wine Ported to OpenBSD
Slashdot
Wine has been successfully ported to OpenBSD, with developers reporting that dozens of 64-bit Windows applications now run on the operating system. “Even the Remaster of The Witcher 3 which released just last month runs, and runs well,” writes Thomas Frohwein, OpenBSD developer and ports maintainer….
4. NVIDIA Advances Agentic Security for Media, From European Public Service Broadcasting to Global Content Production
NVIDIA Blog
At a creative industry event that took place today in Berlin — hosted in collaboration with Adobe, Bertelsmann, NVIDIA and UFA — the latest advancements in intelligent media were on display. Key to the news: the European Broadcasting Union (EBU) is collaborating with NVIDIA to expand its sovereign f…
5. Inception Releases Mercury 3 Diffusion LLM at 1,068 Tokens per Second
MarkTechPost
Inception released Mercury 3, a diffusion LLM it says runs at 1,068 tokens per second with a 1M-token context. On Inception’s own table it leads GPT-6 Luna and Claude Haiku 5.5 on 8 of 12 benchmarks. The post Inception Releases Mercury 3 Diffusion LLM at 1,068 Tokens per Second appeared first on Mar…
6. Meet Drex Max: Nace AI’s Multimodal Decision Model for Computer-Use Agents with a Score 67.59
MarkTechPost
Nace AI released Drex Max, a multimodal decision model that reads text and screenshots and returns calibrated probabilities. Nace reports 67.59 on Decision Index 0.3.2 at $0.06 per 1M input tokens. The post Meet Drex Max: Nace AI’s Multimodal Decision Model for Computer-Use Agents with a Score 67.59…
7. MiniMax Just Launched M3.1 Flash: 428B MoE Agentic Model at $0.15/M Input
MarkTechPost
MiniMax has released M3.1 Flash, a 428B-parameter MoE model with about 23B active parameters and a 1M-token context. API pricing is $0.15/$0.60 per 1M tokens, with weights due October 23 The post MiniMax Just Launched M3.1 Flash: 428B MoE Agentic Model at $0.15/M Input appeared first on MarkTechPost…
8. TorchTPU - A Hybrid Eager/Compiled Architecture for Native PyTorch Execution on TPU
arXiv CS.AI
arXiv:2610.24161v1 Announce Type: new Abstract: TorchTPU is an open-source collaboration between Google and Meta that introduces a native and principled PyTorch backend for Google TPUs. It enables seamless out-of-the-box model execution and TPU migration while delivering industry-competitive infere…
9. Training AIs to Solve CAPTCHAs
Schneier on Security
Interesting research : Abstract : Vision-language models (VLMs) can solve visual CAPTCHAs without task-specific training, but the agents built on them approach every challenge from scratch. For such an agent, the hundredth instance of a familiar puzzle costs as much time and compute as the first. Sp…
10. Hackers hit AI coding tool OpenCode, running up charges on users’ AI accounts
TechCrunch AI
OpenCode is an open source AI coding agent that’s become a popular alternative to Claude Code and Cursor. But hackers got into part of its online platform and stole some users’ API keys, and the AI tools it turned to for help kept refusing to look at the malicious code….
11. A Hacker Broke Into John Deere, SharkNinja With an Army of AI Agents
Slashdot
An anonymous reader quotes a report from The Verge: A hacker broke into more than 100 websites using an army of AI agents that mostly operated on their own, according to findings from the AI security firm Norforge. Though Norforge doesn’t name the targets, Reuters reports that the victims included c…
12. GardenBench: A lightweight, daily evaluation of LLM capabilities
arXiv CS.AI
arXiv:2610.23866v1 Announce Type: new Abstract: We introduce GardenBench, a benchmark with 11 test suites that evaluate the capabilities of Large Language Models (LLMs) across a range of tasks. By systematically prompting LLMs with various lightweight queries, we document day-to-day changes in mode…