We're sharing new research on how models hack public benchmarks. The latest models, including Opus ... - 惯性聚合

推荐订阅源

Hacker News - Newest: "LLM"

Security @ Cisco Blogs

Heimdal Security Blog

cs.CV updates on arXiv.org

www.infosecurity-magazine.com

News and Events Feed by Topic

Hacker News: Front Page

KPMG report finds enterprise disconnect between AI and its ROI | CIO

cs.AI updates on arXiv.org

Application and Cybersecurity Blog

News | PayPal Newsroom

Tor Project blog

Security Archives - TechRepublic

Secure Thoughts

CERT Recently Published Vulnerability Notes

Google Online Security Blog

大猫的无限游戏

The Blog of Author Tim Ferriss

人人都是产品经理

Recent Announcements

Google DeepMind News

Hacker News: Ask HN

Blog — PlanetScale

博客园 - 【当耐特】

Engineering at Meta

Stack Overflow Blog

Recorded Future

罗磊的独立博客

Exploit-DB.com RSS Feed

The Exploit Database - CXSecurity.com

DataBreaches.Net

Security Affairs

WordPress大学

Microsoft Security Blog

Vulnerabilities – Threatpost

The Hacker News

SegmentFault 最新的问题

Proofpoint News Feed

Cursor(@cursor_ai)

Cursor (@cursor_ai) on X Cursor (@cursor_ai) on X Grok 4.5 and Composer are two different model weight classes. Composer 2.5 will remain offered, and... Try it out in Cursor with double usage for the first week. https://t.co/YBKMn66zS1 We've partnered with SpaceXAI to train Grok 4.5. It’s our most powerful model yet and the first we'... Claude Fable 5 is available again in Cursor. It leads all models on CursorBench, but is the most ex... See how Claude Fable 5 compares across every model: https://t.co/61FZktI8K1 See our full model rankings: https://t.co/61FZktIGzz Claude Sonnet 5 is now available in Cursor. On CursorBench, it's a meaningful step up from Sonnet 4... Stay in the loop with Live Activities, and get notified when an agent finishes or needs your input. ... Download Cursor for iOS on the App Store: https://t.co/Vjz5iixcaZ Introducing Cursor for iOS. Build from anywhere by launching always-on cloud agents. Or remotely co... More on how we're constraining eval environments so that scores better reflect model intelligence: h... More on how Notion built it with the Cursor SDK: https://t.co/0oEaafruBf You can now delegate tasks to Cursor directly from Notion. It's built on the Cursor SDK, so every c... Plugins can now include prebuilt canvases. Use the Atlassian canvas to see a real-time view of all ... You can also now use team marketplaces with GitLab, Bitbucket, and Azure DevOps in addition to local... Cursor now shows you a leaderboard of the most popular plugins, skills, and MCPs across your team. ... https://t.co/OQf0LoflrT Three announcements from our keynote at Compile, including how we're training a new model with Space... Cursor Automations now support an emoji trigger in Slack. React to any message to kick off a run. We've also added GitHub triggers for issues, reviews, and workflow runs, plus computer use for cloud... Introducing /automate, a skill for agents to set up automations for you. Describe your task in plai... Use /in-cloud to start a subagent in its own cloud VM. This is especially useful for isolating long... Learn more: https://t.co/8nSkAl9MPY It's now faster and easier to set up cloud environments. Environments are saved as snapshots so fut... Learn more: https://t.co/tYTdAuDF5a Use /in-cloud to start a subagent in its own cloud VM. This is especially useful for isolating long... It’s now easier to move local agents to the cloud so they can keep working with your laptop closed. ... We're launching code storage and git hosting. Origin gives teams and agents a place to host, review... We're excited to join forces with @SpaceX to advance the frontier of useful AI. Expect significant i... Auto-review is now the default for all new users. A classifier subagent reviews actions in context ... More on how we built it: https://t.co/Mspjs8bGA1 Learn more: https://t.co/noYPZbgNJD Cursor’s code review agent is now over 3x faster, 22% cheaper, and finds 10% more bugs. You can als... Claude Fable 5 is now available in Cursor. It sets a new state of the art on CursorBench at 72.9%, ... See how Claude Fable 5 compares across every model: https://t.co/61FZktIGzz With Design Mode, you can now point, draw, or talk to update your UI. Give Cursor visual prompts to shrink the gap between what you see and what the agent understands. L... Learn more: https://t.co/VZBavcKlNW With canvases, Cursor can create apps like dashboards, reports, and internal tools. Now you can pub... Cursor can now show your agent's context usage as an interactive report in a canvas. The context ex... Design Mode is now available in canvases. Select and annotate UI elements directly in a canvas to g... A great cloud agent experience involves a lot more than moving a local agent to a server. We've lea... Read the full announcement: https://t.co/TlsdTOI4WJ We’re increasing usage limits for every Teams user. Inspired by the success of our Ultra plan, we'r... Agent actions that aren't on your allowlist or can't be sandboxed go to a classifier subagent. This... Auto-review mode is now available in Cursor. It allows agents to run tool calls with fewer approval... Claude Opus 4.8 is now available in Cursor. On CursorBench, it's able to work much more efficiently... As agents use more context, input tokens have become the majority of price-equivalent token costs. Power users account for a large share of AI activity, and the gap is widening.

We're sharing new research on how models hack public benchmarks. The latest models, including Opus ...

Cursor (@cur · 2026-06-26 · via Cursor(@cursor_ai)

此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。原文来自 — 版权归原作者所有。