Tag: R1

  • Simon Willison’s Weblog: AbsenceBench: Language Models Can’t Tell What’s Missing

    Source URL: https://simonwillison.net/2025/Jun/20/absencebench/#atom-everything Source: Simon Willison’s Weblog Title: AbsenceBench: Language Models Can’t Tell What’s Missing Feedly Summary: AbsenceBench: Language Models Can’t Tell What’s Missing Here’s another interesting result to file under the “jagged frontier" of LLMs, where their strengths and weaknesses are often unintuitive. Long context models have been getting increasingly good at passing "Needle…

  • The Register: DeepSeek installer or just malware in disguise? Click around and find out

    Source URL: https://www.theregister.com/2025/06/11/deepseek_installer_or_infostealing_malware/ Source: The Register Title: DeepSeek installer or just malware in disguise? Click around and find out Feedly Summary: ‘BrowserVenom’ is pure poison Suspected cybercriminals have created a fake installer for Chinese AI model DeepSeek-R1 and loaded it with previously unknown malware called “BrowserVenom".… AI Summary and Description: Yes Summary: The text discusses…

  • Slashdot: Apple Researchers Challenge AI Reasoning Claims With Controlled Puzzle Tests

    Source URL: https://apple.slashdot.org/story/25/06/09/1151210/apple-researchers-challenge-ai-reasoning-claims-with-controlled-puzzle-tests?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: Apple Researchers Challenge AI Reasoning Claims With Controlled Puzzle Tests Feedly Summary: AI Summary and Description: Yes Summary: Apple researchers have discovered that advanced reasoning AI models, including OpenAI’s o3-mini and Gemini, exhibit a performance collapse at higher complexity levels in puzzle-solving tasks. This finding challenges existing assumptions about…

  • Cloud Blog: Accelerate your gen AI: Deploy Llama4 & DeepSeek on AI Hypercomputer with new recipes

    Source URL: https://cloud.google.com/blog/products/ai-machine-learning/deploying-llama4-and-deepseek-on-ai-hypercomputer/ Source: Cloud Blog Title: Accelerate your gen AI: Deploy Llama4 & DeepSeek on AI Hypercomputer with new recipes Feedly Summary: The pace of innovation in open-source AI is breathtaking, with models like Meta’s Llama4 and DeepSeek AI’s DeepSeek. However, deploying and optimizing large, powerful models can be  complex and resource-intensive. Developers and…

  • Simon Willison’s Weblog: Shisa V2 405B: Japan’s Highest Performing LLM

    Source URL: https://simonwillison.net/2025/Jun/3/shisa-v2/ Source: Simon Willison’s Weblog Title: Shisa V2 405B: Japan’s Highest Performing LLM Feedly Summary: Shisa V2 405B: Japan’s Highest Performing LLM Leonard Lin and Adam Lensenmayer have been working on Shisa for a while. They describe their latest release as “Japan’s Highest Performing LLM". Shisa V2 405B is the highest-performing LLM ever…

  • Simon Willison’s Weblog: How often do LLMs snitch? Recreating Theo’s SnitchBench with LLM

    Source URL: https://simonwillison.net/2025/May/31/snitchbench-with-llm/#atom-everything Source: Simon Willison’s Weblog Title: How often do LLMs snitch? Recreating Theo’s SnitchBench with LLM Feedly Summary: A fun new benchmark just dropped! Inspired by the Claude 4 system card – which showed that Claude 4 might just rat you out to the authorities if you told it to “take initiative" in…

  • Simon Willison’s Weblog: deepseek-ai/DeepSeek-R1-0528

    Source URL: https://simonwillison.net/2025/May/31/deepseek-aideepseek-r1-0528/ Source: Simon Willison’s Weblog Title: deepseek-ai/DeepSeek-R1-0528 Feedly Summary: deepseek-ai/DeepSeek-R1-0528 Sadly the trend for terrible naming of models has infested the Chinese AI labs as well. DeepSeek-R1-0528 is a brand new and much improved open weights reasoning model from DeepSeek, a major step up from the DeepSeek R1 they released back in January.…