Experimental News Clipping Site

Tag: large language models

  • Unit 42: Logit-Gap Steering: A New Frontier in Understanding and Probing LLM Safety

    —

    by

    Kurt Seifried
    in

    Source URL: https://unit42.paloaltonetworks.com/logit-gap-steering-impact/ Source: Unit 42 Title: Logit-Gap Steering: A New Frontier in Understanding and Probing LLM Safety Feedly Summary: New research from Unit 42 on logit-gap steering reveals how internal alignment measures can be bypassed, making external AI security vital. The post Logit-Gap Steering: A New Frontier in Understanding and Probing LLM Safety appeared…

  • Wired: Do Large Language Models Dream of AI Agents?

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.wired.com/story/sleeptime-compute-chatbots-memory/ Source: Wired Title: Do Large Language Models Dream of AI Agents? Feedly Summary: For AI models, knowing what to remember might be as important as knowing what to forget. Welcome to the era of “sleeptime compute.” AI Summary and Description: Yes Summary: The text introduces the concept of “sleeptime compute,” which emphasizes…

  • The Register: Open the pod bay door, GPT-4o

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.theregister.com/2025/08/20/gpt4o_pod_bay_door/ Source: The Register Title: Open the pod bay door, GPT-4o Feedly Summary: Researchers use LLM in ‘AI Space Cortex’ to automate robotic extraterrestrial exploration Businesses may be struggling to find meaningful ways to use artificial intelligence software, but space scientists at least have a few ideas about how to deploy AI models.……

  • Cloud Blog: An efficient path to production AI: Kakao’s journey with JAX and Cloud TPUs

    —

    by

    Kurt Seifried
    in

    Source URL: https://cloud.google.com/blog/products/infrastructure-modernization/kakaos-journey-with-jax-and-cloud-tpus/ Source: Cloud Blog Title: An efficient path to production AI: Kakao’s journey with JAX and Cloud TPUs Feedly Summary: When your messaging platform serves 49 million people – 93% of South Korea’s population – every technical decision carries enormous weight. The engineering team at Kakao faced exactly this challenge when their existing…

  • Simon Willison’s Weblog: TIL: Running a gpt-oss eval suite against LM Studio on a Mac

    —

    by

    Kurt Seifried
    in

    Source URL: https://simonwillison.net/2025/Aug/17/gpt-oss-eval-suite/#atom-everything Source: Simon Willison’s Weblog Title: TIL: Running a gpt-oss eval suite against LM Studio on a Mac Feedly Summary: TIL: Running a gpt-oss eval suite against LM Studio on a Mac The other day I learned that OpenAI published a set of evals as part of their gpt-oss model release, described in…

  • The Register: Little LLM on the RAM: Google’s Gemma 270M hits the scene

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.theregister.com/2025/08/15/little_llm_on_the_ram/ Source: The Register Title: Little LLM on the RAM: Google’s Gemma 270M hits the scene Feedly Summary: A tiny model trained on trillions of tokens, ready for specialized tasks Google has unveiled a pint-sized new addition to its “open" large language model lineup: Gemma 3 270M.… AI Summary and Description: Yes Summary:…

  • The Register: LLM chatbots trivial to weaponise for data theft, say boffins

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.theregister.com/2025/08/15/llm_chatbots_trivial_to_weaponise/ Source: The Register Title: LLM chatbots trivial to weaponise for data theft, say boffins Feedly Summary: System prompt engineering turns benign AI assistants into ‘investigator’ and ‘detective’ roles that bypass privacy guardrails A team of boffins is warning that AI chatbots built on large language models (LLM) can be tuned into malicious…

  • The Register: Dodgy Huawei chips nearly sunk DeepSeek’s next-gen R2 model

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.theregister.com/2025/08/14/dodgy_huawei_deepseek/ Source: The Register Title: Dodgy Huawei chips nearly sunk DeepSeek’s next-gen R2 model Feedly Summary: Chinese AI model dev still plans to use homegrown silicon for inferencing Unhelpful Huawei AI chips are reportedly why Chinese model dev DeepSeek’s next-gen LLMs are taking so long.… AI Summary and Description: Yes Summary: The text…