Experimental News Clipping Site

Tag: llm

  • 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…

  • Schneier on Security: Subverting AIOps Systems Through Poisoned Input Data

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.schneier.com/blog/archives/2025/08/subverting-aiops-systems-through-poisoned-input-data.html Source: Schneier on Security Title: Subverting AIOps Systems Through Poisoned Input Data Feedly Summary: In this input integrity attack against an AI system, researchers were able to fool AIOps tools: AIOps refers to the use of LLM-based agents to gather and analyze application telemetry, including system logs, performance metrics, traces, and alerts,…

  • 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.……

  • Tomasz Tunguz: When One AI Grades Another’s Work

    —

    by

    Kurt Seifried
    in

    Source URL: https://www.tomtunguz.com/evolution-of-ai-judges-improving-evoblog/ Source: Tomasz Tunguz Title: When One AI Grades Another’s Work Feedly Summary: Since launching EvoBlog internally, I’ve wanted to improve it. One way of doing this is having an LLM judge the best posts rather than a static scoring system. I appointed Gemini 2.5 to be that judge. This post is a…

  • Simon Willison’s Weblog: llama.cpp guide: running gpt-oss with llama.cpp

    —

    by

    Kurt Seifried
    in

    Source URL: https://simonwillison.net/2025/Aug/19/gpt-oss-with-llama-cpp/ Source: Simon Willison’s Weblog Title: llama.cpp guide: running gpt-oss with llama.cpp Feedly Summary: llama.cpp guide: running gpt-oss with llama.cpp Really useful official guide to running the OpenAI gpt-oss models using llama-server from llama.cpp – which provides an OpenAI-compatible localhost API and a neat web interface for interacting with the models. TLDR version…

  • Simon Willison’s Weblog: Google Gemini URL Context

    —

    by

    Kurt Seifried
    in

    Source URL: https://simonwillison.net/2025/Aug/18/google-gemini-url-context/ Source: Simon Willison’s Weblog Title: Google Gemini URL Context Feedly Summary: Google Gemini URL Context New feature in the Gemini API: you can now enable a url_context tool which the models can use to request the contents of URLs as part of replying to a prompt. I released llm-gemini 0.25 with a…

  • Shabie’s blog: Agents are search over action space

    —

    by

    Kurt Seifried
    in

    Source URL: https://shabie.github.io/2025/08/18/agents-are-search-over-action-space.html Source: Shabie’s blog Title: Agents are search over action space Feedly Summary: It’s no secret that today’s LLM-based agents are unreliable. This makes them a gamble for most critical tasks, so where can they be safely applied? The answer lies in finding asymmetry: we should use them in domains where the downside…

  • Slashdot: LLM Found Transmitting Behavioral Traits to ‘Student’ LLM Via Hidden Signals in Data

    —

    by

    Kurt Seifried
    in

    Source URL: https://slashdot.org/story/25/08/17/0331217/llm-found-transmitting-behavioral-traits-to-student-llm-via-hidden-signals-in-data?utm_source=rss1.0mainlinkanon&utm_medium=feed Source: Slashdot Title: LLM Found Transmitting Behavioral Traits to ‘Student’ LLM Via Hidden Signals in Data Feedly Summary: AI Summary and Description: Yes Summary: The study highlights a concerning phenomenon in AI development known as subliminal learning, where a “teacher” model instills traits in a “student” model without explicit instruction. This can…

  • 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…