LIVE PULSE
3.9 Anthropic CEO Amodei calls for slower AI development and shared safety rules — 11 src2.1 Agility Robotics unveils Digit 5 humanoid for warehouses and factories — 2 src2.0 Apple ships rebuilt Siri with Google Gemini, but not in the EU — 2 src1.7 Siri AI in macOS 27 Golden Gate: FAQ, Germany availability, privacy questions — 2 src1.4 Sam Altman says OpenAI will not go public in 2026, citing AI safety concerns — 5 src1.1 OpenAI contractors review real ChatGPT conversations to rate responses, report says — 2 src1.1 Anthropic data retention policy prompts firms to limit Claude use for sensitive work — 1 src1.1 Study examines issue bias in LLMs used as writing assistants before Swedish 2026 election — 1 src1.1 Study Audits Misalignment in Multi-Modal World Models — 1 src1.1 Retrieval-Grounded Reasoning Approach Proposed for Universal Multimodal Embeddings — 1 src3.9 Anthropic CEO Amodei calls for slower AI development and shared safety rules — 11 src2.1 Agility Robotics unveils Digit 5 humanoid for warehouses and factories — 2 src2.0 Apple ships rebuilt Siri with Google Gemini, but not in the EU — 2 src1.7 Siri AI in macOS 27 Golden Gate: FAQ, Germany availability, privacy questions — 2 src1.4 Sam Altman says OpenAI will not go public in 2026, citing AI safety concerns — 5 src1.1 OpenAI contractors review real ChatGPT conversations to rate responses, report says — 2 src1.1 Anthropic data retention policy prompts firms to limit Claude use for sensitive work — 1 src1.1 Study examines issue bias in LLMs used as writing assistants before Swedish 2026 election — 1 src1.1 Study Audits Misalignment in Multi-Modal World Models — 1 src1.1 Retrieval-Grounded Reasoning Approach Proposed for Universal Multimodal Embeddings — 1 src
HEATPULSEAI MAGAZINES
FLIP · FOLLOW · SAVE
Log inSign up

persona-vectors

topic3 events
papersTODAY 04:00 UTC

arXiv paper compares off-the-shelf persona vectors with targeted steering for sycophancy

A new arXiv preprint examines sycophancy, the tendency of language models to agree with users even when the user is wrong. The authors build on earlier work that derived sycophancy persona vectors and used activation steering to control the behaviour, and they test whether generic, off-the-shelf persona vectors can match purpose-built steering methods. The results suggest the simpler approach performs competitively.

sycophancy/activation-steering/interpretability/alignment
papersSEP 10 04:00 UTC

Study Examines How Preventative Steering Defenses Hold Up During Adversarial Fine-Tuning

New arXiv research explores how language models resist harmful behavior shifts caused by malicious fine-tuning. The work evaluates preventative steering, a training-time method that injects undesirable persona vectors during fine-tuning and removes them at inference, and analyzes how the defense's effectiveness changes over the course of training. The findings suggest these safeguards require active adjustment across training phases rather than a fixed configuration.

safety/llm-security/fine-tuning/research