LIVE PULSE
5.0 Anthropic CEO Amodei calls for slower AI development and shared safety rules11 src2.7 Agility Robotics unveils Digit 5 humanoid for warehouses and factories2 src2.5 Apple ships rebuilt Siri with Google Gemini, but not in the EU2 src2.2 Siri AI in macOS 27 Golden Gate: FAQ, Germany availability, privacy questions2 src1.8 Sam Altman says OpenAI will not go public in 2026, citing AI safety concerns5 src1.4 OpenAI contractors review real ChatGPT conversations to rate responses, report says2 src1.4 Anthropic data retention policy prompts firms to limit Claude use for sensitive work1 src1.4 VoiceCodeBench arXiv paper proposes benchmark for exact structured-token recovery in speech recognition1 src1.4 Arabic-Russian Parallel Corpus and LLM Benchmark for Scientific Text1 src1.4 Study Analyzes Self-Reported Limitations in NLP Research1 src5.0 Anthropic CEO Amodei calls for slower AI development and shared safety rules11 src2.7 Agility Robotics unveils Digit 5 humanoid for warehouses and factories2 src2.5 Apple ships rebuilt Siri with Google Gemini, but not in the EU2 src2.2 Siri AI in macOS 27 Golden Gate: FAQ, Germany availability, privacy questions2 src1.8 Sam Altman says OpenAI will not go public in 2026, citing AI safety concerns5 src1.4 OpenAI contractors review real ChatGPT conversations to rate responses, report says2 src1.4 Anthropic data retention policy prompts firms to limit Claude use for sensitive work1 src1.4 VoiceCodeBench arXiv paper proposes benchmark for exact structured-token recovery in speech recognition1 src1.4 Arabic-Russian Parallel Corpus and LLM Benchmark for Scientific Text1 src1.4 Study Analyzes Self-Reported Limitations in NLP Research1 src
HEATPULSEAI MAGAZINES
FLIP · FOLLOW · SAVE
papersSEP 12 04:00 UTC

Audit Finds Batch-Normalization Stats Skew Machine Unlearning Evaluations

A new audit examines 263 publicly released checkpoints that use batch normalization and finds that reported unlearning results shift depending on which version of those statistics is used. Because batch-norm statistics are not produced by gradient updates and are rarely documented in model releases, refitting them on retained data can change the numbers an evaluation relies on. The authors argue that this makes some unlearning verdicts unreliable, since apparent forgetting may reflect checkpoint bookkeeping rather than the removed data genuinely being gone.

batch normalizationmachine-unlearningmodel checkpointsmodel evaluationmodel release documentationreproducibility

COVERAGE · 2 REPORTS · LINKS GO TO THE ORIGINAL OUTLETS