Yanai Elazar surveys concept removal in activation space at Copenhagen NLP Symposium
yanaiela · x · 2026-10-22
Yanai Elazar (Bar-Ilan University) is speaking at the Copenhagen NLP Symposium (Oct 22, 2026) on "Lost and Found in Activation Space: Methods and Evaluations for Concept Removal."
The talk traces the arc of concept removal — locating and removing specific concepts from model activation spaces and evaluating such interventions — from 2018 (his first paper, still relevant today) to 2026. A useful systematic overview for anyone working on interpretability and safety interventions.
More from Research
- Forward-only pretraining matches backprop: val 3.704 in 2.5 min on 8×H100, no autograd — tensorqt · 2026-10-22
- Wikipedia editors gut mech interp article, citing preprints and blog posts as unusable — Sauers_ · 2026-10-22
- HKU's PrismaCoder: Open Multimodal Code Models Rival Proprietary Systems — hkunlp · 2026-10-22
- Microsoft researcher cracks 31-year-old KLS conjecture with AI agents exploring 100+ proof paths — rohanpaul_ai · 2026-10-22
- GDPval benchmark under fire: first seed artifact in public split looks off — vincent_koc · 2026-10-22
- GPT-6-Astra nears saturation on 12 robot benchmarks but burns 9M tokens per task — rohanpaul_ai · 2026-10-22