Tag
This paper derives lower bounds on how much a company can manipulate fairness metrics after an audit, showing that finite-budget fairness certification cannot fully eliminate post-audit manipulation.
The tweet highlights a fireside chat at AIE Worlds Fair where Simon Eskildsen emphasizes the value of napkin math for estimating theoretical system throughput limits and understanding why actual performance differs.
This paper presents a theoretical analysis of parameter-based knowledge editing in LLMs, revealing how localized edits can cause global reasoning collapse, and provides empirical evidence that retrieval-based methods outperform parameter-editing approaches.