Hallucination detection is a critical step toward understanding the trustworthiness of modern language models (LMs). To achieve this goal, we re-examine existing detection approaches based on the self-consistency of LMs and uncover two types of hallucinations resulting from 1) question-level and 2) model-level, which cannot be effectively identified through self-consistency check alone. Building upon this discovery, we propose a novel sampling-based method, i.e., semantic-aware cross-check consistency (SAC$^3$) that expands on the principle of self-consistency checking. Our SAC$^3$ approach incorporates additional mechanisms to detect both question-level and model-level hallucinations by leveraging advances including semantically equivalent question perturbation and cross-model response consistency checking. Through extensive and systematic empirical analysis, we demonstrate that SAC$^3$ outperforms the state of the art in detecting both non-factual and factual statements across multiple question-answering and open-domain generation benchmarks.

Two US senators requested a federal investigation of Tesla's statements about its Autopilot driver assistance system Wednesday, asserting in the wake of multiple crashes that the automaker's exaggerations have put consumers at risk. The letter to the Federal Trade Commission concerning "Tesla's misleading advertising" comes two days after another federal agency launched a probe into Autopilot. Tesla and Chief Executive Elon Musk's "repeated overstatements of the vehicles' capabilities -- despite clear and frequent warnings -- demonstrate a deeply concerning disregard for the safety of those on the road and require real accountability," Democratic senators Richard Blumenthal and Ed Markey wrote. "We urge you to swiftly open an investigation into Tesla's repeated and overstated claims about their Autopilot and Full Self-Driving features and take appropriate enforcement action to prevent further injury or death as a result of any Tesla feature," they said in the letter to FTC Chair Lina Khan. The Autopilot system assists with steering and automatic braking and can be employed to help drivers navigate past slow cars, according to Tesla's website, which says the system's features "require active driver supervision and do not make the vehicle autonomous."

Republican and Democratic senators introduced legislation on Thursday that would block international sales of United States-made drones to countries that are not close US allies, mentioning Saudi Arabia in particular. Reuters broke the news in June that President Donald Trump's administration planned to reinterpret the Missile Technology Control Regime, a Cold War arms agreement between 35 nations, with the goal of allowing US defence contractors to sell more drones to an array of nations. Republican Senators Mike Lee and Rand Paul, Democratic Senators Chris Murphy and Chris Coons, and Senator Bernie Sanders, an independent who caucuses with Democrats, introduced the measure. It would amend the Arms Export Control Act to prohibit the export, transfer or trade of many advanced drones except to countries that are NATO members and to Australia, New Zealand, South Korea, Japan and Israel, they said in a news release. US lawmakers have tried before to rein in Trump administration plans for arms sales, particularly to Saudi Arabia and the United Arab Emirates for use in the war in Yemen.