Large-language models (LLMs) can support a wide range of applications like conversational agents, creative writing or general query answering. However, they are ill-suited for query answering in high-stake domains like medicine because they are typically not robust - even the same query can result in different answers when prompted multiple times. In order to improve the robustness of LLM queries, we propose using ranking queries repeatedly and to aggregate the queries using methods from social choice theory. We study ranking queries in diagnostic settings like medical and fault diagnosis and discuss how the Partial Borda Choice function from the literature can be applied to merge multiple query results. We discuss some additional interesting properties in our setting and evaluate the robustness of our approach empirically.

Robustness to distribution shift has become a growing concern for text and image models as they transition from research subjects to deployment in the real world. However, high-quality benchmarks for distribution shift in tabular machine learning tasks are still lacking despite the widespread real-world use of tabular data and differences in the models used for tabular data in comparison to text and images. As a consequence, the robustness of tabular models to distribution shift is poorly understood. To address this issue, we introduce TableShift, a distribution shift benchmark for tabular data. TableShift contains 15 binary classification tasks in total, each with an associated shift, and includes a diverse set of data sources, prediction targets, and distribution shifts. The benchmark covers domains including finance, education, public policy, healthcare, and civic participation, and is accessible using only a few lines of Python code via the TableShift API. We conduct a large-scale study comparing several state-of-the-art tabular data models alongside robust learning and domain generalization methods on the benchmark tasks. Our study demonstrates (1) a linear trend between in-distribution (ID) and out-of-distribution (OOD) accuracy; (2) domain robustness methods can reduce shift gaps but at the cost of reduced ID accuracy; (3) a strong relationship between shift gap (difference between ID and OOD performance) and shifts in the label distribution. The benchmark data, Python package, model implementations, and more information about TableShift are available at https://github.com/mlfoundations/tableshift and https://tableshift.org .

Stance detection is important for understanding different attitudes and beliefs on the Internet. However, given that a passage's stance toward a given topic is often highly dependent on that topic, building a stance detection model that generalizes to unseen topics is difficult. In this work, we propose using contrastive learning as well as an unlabeled dataset of news articles that cover a variety of different topics to train topic-agnostic/TAG and topic-aware/TAW embeddings for use in downstream stance detection. Combining these embeddings in our full TATA model, we achieve state-of-the-art performance across several public stance detection datasets (0.771 $F_1$-score on the Zero-shot VAST dataset). We release our code and data at https://github.com/hanshanley/tata.

Approximation rates are analyzed for deep surrogates of maps between infinite-dimensional function spaces, arising e.g. as data-to-solution maps of linear and nonlinear partial differential equations. Specifically, we study approximation rates for Deep Neural Operator and Generalized Polynomial Chaos (gpc) Operator surrogates for nonlinear, holomorphic maps between infinite-dimensional, separable Hilbert spaces. Operator in- and outputs from function spaces are assumed to be parametrized by stable, affine representation systems. Admissible representation systems comprise orthonormal bases, Riesz bases or suitable tight frames of the spaces under consideration. Algebraic expression rate bounds are established for both, deep neural and spectral operator surrogates acting in scales of separable Hilbert spaces containing domain and range of the map to be expressed, with finite Sobolev or Besov regularity. We illustrate the abstract concepts by expression rate bounds for the coefficient-to-solution map for a linear elliptic PDE on the torus.

US Central Command (Centcom) confirmed that the attack had been carried out at 21:30 local time (18:30GMT), and had killed the "commander responsible for directly planning and participating in attacks on US forces in the region".

A senior commander from Kataib Hezbollah, an Iran-backed armed group in Iraq that the Pentagon linked to an attack that killed three US troops, died in a drone strike on a vehicle in eastern Baghdad, according to security sources and media reports. One of the sources said three people were killed and that the vehicle targeted on Wednesday night was used by Iraq's Popular Mobilisation Forces (PMF), a state security agency composed of dozens of armed groups, many of them close to Iran. Two officials with Iran-backed armed groups in Iraq said that senior commander Abu Baqir al-Saadi was among those killed, the Associated Press news agency reported. Local outlet Sabereen News also reported al-Saadi had been killed in the blast. Al Jazeera's Ali Hashem, reporting from Baghdad, said that "several explosions" were heard across the Iraqi capital and that security sources said three people have been killed.

Fox News chief national security correspondent Jennifer Griffin has the latest on the strike on'The Story.' The U.S. carried out a drone strike in Baghdad late Wednesday that killed three members of the powerful Kataib Hezbollah militia – including a high-ranking commander connected with a drone strike that killed three U.S. troops in Jordan late last month. U.S. Central Command (CENTCOM) said forces conducted a unilateral strike in Iraq around 9:30 p.m. in response to a drone strike that killed three U.S. troops in Jordan on Jan. 28. The strike, which occurred on a main thoroughfare in Baghdad's Mashtal neighborhood, was considered a "high-value individual target," Fox News is told. People inspect the vehicle targeted by airstrike in Baghdad, Iraq on February 07, 2024.

Fox News Flash top sports headlines are here. Check out what's clicking on Foxnews.com. A Pennsylvania man could face up to four years in prison after he was charged in a federal criminal complaint for illegally flying a drone over the Baltimore Ravens stadium during the AFC Championship last month, causing an unusual delay of game. The U.S. Attorney's Office for the District of Maryland announced the charges on Monday, alleging that Matthew Hebert, 44, violated a temporary flight restriction placed on M&T Bank Stadium when he flew a drone over the area during the NFL game. Zay Flowers of the Ravens makes a catch for touchdown during the AFC Championship game against the Kansas City Chiefs at M&T Bank Stadium on Jan. 28, 2024, in Baltimore.

Houthi militants in Yemen are using Iranian-supplied missiles and drones to attack civilian and military targets across the Middle East, analysis from the Defense Intelligence Agency (DIA) shows. The report, "Iran: Enabling Houthi Attacks Across the Middle East," aims to provide more insight into the relationship between Iran and the Houthis. The militant group, stationed in Yemen, has for months been striking commercial vessels traveling through the Red Sea in protest of Palestinian civilians killed during Israel's ongoing offensive against Hamas members in Gaza. Houthi fighters stage a rally in support of the Palestinians in the Gaza Strip and against the U.S.-led airstrikes on Yemen, in Sanaa, Yemen, Monday, Jan. 29, 2024. Most recently, Houthi rebels fired ballistic missiles at two ships traveling through Middle East waters.

A group of scientists from across the U.S. claim to have created the first artificial intelligence capable of generating AI without human supervision. The Department of Homeland Security is recruiting dozens of artificial intelligence experts for an "AI Corps" that will use the blossoming tech to advance national security goals, Secretary Alejandro Mayorkas announced Tuesday. The 50 experts will be part of a DHS initiative to leverage AI for a variety of efforts, including combating fentanyl distribution, online child exploitation and cyberattacks, according to Mayorkas. He announced the AI Corps alongside DHS Chief Information Officer Eric Hysen at a Mountain View, California, event as the House tried and failed to impeach the secretary. Homeland Security Secretary Alejandro Mayorkas on Tuesday launched a hiring spree for 50 artificial intelligence experts as the House pursued a doomed impeachment case against him.