The use of large language models for code generation is a rapidly growing trend in software development. However, without effective methods for ensuring the correctness of generated code, this trend could lead to any number of undesirable outcomes. In this paper, we lay out a vision for addressing this challenge: the Clover paradigm, short for Closed-Loop Verifiable Code Generation, which reduces correctness checking to the more accessible problem of consistency checking. At the core of Clover lies a checker that performs consistency checks among code, docstrings, and formal annotations. The checker is implemented using a novel integration of formal verification tools and large language models. We provide a theoretical analysis to support our thesis that Clover should be effective at consistency checking. We also empirically investigate its feasibility on a hand-designed dataset (CloverBench) featuring annotated Dafny programs at a textbook level of difficulty. Experimental results show that for this dataset, (i) LLMs are reasonably successful at automatically generating formal specifications; and (ii) our consistency checker achieves a promising acceptance rate (up to 87%) for correct instances while maintaining zero tolerance for incorrect ones (no false positives).

In-context learning (ICL) i.e. showing LLMs only a few task-specific demonstrations has led to downstream gains with no task-specific fine-tuning required. However, LLMs are sensitive to the choice of prompts, and therefore a crucial research question is how to select good demonstrations for ICL. One effective strategy is leveraging semantic similarity between the ICL demonstrations and test inputs by using a text retriever, which however is sub-optimal as that does not consider the LLM's existing knowledge about that task. From prior work (Lyu et al., 2023), we already know that labels paired with the demonstrations bias the model predictions. This leads us to our hypothesis whether considering LLM's existing knowledge about the task, especially with respect to the output label space can help in a better demonstration selection strategy. Through extensive experimentation on three text classification tasks, we find that it is beneficial to not only choose semantically similar ICL demonstrations but also to choose those demonstrations that help resolve the inherent label ambiguity surrounding the test example. Interestingly, we find that including demonstrations that the LLM previously mis-classified and also fall on the test example's decision boundary, brings the most performance gain.

Graph Neural Networks (GNNs) have emerged as promising solutions for collaborative filtering (CF) through the modeling of user-item interaction graphs. The nucleus of existing GNN-based recommender systems involves recursive message passing along user-item interaction edges to refine encoded embeddings. Despite their demonstrated effectiveness, current GNN-based methods encounter challenges of limited receptive fields and the presence of noisy ``interest-irrelevant'' connections. In contrast, Transformer-based methods excel in aggregating information adaptively and globally. Nevertheless, their application to large-scale interaction graphs is hindered by inherent complexities and challenges in capturing intricate, entangled structural information. In this paper, we propose TransGNN, a novel model that integrates Transformer and GNN layers in an alternating fashion to mutually enhance their capabilities. Specifically, TransGNN leverages Transformer layers to broaden the receptive field and disentangle information aggregation from edges, which aggregates information from more relevant nodes, thereby enhancing the message passing of GNNs. Additionally, to capture graph structure information effectively, positional encoding is meticulously designed and integrated into GNN layers to encode such structural knowledge into node attributes, thus enhancing the Transformer's performance on graphs. Efficiency considerations are also alleviated by proposing the sampling of the most relevant nodes for the Transformer, along with two efficient sample update strategies to reduce complexity. Furthermore, theoretical analysis demonstrates that TransGNN offers increased expressiveness compared to GNNs, with only a marginal increase in linear complexity. Extensive experiments on five public datasets validate the effectiveness and efficiency of TransGNN.

The rise of Generative Artificial Intelligence systems ("AI systems") has created unprecedented social engagement. AI code generation systems provide responses (output) to questions or requests by accessing the vast library of open-source code created by developers over the past few decades. However, they do so by allegedly stealing the open-source code stored in virtual libraries, known as repositories. This Article focuses on how this happens and whether there is a solution that protects innovation and avoids years of litigation. We also touch upon the array of issues raised by the relationship between AI and copyright. Looking ahead, we propose the following: (a) immediate changes to the licenses for open-source code created by developers that will limit access and/or use of any open-source code to humans only; (b) we suggest revisions to the Massachusetts Institute of Technology ("MIT") license so that AI systems are required to procure appropriate licenses from open-source code developers, which we believe will harmonize standards and build social consensus for the benefit of all of humanity, rather than promote profit-driven centers of innovation; (c) we call for urgent legislative action to protect the future of AI systems while also promoting innovation; and (d) we propose a shift in the burden of proof to AI systems in obfuscation cases.

Even before the drone strike that killed three U.S. service members in Jordan on Sunday, the Biden administration was planning for a moment just like this, debating how it might strike back in ways that would deter Iran's proxy forces and send a message that Tehran would not miss. But the options range from the unsatisfying to the highly risky. Mr. Biden could order strikes on the proxy forces, a major escalation of the whack-a-mole attacks it has conducted in recent weeks in Syria, Iraq and Yemen. So far, those attacks have put a dent into the abilities of the Iranian-backed groups that have mounted more than 160 attacks. But they have failed, as Mr. Biden himself noted 10 days ago, to deter those groups.

There have been at least 160 attacks on U.S. troops in the Middle East since mid-October, following this weekend's attack on a base in Jordan near the Syrian border that left three American soldiers dead and dozens of others injured, U.S. officials said. Defense Secretary Lloyd Austin addressed Sunday's attack and vowed the U.S. would "take all necessary actions" to keep U.S. troops in the region safe. "Let me start with my outrage and sorrow for the deaths of three brave U.S. troops in Jordan and for the other troops who were wounded," Austin said. He added, "The president and I will not tolerate attack on U.S. forces. And we will take all necessary actions to defend the U.S. and our troops."

Former U.S. Navy SEAL Jonathan Gilliam joined'Fox & Friends First' to discuss what he sees as the'common denominator' to the Biden administration's response to attacks in the Middle East and how it will serve as a'litmus test' in 2024. A U.S. official confirmed to Fox News the drone from an Iranian proxy that killed 3 American service members in Jordan and injured others got past the air defenses for Tower 22 because it was mistaken for a U.S. drone expected to return to the base at the same time. The Wall Street Journal initially reported on this development on Monday. A U.S. official confirmed the information to Fox News. President Biden has vowed to take action against Iranian-backed militants in the Middle East after the drone attack at Tower 22, a post in Jordan near Syria's border, over the weekend.

The United States military announced on Sunday that three US soldiers were killed and at least 34 were wounded in a drone attack targeting Tower 22, a remote logistics outpost near the Jordan-Syrian border. The attack has elicited a strong reaction from Washington with President Joe Biden pledging to hold the attackers to account. The Islamic Resistance in Iraq, an umbrella group of Iran-backed armed groups in the region, claimed the attacks, saying it was in response to US support to Israel's war on Gaza, which has killed more than 26,000 people. Tower 22, which houses a small US logistics outpost, is located in Jordan's northeast close to the borders with Iraq and Syria. Public information about the outpost is limited.

Iran is claiming that accusations of its involvement in an attack that left three U.S. service members dead in Jordan over the weekend are "baseless." Iranian foreign ministry spokesperson Nasser Kanaani also claimed that "resistance groups" in the region do not take orders from Iran, according to Reuters. The remarks come after three American service members were killed and other were wounded in a drone attack near the Syrian border over the weekend. "While we are still gathering the facts of this attack, we know it was carried out by radical Iran-backed militant groups operating in Syria and Iraq," President Biden said in response to the strike. Some Republicans have pressed Biden to authorize stronger action against Iran, with members of both parties concerned about the safety of U.S. troops overseas.

This was the day that President Biden and his team had feared for more than three months, the day that relatively low-level attacks by Iranian proxy groups on American troops in the Middle East turned deadly and intensified the pressure on the president to respond in kind. With three American service members killed and two dozen more injured by a drone in Jordan, Mr. Biden must decide how far he is willing to go in terms of retaliation at the risk of a wider war that he has sought to avoid ever since the Oct. 7 terrorist attack by Hamas touched off the current Middle East crisis. Until now, the president had carefully calibrated his responses to the more than 150 attacks by Iranian-backed militias on American forces in the region since Oct. 7. He essentially ignored the majority that were successfully intercepted or did little to no damage while authorizing limited U.S. strikes focused mainly on buildings, weapons and infrastructure after attacks that were more brazen, most notably against the Houthis in Yemen who have targeted shipping in the Red Sea. The first deaths of American troops under fire, however, will require a different level of response, American officials said, and the president's advisers were in consensus about that as they consulted with him by secure videoconference on Sunday.