We introduce CLEVER, the first curated benchmark for evaluating the generation of specifications and formally verified code in Lean. The benchmark comprises of 161 programming problems; it evaluates …

22 Ean 2025 · Promoting openness in scientific communication and the peer-review process

1 Beal 2025 · One common approach is training models to refuse unsafe queries, but this strategy can be vulnerable to clever prompts, often referred to as jailbreak attacks, which can trick the AI into …

16 MFómh 2025 · A fundamental limitation of current AI agents is their inability to learn complex skills on the fly at test time, often behaving like “clever but clueless interns” in novel environments. This …

ACII 2021 CaSiNo: A Corpus of Campsite Negotiation Dialogues for Automatic Negotiation Systems Kushal Chawla, Jaysa Ramirez, Rene Clever, Gale M. Lucas, Jonathan May, Jonathan Gratch 2021 …

Promoting openness in scientific communication and the peer-review process

15 Feabh 2018 · Our analysis yields a novel robustness metric called CLEVER, which is short for Cross Lipschitz Extreme Value for nEtwork Robustness. The proposed CLEVER score is attack-agnostic and …

12 DFómh 2024 · Keywords: histopathology, foundation models, batch effects, Clever Hans effect, robustness, generalization Abstract: Deep learning has led to remarkable advancements in …

26 MFómh 2024 · TL;DR: This paper proposes CLeVER, a novel equivariant-based contrastive learning framework that improves training efficiency and robustness in downstream tasks by incorporating …

579 In this paper, we have proposed a novel counter- factual framework CLEVER for debiasing fact- checking models. Unlike existing works, CLEVER is augmentation-free and mitigates biases on infer- …