Understanding why real-world events occur is important for both natural language processing and practical decision-making, yet direct-cause inference remains underexplored in evidence-rich settings. To address this gap, we organized SemEval-2026 Task 12: Abductive Event Reasoning (AER). The task asks systems to identify the most plausible direct cause of a target event from supporting evidence. We formulate AER as an evidence-grounded multiple choice benchmark that captures key challenges of real-world causal reasoning, including distributed evidence, indirect background factors, and semantically related but non-causal distractors. The shared task attracted 122 participants and received 518 submissions. This paper presents the task formulation, dataset construction pipeline, evaluation setup, and system results. AER provides a focused benchmark for abductive reasoning over real-world events and highlights challenges for future work on causal reasoning and multi-document understanding.

Language models (LMs) are increasingly used to generate decisions and explanations for high-stakes financial tasks, but these explanations can contain factual inconsistencies with the underlying claim or invalid causal reasoning. We propose a two-step auditing framework that decomposes each explanation into reasoning points and verifies them using natural language inference (NLI): (i) claim ⇒ point for factuality verification and (ii) point ⇒ conclusion for causality verification. To reduce reliance on expensive expert annotations, we study effective fine-tuning strategies based on semi-supervised learning (SSL) for NLI verifiers, leveraging abundant unlabeled explanations via pseudo-labeling and FreeMatch with self-adaptive confidence thresholds. Experiments across small language models and both encoder-only and decoder-only verifiers under varying labeling budgets show that SSL consistently outperforms the pre-trained baseline and supervised fine-tuning. FreeMatch is generally more robust than simple pseudo-labeling, suggesting better utilization of unlabeled data for scaling error detection of LMs’ generations.

Adversarial texts help explore vulnerabilities in language models, improve model robustness, and explain their working mechanisms. However, existing word-level attack methods trap in a one-to-one attack pattern, i.e., only a single word can be modified in one transformation round, and they ignore the interactions between several consecutive words. In this paper, we propose ValCAT, a black-box attack framework that misleads the language model by applying variable-length contextualized transformations to the original text. Compared to word-level methods, ValCAT expands the basic units of perturbation from single words to spans composed of multiple consecutive words, enhancing the perturbation capability. Experiments show that our method outperforms state-of-the-art methods in terms of attack success rate, perplexity, and semantic similarity on several classification tasks and inference tasks. The comprehensive human evaluation demonstrates that ValCAT has a significant advantage in ensuring the fluency of the adversarial examples and achieves better semantic consistency. We release the code at https://github.com/linerxliner/ValCAT.