Abstract: Systems and methods for pre-training and fine-tuning of neural-network-based language models to reason directly over tables without generating logical forms. In some examples, a language model can be pre-trained using masked-language modeling tasks synthetically generated from tables pulled from a knowledge corpus. In some examples, the language model may be further pre-trained using pairs of counterfactual statements generated from those tables, and/or one or more statements that compare selected data from those tables. The language model may then be fine-tuned using examples that include only a question, an answer, and a table, allowing fine-tuning examples to be harvested directly from existing benchmark datasets or synthetically generated.

Abstract: A method for closing a fillable collecting tank, in particular a fillable collecting tank of a heat exchanger for storing a fluid, having walls forming the collecting tank, wherein one of the walls is formed as a baseplate having openings for receiving pipes, wherein a filling opening for adding the fluid is provided in one of the walls, wherein the filling opening can be closed by the provision of a closure element that can be inserted into the filling opening or can be placed onto the filling opening after the fluid has been added to the collecting tank. A heat exchanger is also provided.

Abstract: A method for closing a fillable collecting tank, in particular a fillable collecting tank of a heat exchanger for storing a fluid, having walls forming the collecting tank, wherein one of the walls is formed as a baseplate having openings for receiving pipes, wherein a filling opening for adding the fluid is provided in one of the walls, wherein the filling opening can be closed by the provision of a closure element that can be inserted into the filling opening or can be placed onto the filling opening after the fluid has been added to the collecting tank. A heat exchanger is also provided.