With the rapid development of Large Language Models (LLMs), it is crucial to have benchmarks which can evaluate the ability of LLMs on different domains. One common use of LLMs is performing tasks on scientific topics, such as writing algorithms, querying databases or giving mathematical proofs. Inspired by the way university students are evaluated on such tasks, in this paper, we propose SciEx - a benchmark consisting of university computer science exam questions, to evaluate LLMs’ ability on solving scientific tasks. SciEx is (1) multilingual, containing both English and German exams, and (2) multi-modal, containing questions that involve images, and (3) contains various types of freeform questions with different difficulty levels, due to the nature of university exams. We evaluate the performance of various state-of-the-art LLMs on our new benchmark. Since SciEx questions are freeform, it is not straightforward to evaluate LLM performance. Therefore, we provide human expert grading of the LLM outputs on SciEx. We show that the free-form exams in SciEx remain challenging for the current LLMs, where the best LLM only achieves 59.4% exam grade on average. We also provide detailed comparisons between LLM performance and student performance on SciEx. To enable future evaluation of new LLMs, we propose using LLM-as-a-judge to grade the LLM answers on SciEx. Our experiments show that, although they do not perform perfectly on solving the exams, LLMs are decent as graders, achieving 0.948 Pearson correlation with expert grading.

Word embeddings are powerful tools that facilitate better analysis of natural language. However, their quality highly depends on the resource used for training. There are various approaches relying on n-gram corpora, such as the Google n-gram corpus. However, n-gram corpora only offer a small window into the full text – 5 words for the Google corpus at best. This gives way to the concern whether the extracted word semantics are of high quality. In this paper, we address this concern with two contributions. First, we provide a resource containing 120 word-embedding models – one of the largest collection of embedding models. Furthermore, the resource contains the n-gramed versions of all used corpora, as well as our scripts used for corpus generation, model generation and evaluation. Second, we define a set of meaningful experiments allowing to evaluate the aforementioned quality differences. We conduct these experiments using our resource to show its usage and significance. The evaluation results confirm that one generally can expect high quality for n-grams with n > 3.