Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by integrating external knowledge, where the LLM's ability to generate responses based on the combination of a given query and retrieved documents is crucial. However, most benchmarks focus on overall RAG system performance, rarely assessing LLM-specific capabilities. Current benchmarks emphasize broad aspects such as noise robustness, but lack a systematic and granular evaluation framework on document utilization. To this end, we introduce \textit{Placeholder-RAG-Benchmark}, a multi-level fine-grained benchmark, emphasizing the following progressive dimensions: (1) multi-level filtering abilities, (2) combination abilities, and (3) reference reasoning. To provide a more nuanced understanding of LLMs' roles in RAG systems, we formulate an innovative placeholder-based approach to decouple the contributions of the LLM's parametric knowledge and the external knowledge. Experiments demonstrate the limitations of representative LLMs in the RAG system's generation capabilities, particularly in error resilience and context faithfulness. Our benchmark provides a reproducible framework for developing more reliable and efficient RAG systems. Our code is available in https://github.com/Alipay-Med/PRGB.

Russia's territorial gains in Ukraine are slowing down dramatically, two analyses have found, continuing a pattern from 2024 at a time when both nations are trying to project strength in the face of United States-mediated negotiations aimed at ending the war. Britain's Ministry of Defence last week estimated that Russian forces seized 143sq km (55sq miles) of Ukrainian land in March, compared with 196sq km (76sq miles) in February and 326sq km (126sq miles) in January. The Institute for the Study of War, a Washington, DC-based think tank, spotted the same trend, estimating Russian gains at 203sq km (78sq miles) in March, 354sq km (137sq miles) in February and 427sq km (165sq miles) in January. These estimates are based on satellite imagery and geolocated open-source photography rather than claims by either side. Should this trend continue, Russian forces could come to a standstill by early summer, roughly coinciding with US President Donald Trump's self-imposed early deadline for achieving a ceasefire.

We present the first neural machine translation system for translation between the endangered Erzya language and Russian and the dataset collected by us to train and evaluate it. The BLEU scores are 17 and 19 for translation to Erzya and Russian respectively, and more than half of the translations are rated as acceptable by native speakers. We also adapt our model to translate between Erzya and 10 other languages, but without additional parallel data, the quality on these directions remains low. We release the translation models along with the collected text corpus, a new language identification model, and a multilingual sentence encoder adapted for the Erzya language. These resources will be available at https://github.com/slone-nlp/myv-nmt.