5 papers
UnWeaving the knots of GraphRAG -- turns out VectorRAG is almost enough
Ryszard Tuora, Mateusz GaliÅski, MichaÅ Godziszewski +4
One of the key problems in Retrieval-augmented generation (RAG) systems is that chunk-based retrieval pipelines represent the source chunks as atomic objects, mixing the informatio…
Query Symbolically or Retrieve Semantically? A Dataset and Method for Semi-Structured Question Answering
Mateusz Czyżnikiewicz, Ryszard Tuora, Adam Kozakiewicz +6
Retrieval-Augmented Generation (RAG) systems for question answering typically retrieve evidence by semantic similarity between the query and document chunks. While effective for un…
A Benchmark for Audio Reasoning Capabilities of Multimodal Large Language Models
Iwona Christop, Mateusz Czyżnikiewicz, PaweŠSkórzewski +4
The present benchmarks for testing the audio modality of multimodal large language models concentrate on testing various audio tasks such as speaker diarization or gender identific…
Preservation of Language Understanding Capabilities in Speech-aware Large Language Models
Marek Kubis, PaweŠSkórzewski, Iwona Christop +4
The paper presents C3T (Cross-modal Capabilities Conservation Test), a new benchmark for assessing the performance of speech-aware large language models. The benchmark utilizes tex…
Augmenting Polish Automatic Speech Recognition System With Synthetic Data
Åukasz Bondaruk, Jakub Kubiak, Mateusz Czyżnikiewicz
This paper presents a system developed for submission to Poleval 2024, Task 3: Polish Automatic Speech Recognition Challenge. We describe Voicebox-based speech synthesis pipeline a…