Real-time Relevance: RAG with Dynamic Context for Improved Natural Language Responses

Visa enkel post
dc.contributor.authorLandgren, Malte
dc.contributor.authorGiljegård, Oskar
dc.contributor.departmentChalmers tekniska högskola / Institutionen för data och informationstekniksv
dc.contributor.departmentChalmers University of Technology / Department of Computer Science and Engineeringen
dc.contributor.examinerJohansson, Richard
dc.contributor.supervisorJohansson, Richard
dc.date.accessioned2024-10-17T13:49:18Z
dc.date.available2024-10-17T13:49:18Z
dc.date.issued2024
dc.date.submitted
dc.description.abstractToday’s Retrieval Augmented Generation (RAG) systems often struggle when trying to answer questions that require complex multi-hop reasoning. In this thesis we investigate an autoregressive Large Language Model (LLM) architecture which can generate a real-time relevant dense search vector for every token generation step. To facilitate this we also develop a synthetic data generation technique to acquire search query vector labels on a token-by-token level, requiring only a generating LLM and a document database. We investigate the quality of the synthetic data, and provide an attention based relabeling method which decreases hallucinations, improving the correctness of the labels by 67%. The architecture is able to produce query vectors 27 times faster than a separate embedder at the cost of retrieval accuracy. Finally, we train and employ the model in an active retrieval question-answering setting.
dc.identifier.coursecodeDATX05
dc.identifier.urihttp://hdl.handle.net/20.500.12380/308927
dc.language.isoeng
dc.setspec.uppsokTechnology
dc.subjectLLM
dc.subjectRAG
dc.subjectactive retrieval
dc.subjectsynthetic data generation
dc.subjectmaster thesis
dc.titleReal-time Relevance: RAG with Dynamic Context for Improved Natural Language Responses
dc.type.degreeExamensarbete för masterexamensv
dc.type.degreeMaster's Thesisen
dc.type.uppsokH
local.programmeComputer science – algorithms, languages and logic (MPALG), MSc
local.programmeData science and AI (MPDSC), MSc
Ladda ner
Original bundle
Visar 1 - 1 av 1
Bild (thumbnail)
Namn:
CSE 24-42 ML OG.pdf
Storlek:
1.97 MB
Format:
Adobe Portable Document Format
Beskrivning:
Ladda ner
License bundle
Visar 1 - 1 av 1
Bild (thumbnail)
Namn:
license.txt
Storlek:
2.35 KB
Format:
Item-specific license agreed upon to submission
Beskrivning:
Ladda ner
Samling
Examensarbeten för masterexamen