LoRA Fine-Tuning of English-Norwegian NMT for the Oil & Gas Industry
EAMT 2026 research on LoRA adaptation for low-resource petroleum-domain machine translation.
Research
Machine Translation
LoRA / PEFT
+5
Projects
I have hands-on experience taking projects from idea and experimentation to implementation and evaluation, covering both research projects and applied AI systems.
Beyond Routing: Diagnosing Modular LoRA Experts for Low-Resource Multilingual Petroleum-Domain Translation
A diagnostic study of modular LoRA experts for low-resource petroleum translation.
Research
Machine Translation
Petroleum-Domain MT
+5
Group-Level Training Data Attribution with Exact Shapley Analysis
Exact Shapley analysis of how training groups shape machine-translation behavior.
Research
Shapley Attribution
Training Data
+4
LongFinRAG: A Benchmark for Finding the Right Evidence in Annual Reports
A reliability-audited benchmark for evidence-grounded RAG on 15 long annual reports, with 720 QA items and traceable report-, page-, and object-level evidence.
Research
RAG
Information Retrieval
+3
When Does GraphRAG Help? A Controlled Study of Evidence Localization in Long Annual Reports
A controlled study of when graph structure improves evidence localization in long annual reports, separating object recovery, page coverage, and candidate competition.
Research
RAG
Information Retrieval
+1
When Relevance Is Not Enough: Aspect-Controlled Biomedical Evidence Retrieval
A controlled study of whether embeddings can retrieve different clinical evidence when the requested information changes.
Research
Information Retrieval
NLP
+2
When Data Cleaning Becomes Bias: Target-Standard Specialization in Norwegian MT
A controlled English-Norwegian MT study showing how Bokmål filtering creates target-standard specialization—and how reference choice can turn that specialization into an evaluation-bias problem.
Research
Responsible AI
Machine Translation
+1
Norwegian Petroleum Corpus
A reproducible pipeline for building English-Norwegian MT data and retrieval-ready text from public Equinor web pages and PDFs.
Research
Data Quality
Data-Centric ML
+2
Replay-Based Continual Adaptation for English-Norwegian Petroleum Machine Translation
A continual LoRA adaptation study examining new-source learning, old-source forgetting, and replay-based retention in petroleum-domain MT.
Research
Machine Translation
LoRA / PEFT
+2
Reusable Bias Evaluation Framework for LMs and VLMs
A reusable evaluation framework for matched-prompt and image-instruction bias studies, covering geographic, gender-occupation, and political/moral VLM framing cases.
Research
Responsible AI
NLP
+1
LoRA Fine-Tuning of English-Norwegian NMT for the Oil & Gas Industry
EAMT 2026 research on LoRA adaptation for low-resource petroleum-domain machine translation.
Research
Machine Translation
LoRA / PEFT
+5
Beyond Routing: Diagnosing Modular LoRA Experts for Low-Resource Multilingual Petroleum-Domain Translation
A diagnostic study of modular LoRA experts for low-resource petroleum translation.
Research
Machine Translation
Petroleum-Domain MT
+5
Group-Level Training Data Attribution with Exact Shapley Analysis
Exact Shapley analysis of how training groups shape machine-translation behavior.
Research
Shapley Attribution
Training Data
+4
LongFinRAG: A Benchmark for Finding the Right Evidence in Annual Reports
A reliability-audited benchmark for evidence-grounded RAG on 15 long annual reports, with 720 QA items and traceable report-, page-, and object-level evidence.
Research
RAG
Information Retrieval
+3
When Does GraphRAG Help? A Controlled Study of Evidence Localization in Long Annual Reports
A controlled study of when graph structure improves evidence localization in long annual reports, separating object recovery, page coverage, and candidate competition.
Research
RAG
Information Retrieval
+1
When Relevance Is Not Enough: Aspect-Controlled Biomedical Evidence Retrieval
A controlled study of whether embeddings can retrieve different clinical evidence when the requested information changes.
Research
Information Retrieval
NLP
+2
When Data Cleaning Becomes Bias: Target-Standard Specialization in Norwegian MT
A controlled English-Norwegian MT study showing how Bokmål filtering creates target-standard specialization—and how reference choice can turn that specialization into an evaluation-bias problem.
Research
Responsible AI
Machine Translation
+1
Norwegian Petroleum Corpus
A reproducible pipeline for building English-Norwegian MT data and retrieval-ready text from public Equinor web pages and PDFs.
Research
Data Quality
Data-Centric ML
+2
Replay-Based Continual Adaptation for English-Norwegian Petroleum Machine Translation
A continual LoRA adaptation study examining new-source learning, old-source forgetting, and replay-based retention in petroleum-domain MT.
Research
Machine Translation
LoRA / PEFT
+2
Reusable Bias Evaluation Framework for LMs and VLMs
A reusable evaluation framework for matched-prompt and image-instruction bias studies, covering geographic, gender-occupation, and political/moral VLM framing cases.
Research
Responsible AI
NLP
+1