Alexander Fraser
Professor
Researches machine translation, language modelling and multilingual NLP, including methods for low-resource languages developed with language communities.
Alexander Fraser's lab at TUM works on multilingual, translation & linguistics and large language models. 43 papers since 2023, 15 at top venues, mostly EMNLP and Findings of ACL. 9 PhD students. 9 alumni, most often to BeOne Stuttgart GmbH and Google.
At a glance
- papers, 2023–2026
- 11/ year
- at T1 venues
- 35%
- PhD students
- 9
- MS students
- 0
- alumni listed
- 9
Papers by yearfrom the lab homepage, 2023–2026
- 0 papers2026 · T1 0 · collecting
- 19 papers2025 · T1 10
- 13 papers2024 · T1 4
- 11 papers2023 · T1 1
T1 venuesOtherVenues: EMNLP, Findings of ACL, ACL, LREC, Findings of EMNLP
Research focuspaper titles that use each topic's words
- Large language models23 papers
- Multilingual, translation & linguistics9 papers
- low-resource languages2 papers
- language modelling0 papers
- multilingual nlp0 papers
Paper list
From the lab homepage.
202519 papers
- FieldMattersA Practical Tool to Help Automate Interlinear Glossing: A Study on Mukrī Kurdish
- NAACL (T1)Beyond Literal Token Overlap: Token Alignability for Multilinguality
- ComputELBilingual Sentence Mining for Low-Resource Languages: A Case Study on Upper and Lower Sorbian
- WOAHCan Prompting LLMs Unlock Hate Speech Detection Across Languages? A Zero-Shot and Few-Shot Study
- EMNLP (T1)Data-Efficient Hate Speech Detection via Cross-Lingual Nearest Neighbour Retrieval with Limited Labeled Data
- In Advances in Neural Information Processing Systems 38 (NeurIPS ), Datasets and Benchmarks Track, San Diego, December.DCAD-2000: A Multilingual Dataset across 2000+ Languages with Data Cleaning as Anomaly Detection
- Findings of EMNLPEmoBench-UA: A Benchmark Dataset for Emotion Detection in Ukrainian
- Findings of ACLEXECUTE: A Multilingual Benchmark for LLM Token Understanding
- EMNLP (T1)Extracting Linguistic Information from Large Language Models: Syntactic Relations and Derivational Knowledge
- WMTFindings of the WMT 2025 Shared Task LLMs with Limited Resources for Slavic Languages: MT and QA
- NAACL (T1)Fine-Grained Transfer Learning for Harmful Content Detection through Label-Specific Soft Prompt Tuning
- EMNLP (T1)From Unaligned to Aligned: Scaling Multilingual LLMs with Multi-Way Parallel Corpora
202413 papers
- WOAHA Study of the Class Imbalance Problem in Abusive Language Detection
- LRECAnalyzing the Understanding of Morphologically Complex Words in Large Language Models
- Proceedings of the BabyLM Challenge at the Conference on Computational Natural Language Learning, Miami, FL, USA, November.Are BabyLMs Second Language Learners?
- EMNLP (T1)CUTE: Measuring LLMs' Understanding of Their Tokens
- EMNLP (T1)Hate Personified: Investigating the role of LLMs in content moderation
- LRECHow to Solve Few-Shot Abusive Content Detection Using the Data We Actually Have
- Proceedings of the Workshop on Natural Language Processing Meets Climate Change (ClimateNLP ), Bangkok, Thailand, August.Informing climate risk analysis using textual information – A research agenda
- Findings of ACLLLMs Beyond English: Scaling the Multilingual Capability of LLMs with Cross-Lingual Feedback
- LRECReconstruction of Cuneiform Literary Texts as Text Matching
- EMNLP (T1)Style-Specific Neurons for Steering LLMs in Text Style Transfer
- EMNLP (T1)Subword Segmentation in LLMs: Looking at Inflection and Consistency
- Findings of ACLUnderstanding Cross-Lingual Alignment—A Survey
202311 papers
- EMNLP (T1)A Study on Accessing Linguistic Information in Pre-Trained Language Models by Using Prompts
- EACLA Survey of Methods for Addressing Class Imbalance in Deep-Learning Based Natural Language Processing
- Findings of EACLAdapterSoup: Weight Averaging to Improve Generalization of Pretrained Language Models
- Findings of ACLExploring anisotropy and outliers in multilingual language models for cross-lingual semantic sentence similarity
- Journal of Language Resources and EvaluationLabel modification and bootstrapping for zero-shot cross-lingual hate speech detection
- LoResMTLanguage-Family Adapters for Low-Resource Multilingual Neural Machine Translation
- Final Workshop of the evaluation campaign EVALITALMU at HaSpeeDe3: Multi-Dataset Training for Cross-Domain Hate Speech Detection
- Findings of EMNLPMitigating Data Imbalance and Representation Degeneration in Multilingual Machine Translation
- MRLMultilingual Word Embeddings for Low-Resource Languages using Anchors and a Chain of Related Languages
- IWSLTOn the Copying Problem of Unsupervised NMT: A Training Schedule with a Language Discriminator Loss
- Findings of ACLSpeaking Multiple Languages Affects the Moral Bias of Language Models
Where alumni went9 listed, 38% to industry
Industry 3Academia 5
Most often to
- BeOne Stuttgart GmbH
- Microsoft
Also to
- Qatar Computing Research Institute. Currently: Associate Professor
- University of Amsterdam with Khalil Sima'an
Apply to TUMRolling
Not announced yet
Computer Science Doctorate (CIT)
- Round
- Rolling (position-based)
- Opens
- —
- GRE
- See program page
- English
- Very good written and spoken English; no minimum test score
- Fee
- See program page
Confirmed on the program page.
Similar labsnearest by research summary
- CNSJTU
Chen Qian
- Large language models
- GBUCL
Pontus Stenetorp
- Generative & foundation models
- CNPeking
Dongyan Zhao
- Knowledge, QA & retrieval
- USCMU
Daniel Fried
- Large language models
- CNTsinghua
Yang Liu
- Multilingual, translation & linguistics
- USUMD
Marine Carpuat
- Multilingual, translation & linguistics
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.