09:00 - 09:10 Opening remarks by Atul Kr. Ojha (on behalf of workshop chairs)
09:10 - 10:10 Invited talk 1: Ondřej Dušek (Institute of Formal and Applied Linguistics, Charles University, Prague (Czech Republic))
Chair: Chao-Hong Liu (ITRI, Potamu Research)
Title - How (Not) to Find Errors in LLM Outputs
Abstract: While LLMs have substantially improved the quality of generated texts, they still tend to make errors in their outputs, which can be subtle and harder to find than for older approaches. This needs to be reflected in the evaluation, where standard metrics or simple scores may not capture errors easily. While human evaluation should produce better results, we find a lot of inconsistency and underspecification in practice.
Building on previous works in machine translation, we examine annotating individual spans of texts for errors in order to get more detailed evaluation feedback. We explore span annotation through both human evaluation and LLM-as-judge evaluation. We provide a unified interface for both LLM and human authored error annotations, we examine different methods of obtaining LLM-annotated spans and introduce LLM ensembles for higher robustness. We directly compare LLMs and humans on the same task, finding that LLMs are able to reach high correlation with human assessments and, depending on the domain, can match trained human crowd workers in performance. However, we also report many caveats on both the human and LLM side, and we discuss potential further improvements of the evaluation setup.
10:10 - 10:30 Session 1: Findings of Turkic Low Resource Machine Translation Challenge
Chair: TBD
10:30 - 11:00 COFFEE/TEA BREAK
11:00 - 12:30 Session 2: Scientific Research Papers
Chair: Nathaniel Oco (De La Salle University)
11:00-11:15 - Are Small Language Models the Silver Bullet to Low-Resource Languages Machine Translation? - Yewei Song, Lujun LI, Cedric Lothritz, Saad Ezzini, Lama Sleem, Niccolo’ Gentile, Radu State, Tegawendé F. Bissyandé and Jacques Klein
11:15- 11:30 - Tao–Filipino Neural Machine Translation: Strategies for Ultra–Low-Resource Settings - Adrian Denzel Macayan, Luis Andrew Sunga Madridijo, Ellexandrei Esponilla and Zachary Mitchell Francisco
11:30-11:45 - Building and Evaluating a High Quality Parallel Corpus for English Urdu Low Resource Machine Translation - Munief Hassan Tahir, Hunain Azam, Sana Shams and Sarmad Hussain
11:45-12:00 - Text Filter Based on Automatically Acquired Vocabularies for Multilingual Machine Translation - Kenji Imamura and Masao Utiyama
12:00-12:15 - Improving Indigenous Language Machine Translation with Synthetic Data and Language-Specific Preprocessing - Aashish Dhawan, Christopher Driggers-Ellis, Christan Grant and Daisy Zhe Wang
12:15-12:30 - Adapting Multilingual NMT to Language Isolates: The Role of Proxy Language Selection and Dialect Handling for Nivkh - Eleonora Izmailova, Alexey Sorokin and Pavel Grashchenkov
12:30 - 14:00 LUNCH
14:00 - 15:00 Invited talk 2: TBD
Chair: Atul Kr. Ojha (University of Galway)
Title - TBD
Abstract: TBD
15:00 - 16:00 Session 3: Poster Session
Chair: Valentin Malykh
Can LLMs Translate Italy’s Language Varieties? - Edoardo Signoroni, Pavel Rychlý
Balancing Fluency and Adherence: Hybrid Fallback Term Injection in Low-Resource Terminology Translation - Kurt Abela, Marc Tanti, Claudia Borg
Assessing and Improving Punctuation Robustness in English-Marathi Machine Translation - Kaustubh Shivshankar Shejole, Sourabh Deoghare, Pushpak Bhattacharyya
Can Linguistically Related Languages Guide LLM Translation in Low-Resource Settings? - Aishwarya Ramasethu, Rohin Garg, Niyathi Allu, Harshwardhan Fartale and Dun Li Chan
Navigating Data Scarcity in Low-Resource English-Tatar Translation using LLM Fine-Tuning - Ahmed Khaled Khamis
No One-Size-Fits-All: Building Systems For Translation to Bashkir, Kazakh, Kyrgyz, Tatar and Chuvash Using Synthetic And Original Data - Dmitry Karpov
DevLake at LoResMT 2026: The Impact of Pre-training and Model Scale on Russian-Bashkir Low-Resource Translation - Vyacheslav Tyurin
A Comparative Evaluation of Open-Source Models for Russian-Kazakh Translation -Gleb Shanshin
Script Correction and Synthetic Pivoting: Adapting Tencent HY-MT for Low-Resource Turkic Translation - Bolgov Maxim
Machine Translation for Low Resource Turkic Languages: English-Tatar - Alexander Dikov
Data-Centric Approach at the LoResMT 2026 Turkic Translation Challenge: Russian-Kyrgyz - Dmitry Novokshanov
LoResMT 2026 Shared Task System Description - Vladimir Panov
Ensemble Methods for Low-Resource Russian-Kyrgyz Machine Translation: When Diverse Models Beat Better Models - Adilet Metinov
15:30 - 16:00 COFFEE/TEA BREAK
16:00 - 17:30 Session 4: Scientific Research Papers
Chair: Ekaterina Vylomova (University of Melbourne, Australia)
16:00-16:15 - Comparing LLM-Based Translation Approaches for Extremely Low-Resource Languages - Jared Coleman, Ruben Rosales, Kira Toal, Diego Cuadros, Nicholas Leeds, Bhaskar Krishnamachari and Khalil Iskarous
16:15-16:30 - Context Volume Drives Performance: Tackling Domain Shift in Extremely Low-Resource Translation via RAG - David Samuel Setiawan, Raphael Merx and Jey Han Lau
16:30-16:45 - Semi-Automatic construction of a Quechua-Spanish dictionary - Maximiliano Duran and Max Silberztein
16:45-17:00 - A Fine-Grained Linguistic Evaluation of Low-Resource Luxembourgish–English MT - Nils Rehlinger
17:00-17:15 - CTC Regularization for Low-Resource Speech-to-Text Translation - Zachary William Hopton and Rico Sennrich
17:15 -17:25 Closing remarks by Atul Kr. Ojha ( on behalf of workshop chairs)