13 papers
Beyond Clean Accuracy: The Practical Challenges and Benefits of Pixel-Based Language Models
Haowen Tang, Wei Xia, Raoyuan Zhao +2
Pixel-based language models process rendered text without relying on a fixed subword vocabulary, offering a promising approach to noisy and multilingual inputs. However, their prac…
Last Translation Benchmark
Vilém Zouhar, Niyati Bafna, Mukund Choudhary +263
For scientific progress, we need benchmarks that test the limits of state-of-the-art models, and evaluation methods that inform us about failure cases. As models get stronger, stan…
Beyond Input Understanding: Diagnosing Multilingual Mathematical Reasoning with Directed Acyclic Trace Graphs
Jiaqiao Zhang, Zhoujun Li, Zhuojun Li +6
Large reasoning models (LRMs) achieve strong mathematical reasoning performance in English, but remain much less reliable in many low- and medium-resource languages. This gap is of…
ReverseMath: Answer Inversion for Scalable and Verifiable Mathematical Problem Generation
Raoyuan Zhao, Yihong Liu, Yupei Du +2
Mathematical reasoning benchmarks are vital for evaluating large language models (LLMs), but many are static and repeatedly exposed through public evaluation and training pipelines…
Crosslingual On-Policy Self-Distillation for Multilingual Reasoning
Yihong Liu, Raoyuan Zhao, Michael A. Hedderich +1
Large language models (LLMs) have achieved remarkable progress in mathematical reasoning, but this ability is not equally accessible across languages. Especially low-resource langu…
Large Reasoning Models Are (Not Yet) Multilingual Latent Reasoners
Yihong Liu, Raoyuan Zhao, Hinrich Schütze +1
Large reasoning models (LRMs) achieve strong performance on mathematical reasoning tasks, often attributed to their capability to generate explicit chain-of-thought (CoT) explanati…
Human Uncertainty-Aware Data Selection and Automatic Labeling in Visual Question Answering
Jian Lan, Zhicheng Liu, Udo Schlegel +5
Large vision-language models (VLMs) achieve strong performance in Visual Question Answering but still rely heavily on supervised fine-tuning (SFT) with massive labeled datasets, wh…
A Comprehensive Evaluation of Multilingual Chain-of-Thought Reasoning: Performance, Consistency, and Faithfulness Across Languages
Raoyuan Zhao, Yihong Liu, Hinrich Schütze +1
Large reasoning models (LRMs) increasingly rely on step-by-step Chain-of-Thought (CoT) reasoning to improve task performance, particularly in high-resource languages such as Englis…
Evaluating Robustness of Large Language Models Against Multilingual Typographical Errors
Raoyuan Zhao, Yihong Liu, Lena Altinger +2
Large language models (LLMs) are increasingly deployed in multilingual, real-world applications with user inputs -- naturally introducing \emph{typographical errors} (typos). Yet m…
Do We Know What LLMs Don't Know? A Study of Consistency in Knowledge Probing
Raoyuan Zhao, Abdullatif Köksal, Ali Modarressi +2
The reliability of large language models (LLMs) is greatly compromised by their tendency to hallucinate, underscoring the need for precise identification of knowledge gaps within L…
What's the Difference? Supporting Users in Identifying the Effects of Prompt and Model Changes Through Token Patterns
Michael A. Hedderich, Anyi Wang, Raoyuan Zhao +3
Prompt engineering for large language models is challenging, as even small prompt perturbations or model changes can significantly impact the generated output texts. Existing evalu…
MAKIEval: A Multilingual Automatic WiKidata-based Framework for Cultural Awareness Evaluation for LLMs
Raoyuan Zhao, Beiduo Chen, Barbara Plank +1
Large language models (LLMs) are used globally across many languages, but their English-centric pretraining raises concerns about cross-lingual disparities for cultural awareness,…
SYNTHEVAL: Hybrid Behavioral Testing of NLP Models with Synthetic CheckLists
Raoyuan Zhao, Abdullatif Köksal, Yihong Liu +3
Traditional benchmarking in NLP typically involves using static held-out test sets. However, this approach often results in an overestimation of performance and lacks the ability t…