Image de couverture de livre

Lost in Literalism: How Supervised Training Shapes Translationese in LLMs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Li, Yafu, Zhang, Ronghao, Wang, Zhilin, Zhang, Huajian, Cui, Leyang, Yin, Yongjing, Xiao, Tong, Zhang, Yue
Format:	Preprint
Publié:	2025
Sujets:	Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

What Have We Achieved on Non-autoregressive Translation?
par: Li, Yafu, et autres
Publié: (2024)

Spotting AI's Touch: Identifying LLM-Paraphrased Spans in Text
par: Li, Yafu, et autres
Publié: (2024)

LexMatcher: Dictionary-centric Data Collection for LLM-based Machine Translation
par: Yin, Yongjing, et autres
Publié: (2024)

MAGE: Machine-generated Text Detection in the Wild
par: Li, Yafu, et autres
Publié: (2023)

Translationese-index: Using Likelihood Ratios for Graded and Generalizable Measurement of Translationese
par: Liu, Yikang, et autres
Publié: (2025)

Lost in Translationese? Reducing Translation Effect Using Abstract Meaning Representation
par: Wein, Shira, et autres
Publié: (2023)

Training Models on Dialects of Translationese Shows How Lexical Diversity and Source-Target Syntactic Similarity Shape Learning
par: Kunz, Jenny
Publié: (2026)

Unveiling Attractor Cycles in Large Language Models: A Dynamical Systems View of Successive Paraphrasing
par: Wang, Zhilin, et autres
Publié: (2025)

Semformer: Transformer Language Models with Semantic Planning
par: Yin, Yongjing, et autres
Publié: (2024)

Characterizing, Evaluating, and Optimizing Complex Reasoning
par: Zhang, Haoran, et autres
Publié: (2026)

Decoding Machine Translationese in English-Chinese News: LLMs vs. NMTs
par: Kong, Delu, et autres
Publié: (2025)

SEE: Continual Fine-tuning with Sequential Ensemble of Experts
par: Wang, Zhilin, et autres
Publié: (2025)

Pretraining Language Models Using Translationese
par: Doshi, Meet, et autres
Publié: (2024)

Reasoning over Boundaries: Enhancing Specification Alignment via Test-time Deliberation
par: Zhang, Haoran, et autres
Publié: (2025)

From Drafts to Answers: Unlocking LLM Potential via Aggregation Fine-Tuning
par: Li, Yafu, et autres
Publié: (2025)

Alleviating Hallucinations of Large Language Models through Induced Hallucinations
par: Zhang, Yue, et autres
Publié: (2023)

Potential and Challenges of Model Editing for Social Debiasing
par: Yan, Jianhao, et autres
Publié: (2024)

Mitigating Translationese Bias in Multilingual LLM-as-a-Judge via Disentangled Information Bottleneck
par: Zhang, Hongbin, et autres
Publié: (2026)

Keys to Robust Edits: from Theoretical Insights to Practical Advances
par: Yan, Jianhao, et autres
Publié: (2024)

New Skills or Sharper Primitives? A Probabilistic Perspective on the Emergence of Reasoning in RLVR
par: Wang, Zhilin, et autres
Publié: (2026)

Measuring Spurious Correlation in Classification: 'Clever Hans' in Translationese
par: Borah, Angana, et autres
Publié: (2023)

GSM-Plus: A Comprehensive Benchmark for Evaluating the Robustness of LLMs as Mathematical Problem Solvers
par: Li, Qintong, et autres
Publié: (2024)

A Dataset for Probing Translationese Preferences in English-to-Swedish Translation
par: Kunz, Jenny, et autres
Publié: (2026)

Beyond English-Centric Training: How Reinforcement Learning Improves Cross-Lingual Reasoning in LLMs
par: Huang, Shulin, et autres
Publié: (2025)

DiffuGuard: How Intrinsic Safety is Lost and Found in Diffusion Large Language Models
par: Li, Zherui, et autres
Publié: (2025)

Understanding How Value Neurons Shape the Generation of Specified Values in LLMs
par: Su, Yi, et autres
Publié: (2025)

The Comparison of Translationese in Machine Translation and Human Transation in terms of Translation Relations
par: Zhou, Fan
Publié: (2024)

Beyond Literal Mapping: Benchmarking and Improving Non-Literal Translation Evaluation
par: Tian, Yanzhi, et autres
Publié: (2026)

Mitigating Translationese in Low-resource Languages: The Storyboard Approach
par: Kuwanto, Garry, et autres
Publié: (2024)

Leveraging Entailment Judgements in Cross-Lingual Summarisation
par: Zhang, Huajian, et autres
Publié: (2024)

Under the Shadow of Babel: How Language Shapes Reasoning in LLMs
par: Wang, Chenxi, et autres
Publié: (2025)

KOSMOS-2.5: A Multimodal Literate Model
par: Lv, Tengchao, et autres
Publié: (2023)

Understanding In-Context Learning from Repetitions
par: Yan, Jianhao, et autres
Publié: (2023)

An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuning
par: Luo, Yun, et autres
Publié: (2023)

Not All Needles Are Found: How Fact Distribution and Don't Make It Up Prompts Shape Literal Extraction, Logical Inference, and Hallucination Risks in Long-Context LLMs
par: Ebrahimzadeh, Amirali, et autres
Publié: (2026)

Right Is Not Enough: The Pitfalls of Outcome Supervision in Training LLMs for Math Reasoning
par: Guo, Jiaxing, et autres
Publié: (2025)

Retracing the Past: LLMs Emit Training Data When They Get Lost
par: Ko, Myeongseob, et autres
Publié: (2025)

CopyBench: Measuring Literal and Non-Literal Reproduction of Copyright-Protected Text in Language Model Generation
par: Chen, Tong, et autres
Publié: (2024)

Exploring the Reliability of Large Language Models as Customized Evaluators for Diverse NLP Tasks
par: Li, Qintong, et autres
Publié: (2023)

TIM: Teaching Large Language Models to Translate with Comparison
par: Zeng, Jiali, et autres
Publié: (2023)