RLSR trains source rewriters via RL with translation-quality improvement as the reward, outperforming prompt baselines at 4B scale while matching larger models.
hub
Findings of the WMT 25 general machine translation shared task: Time to stop evaluating on easy test sets
19 Pith papers cite this work. Polarity classification is still indexing.
19
Pith papers citing it
hub tools
citation-role summary
other 1
citation-polarity summary
years
2026 19roles
other 1polarities
unclear 1representative citing papers
Ouvia is a user-centered evaluation framework for speech translation usability in real-world scenarios, showing limited usability rates and the superiority of QA-based metrics.
Automatic evaluation tools for literary translations correlate poorly with expert human judgments on creativity and exhibit bias favoring machine-translated texts.
Document-level machine translation followed by segment-level LLM refinement provides the strongest and most stable improvements in literary translation quality, mainly enhancing fluency and style rather than adequacy.
Human readers prefer human literary translations over AI-generated ones for immersion and clarity despite finding MT adequate and struggling to identify the source.
Reward models for LLMs frequently select socially undesirable options across four social domains, show no overall best performer, and exhibit a bias-avoidance versus context-sensitivity trade-off.
Empirical study finds verbalized per-token confidence methods in LLMs for MT perform similarly to internal signals on error detection and calibration but show little correlation.
Multi-aspect iterative refinement with specialized LLMs generates superior literary translation data, enabling SFT and GRPO to produce LitMT-8B and LitMT-14B models scoring 67.25 and 69.07 CEA100 on MetaphorTrans, competitive with Claude Sonnet 4.5.
Lexical richness is a robust linguistic signal for AI-generated text detection across models and domains, while most other features are context-dependent.
Cross-lingual transfer and language-specific data efforts are interdependent and complementary for effective low-resource NLP, as demonstrated through Luxembourgish case studies and synthesis.
Compact 0.8B-7B models for bidirectional Japanese-English translation outperform large multilingual models on real-world domain benchmarks.
Introduces LLM Consumer Behavior Theory to analyze consumer behavior when LLMs serve as autonomous decision-making agents in markets.
A cascaded SimulST system using Parakeet and Qwen 3.5 with adaptive black-box policies and RAG context achieves +5.82 XCOMET-XL improvement on En→De for IWSLT 2026.
A feature-based decision tree with parsing-derived signals and heuristics detects LLM-generated code in a lightweight, CPU-only setup for SemEval-2026 Task 13.
citing papers explorer
-
Rewrite to Translate, Translate to Reward: Reinforcement Learning for Source Rewriting in Machine Translation
RLSR trains source rewriters via RL with translation-quality improvement as the reward, outperforming prompt baselines at 4B scale while matching larger models.
-
Ouvia: A User-centered Framework for Measuring Usability of Speech Translation in Real-World Communication Scenarios
Ouvia is a user-centered evaluation framework for speech translation usability in real-world scenarios, showing limited usability rates and the superiority of QA-based metrics.
-
Creativity Bias: How Machine Evaluation Struggles with Creativity in Literary Translations
Automatic evaluation tools for literary translations correlate poorly with expert human judgments on creativity and exhibit bias favoring machine-translated texts.
-
What Does LLM Refinement Actually Improve? A Systematic Study on Document-Level Literary Translation
Document-level machine translation followed by segment-level LLM refinement provides the strongest and most stable improvements in literary translation quality, mainly enhancing fluency and style rather than adequacy.
-
AI translation of literary texts is "fine", but readers still prefer human translations
Human readers prefer human literary translations over AI-generated ones for immersion and clarity despite finding MT adequate and struggling to identify the source.
-
Misaligned by Reward: Socially Undesirable Preferences in LLMs
Reward models for LLMs frequently select socially undesirable options across four social domains, show no overall best performer, and exhibit a bias-avoidance versus context-sensitivity trade-off.
-
Speaking in Self-Assessing Tongues: On the Verbalized Confidence of LLMs in Machine Translation
Empirical study finds verbalized per-token confidence methods in LLMs for MT perform similarly to internal signals on error detection and calibration but show little correlation.
-
Better Literary Translation: A Multi-Aspect Data Generation and LLM Training Approach
Multi-aspect iterative refinement with specialized LLMs generates superior literary translation data, enabling SFT and GRPO to produce LitMT-8B and LitMT-14B models scoring 67.25 and 69.07 CEA100 on MetaphorTrans, competitive with Claude Sonnet 4.5.
-
A Systematic Analysis of Linguistic Features in AI-Generated Text Detection Across Domains and Models
Lexical richness is a robust linguistic signal for AI-generated text detection across models and domains, while most other features are context-dependent.
-
Why Low-Resource NLP Needs More Than Cross-Lingual Transfer: Lessons Learned from Luxembourgish
Cross-lingual transfer and language-specific data efforts are interdependent and complementary for effective low-resource NLP, as demonstrated through Luxembourgish case studies and synthesis.
-
CAT-Translate: Building Compact Open-Source Models for Japanese-English Translation
Compact 0.8B-7B models for bidirectional Japanese-English translation outperform large multilingual models on real-world domain benchmarks.
-
LLM Consumer Behavior Theory: Foundations of a Novel Research Field
Introduces LLM Consumer Behavior Theory to analyze consumer behavior when LLMs serve as autonomous decision-making agents in markets.
-
MLLP-VRAIN UPV system for the IWSLT 2026 Simultaneous Speech Translation task
A cascaded SimulST system using Parakeet and Qwen 3.5 with adaptive black-box policies and RAG context achieves +5.82 XCOMET-XL improvement on En→De for IWSLT 2026.
-
FMI_SU_Yotkova_Kastreva at SemEval-2026 Task 13: Lightweight Detection of LLM-Generated Code via Stylometric Signals
A feature-based decision tree with parsing-derived signals and heuristics detects LLM-generated code in a lightweight, CPU-only setup for SemEval-2026 Task 13.
- Hy-MT2: A Family of Fast, Efficient and Powerful Multilingual Translation Models in the Wild
- A Recipe for Long-Context Reasoning in Large Language Models via On-Policy Optimization and Distillation
- Dynamic Meta-Metrics: Source-Sentence Conditioned Weighting for MT Evaluation
- Psychologically Potent, Computationally Invisible: LLMs Generate Social-Comparison-Eliciting Posts They Fail to Detect
- Syntax as a Rosetta Stone: Universal Dependencies for In-Context Coptic Translation