CheckMIABench converts LLMs with intermediate checkpoints into clean MIA testbeds by using pre- and post-checkpoint training data from the same distribution and evaluates published attacks on Pythia and OLMo models while releasing an open-source library.
hub
A Call for Clarity in Reporting BLEU Scores
19 Pith papers cite this work, alongside 48 external citations. Polarity classification is still indexing.
19
Pith papers citing it
48
external citations · OpenAlex
hub tools
citation-role summary
method 3
background 1
citation-polarity summary
representative citing papers
LLM in-context translation accuracy falls sharply with larger grammars and longer sentences, and drops further when source and target languages differ in morphology or writing system, with common errors including wrong word recall, hallucinations, and untranslated source words.
IPQA is a new benchmark that measures how well models identify core user intents from history in personalized question answering, finding that performance is poor and declines with greater question complexity.
Low-resource languages are structurally more different from English in LLMs than high- or mid-resource ones, and language-specific post-training alters structures while preserving inter-language relationships.
A multi-reference audit framework for LLM translations of the Pali Canon uses embedding drift from a human reference centroid to triage candidates for LLM-judge adjudication, showing drift correlates with major error rates and model-specific differences in the high-drift tail.
ATD-Trans is a new geographically annotated Japanese-English travelogue dataset that reveals Japanese-enhanced models perform better on geo-entity translation while domestic Japanese locations remain harder to translate accurately.
Dynamic Meta-Metrics learns source-sentence conditioned combinations of MT metrics, with MLP-based and soft-conditioned versions showing gains over linear and GP ensembles on WMT data.
BLOOM is a 176B-parameter open-access multilingual language model trained on the ROOTS corpus that achieves competitive performance on benchmarks, with improved results after multitask prompted finetuning.
A sparsely gated mixture-of-experts model trained on newly mined low-resource data achieves 44% relative BLEU improvement across 200 languages while adding human safety evaluation.
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.
Introduces LLM-FACETS, a privacy-preserving open-source framework for LLM evaluation using deterministic metrics locally, LLM-judge metrics with user-controlled APIs, and mechanisms for uncertainty visualization and hallucination detection.
Experiments show domain match and language relatedness drive knowledge transfer in multilingual MT more than vocabulary overlap.
Label noise hurts fine-tuning performance most while grammatical and typographical noise sometimes act as mild regularizers, with changes concentrated in task-specific layers.
HAMR combines bi-level meta-learned instance reweighting with KNN-based neighborhood resampling to improve NLP performance under class imbalance, evaluated on six NER and classification benchmarks.
Mistral uses careful lexical simplification to raise readability while keeping BERTScore at 0.91 comparable to humans, whereas QWen improves readability but shows a disconnect with its 0.89 BERTScore in biomedical text simplification.
Compact 0.8B-7B models for bidirectional Japanese-English translation outperform large multilingual models on real-world domain benchmarks.
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.
Lius improves LLM translation for Kupang Malay by 4-13 points over baselines via continual instruction tuning with dictionary-derived instructions.
A literature survey that organizes prompting, fine-tuning, preference optimization, and context-aware techniques for LLM-based machine translation with emphasis on low-resource languages.
citing papers explorer
-
CheckMIABench: Firm Foundations For Membership Inference Attacks on Language Models
CheckMIABench converts LLMs with intermediate checkpoints into clean MIA testbeds by using pre- and post-checkpoint training data from the same distribution and evaluates published attacks on Pythia and OLMo models while releasing an open-source library.
-
Evaluating In-Context Translation with Synchronous Context-Free Grammar Transduction
LLM in-context translation accuracy falls sharply with larger grammars and longer sentences, and drops further when source and target languages differ in morphology or writing system, with common errors including wrong word recall, hallucinations, and untranslated source words.
-
IPQA: A Benchmark for Core Intent Identification in Personalized Question Answering
IPQA is a new benchmark that measures how well models identify core user intents from history in personalized question answering, finding that performance is poor and declines with greater question complexity.
-
Multilinguality of Large Language Models From a Structural Perspective
Low-resource languages are structurally more different from English in LLMs than high- or mid-resource ones, and language-specific post-training alters structures while preserving inter-language relationships.
-
From Outliers to Errors: Auditing Pali-to-English LLM Translations with Multi-Reference Adjudication
A multi-reference audit framework for LLM translations of the Pali Canon uses embedding drift from a human reference centroid to triage candidates for LLM-judge adjudication, showing drift correlates with major error rates and model-specific differences in the high-drift tail.
-
ATD-Trans: A Geographically Grounded Japanese-English Travelogue Translation Dataset
ATD-Trans is a new geographically annotated Japanese-English travelogue dataset that reveals Japanese-enhanced models perform better on geo-entity translation while domestic Japanese locations remain harder to translate accurately.
-
Dynamic Meta-Metrics: Source-Sentence Conditioned Weighting for MT Evaluation
Dynamic Meta-Metrics learns source-sentence conditioned combinations of MT metrics, with MLP-based and soft-conditioned versions showing gains over linear and GP ensembles on WMT data.
-
BLOOM: A 176B-Parameter Open-Access Multilingual Language Model
BLOOM is a 176B-parameter open-access multilingual language model trained on the ROOTS corpus that achieves competitive performance on benchmarks, with improved results after multitask prompted finetuning.
-
No Language Left Behind: Scaling Human-Centered Machine Translation
A sparsely gated mixture-of-experts model trained on newly mined low-resource data achieves 44% relative BLEU improvement across 200 languages while adding human safety evaluation.
-
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.
-
LLM-FACETS: A Privacy-Preserving Framework for Evaluating LLM Transparency and Accountability
Introduces LLM-FACETS, a privacy-preserving open-source framework for LLM evaluation using deterministic metrics locally, LLM-judge metrics with user-controlled APIs, and mechanisms for uncertainty visualization and hallucination detection.
-
The Impact of Vocabulary Overlaps on Knowledge Transfer in Multilingual Machine Translation
Experiments show domain match and language relatedness drive knowledge transfer in multilingual MT more than vocabulary overlap.
-
Analyzing the Effect of Noise in LLM Fine-tuning
Label noise hurts fine-tuning performance most while grammatical and typographical noise sometimes act as mild regularizers, with changes concentrated in task-specific layers.
-
Syntax as a Rosetta Stone: Universal Dependencies for In-Context Coptic Translation
HAMR combines bi-level meta-learned instance reweighting with KNN-based neighborhood resampling to improve NLP performance under class imbalance, evaluated on six NER and classification benchmarks.
-
Making Knowledge Accessible: Divergent Readability-Accuracy Strategies of Mistral and QWen in Biomedical Text Simplification
Mistral uses careful lexical simplification to raise readability while keeping BERTScore at 0.91 comparable to humans, whereas QWen improves readability but shows a disconnect with its 0.89 BERTScore in biomedical text simplification.
-
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.
-
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.
-
Lius: Translation Model Based Instructional Lingustic Using Continual Instruction Tuning In Kupang Malay
Lius improves LLM translation for Kupang Malay by 4-13 points over baselines via continual instruction tuning with dictionary-derived instructions.
-
Bridging the Linguistic Divide: A Survey on Leveraging Large Language Models for Machine Translation
A literature survey that organizes prompting, fine-tuning, preference optimization, and context-aware techniques for LLM-based machine translation with emphasis on low-resource languages.