Imagen de Portada

When Life Gives You Samples: The Benefits of Scaling up Inference Compute for Multilingual LLMs

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Khairi, Ammar, D'souza, Daniel, Shen, Ye, Kreutzer, Julia, Hooker, Sara
Formato:	Preprint
Publicado:	2025
Materias:	Computation and Language Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Making, not Taking, the Best of N
por: Khairi, Ammar, et al.
Publicado: (2025)

Treasure Hunt: Real-time Targeting of the Long Tail using Training-Time Markers
por: D'souza, Daniel, et al.
Publicado: (2025)

Multilingual Arbitrage: Optimizing Data Pools to Accelerate Multilingual Progress
por: Odumakinde, Ayomide, et al.
Publicado: (2024)

Tiny Aya: Bridging Scale and Multilingual Depth
por: Salamanca, Alejandro R., et al.
Publicado: (2026)

RLHF Can Speak Many Languages: Unlocking Multilingual Preference Optimization for LLMs
por: Dang, John, et al.
Publicado: (2024)

The Art of Asking: Multilingual Prompt Optimization for Synthetic Data
por: Mora, David, et al.
Publicado: (2025)

Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model
por: Üstün, Ahmet, et al.
Publicado: (2024)

The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm
por: Aakanksha, et al.
Publicado: (2024)

LLM See, LLM Do: Guiding Data Generation to Target Non-Differentiable Objectives
por: Shimabucoro, Luísa, et al.
Publicado: (2024)

On the Limitations of Compute Thresholds as a Governance Strategy
por: Hooker, Sara
Publicado: (2024)

How Does Quantization Affect Multilingual LLMs?
por: Marchisio, Kelly, et al.
Publicado: (2024)

Critical Learning Periods: Leveraging Early Training Dynamics for Efficient Data Pruning
por: Chimoto, Everlyn Asiko, et al.
Publicado: (2024)

Déjà Vu: Multilingual LLM Evaluation through the Lens of Machine Translation Evaluation
por: Kreutzer, Julia, et al.
Publicado: (2025)

The State of Multilingual LLM Safety Research: From Measuring the Language Gap to Mitigating It
por: Yong, Zheng-Xin, et al.
Publicado: (2025)

The Multilingual Divide and Its Impact on Global AI Safety
por: Peppin, Aidan, et al.
Publicado: (2025)

Multi-Agent Sampling: Scaling Inference Compute for Data Synthesis with Tree Search-Based Agentic Collaboration
por: Ye, Hai, et al.
Publicado: (2024)

Scaling LLM Inference with Optimized Sample Compute Allocation
por: Zhang, Kexun, et al.
Publicado: (2024)

Synthetic Tabular Data Generation for Imbalanced Classification: The Surprising Effectiveness of an Overlap Class
por: D'souza, Annie, et al.
Publicado: (2024)

Novel Pathogenic Variant in Exon 31 of the TSC2 Gene Associated With a Severe Phenotype of Tuberous Sclerosis
por: Tabitha D'souza, et al.
Publicado: (2025)

Aya 23: Open Weight Releases to Further Multilingual Progress
por: Aryabumi, Viraat, et al.
Publicado: (2024)

Beyond Factual QA: Mentorship-Oriented Question Answering over Long-Form Multilingual Content
por: Bhalerao, Parth, et al.
Publicado: (2026)

LLMs Corrupt Your Documents When You Delegate
por: Laban, Philippe, et al.
Publicado: (2026)

Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning
por: Singh, Shivalika, et al.
Publicado: (2024)

From Unaligned to Aligned: Scaling Multilingual LLMs with Multi-Way Parallel Corpora
por: Shen, Yingli, et al.
Publicado: (2025)

LLMs Beyond English: Scaling the Multilingual Capability of LLMs with Cross-Lingual Feedback
por: Lai, Wen, et al.
Publicado: (2024)

Scale-free Characteristics of Multilingual Legal Texts and the Limitations of LLMs
por: Chen, Haoyang, et al.
Publicado: (2025)

LLMs vs Established Text Augmentation Techniques for Classification: When do the Benefits Outweight the Costs?
por: Cegin, Jan, et al.
Publicado: (2024)

When Benchmarks Leak: Inference-Time Decontamination for LLMs
por: Chai, Jianzhe, et al.
Publicado: (2026)

Investigating and Scaling up Code-Switching for Multilingual Language Model Pre-Training
por: Wang, Zhijun, et al.
Publicado: (2025)

Test-Time Scaling with Repeated Sampling Improves Multilingual Text Generation
por: Gupta, Ashim, et al.
Publicado: (2025)

Do LLMs Give Psychometrically Plausible Responses in Educational Assessments?
por: Säuberli, Andreas, et al.
Publicado: (2025)

Evaluating Language Model Context Windows: A "Working Memory" Test and Inference-time Correction
por: Dsouza, Amanda, et al.
Publicado: (2024)

When Does Monolingual Data Help Multilingual Translation: The Role of Domain and Model Scale
por: Baziotis, Christos, et al.
Publicado: (2023)

When Less Language is More: Language-Reasoning Disentanglement Makes LLMs Better Multilingual Reasoners
por: Zhao, Weixiang, et al.
Publicado: (2025)

Informed Routing in LLMs: Smarter Token-Level Computation for Faster Inference
por: Han, Chao, et al.
Publicado: (2025)

Expanding Computation Spaces of LLMs at Inference Time
por: Jang, Yoonna, et al.
Publicado: (2025)

Pruning Multilingual Large Language Models for Multilingual Inference
por: Kim, Hwichan, et al.
Publicado: (2024)

"Knowing When You Don't Know": A Multilingual Relevance Assessment Dataset for Robust Retrieval-Augmented Generation
por: Thakur, Nandan, et al.
Publicado: (2023)

Transforming mental health systems: The role of embedded researchers in advancing learning health systems
por: Miranda Field, et al.
Publicado: (2025)

When Scale Meets Diversity: Evaluating Language Models on Fine-Grained Multilingual Claim Verification
por: Shcharbakova, Hanna, et al.
Publicado: (2025)