Adapting chat models to support non-English languages for broader access.
― 7 min read
Computer Science - Computation and Language
RSSA new method enhances AI responses by learning from multiple feedback sources.
― 6 min read
A new approach enhances the performance of code generated by large language models.
― 7 min read
This study compares how language models express uncertainty compared to human communication.
― 5 min read
New system enhances academic searches using language models.
― 6 min read
A system to improve cultural representation in AI through simulated dialogues.
― 7 min read
New method combines Go-Explore with advanced models for better machine learning.
― 5 min read
OptLLM aids users in choosing the best LLMs for budgets and tasks.
― 6 min read
This research highlights the need for better evaluation of dialogue systems' use of conversation history.
― 5 min read
LangCell combines cell data and natural language for better cell analysis.
― 7 min read
A new method improves how systems handle errors in spoken language understanding.
― 6 min read
This study analyzes how language models recover from reasoning errors during tasks.
― 8 min read
This article discusses a flexible ranking method using multi-vector embeddings for better search results.
― 6 min read
A new method enhances model compression while maintaining accuracy.
― 5 min read
Exploring the link between AI attention heads and human memory processes.
― 6 min read
New methods improve multimodal models by adding questioning and evaluation tasks.
― 8 min read
A new method for combining training data enhances language model performance.
― 5 min read
Researchers use Brainscore to evaluate AI models against human brain activity.
― 5 min read
Examining fairness and ethics of how language models respond to users.
― 5 min read
This article discusses issues and best practices for evaluating language models.
― 7 min read
A new method for understanding preferences using feature-level comparisons.
― 4 min read
This article reviews the performance of language models in public health data analysis.
― 4 min read
A new way to improve transformer models using adaptable positional encoding techniques.
― 6 min read
I2CL improves efficiency and performance in language model tasks.
― 6 min read
AdvEval exposes weaknesses in Natural Language Generation evaluation metrics.
― 6 min read
Research presents STA-1 and STA-M methods for efficient watermarking of AI-generated text.
― 5 min read
A study introduces RepNoise to protect large language models from harmful fine-tuning.
― 6 min read
A new method boosts AI performance by effectively using all available experts.
― 4 min read
A new method combines technology for accurate carbon footprint calculation and understandable explanations.
― 7 min read
A framework for better detecting false claims in language models.
― 4 min read
A new method to enhance query rewriting without labeled data.
― 5 min read
New methods improve model performance during quantization in language models.
― 6 min read
Innovative method improves student assessment with fewer data requirements.
― 6 min read
Researchers investigate the challenges faced by LLMs with ambiguous questions.
― 5 min read
A new method to improve memory usage in language models.
― 5 min read
A two-stage active learning method enhances speech recognition accuracy with less data.
― 5 min read
A study using TDA reveals historical links between languages through word embeddings.
― 5 min read
ReAd enhances robot teamwork by refining planning and feedback mechanisms.
― 5 min read
New methods improve classification of legal texts by considering their temporal context.
― 8 min read
Exploring GPT-4's effects on student learning and engagement in coding classes.
― 5 min read