Latest Articles for Language Models

Learn how to reduce BERT's size while maintaining performance through knowledge distillation.

A new method enhances attribution and correctness in language models' outputs.

A new approach to understanding dialogue agents through role-play and simulation.

This article analyzes GPT-4's abilities on abstract reasoning tasks and the impact of object representation.

A tool to assess large language models' multi-step reasoning capabilities.

This research shows how entailment and self-training improve language models without needing human-labeled data.

An in-depth analysis of ChatGPT's capabilities across various tasks and challenges.

This article explores how language models enhance AI's strategic reasoning in games.

Research examines syntax understanding in spoken language models using various methods.

Introducing TopEx, a fresh approach to understand language model differences.

Introducing a French model that outperforms leading benchmarks with less data.

Exploring methods to ensure personal information safety in language models.

A study on Auto-GPT performance in decision-making tasks.

LexGPT aims to assist legal professionals with understanding and generating legal text.

This paper explores how language models streamline project planning and execution.

This study highlights the need for better recognition of non-binary pronouns in language models.

A new method enhances reasoning accuracy in language models using structured prompts.

WOGLI focuses on word order impacts in German language inference.

PandaLM automates evaluation processes to improve large language models' instruction following.

ToolAlpaca aims to help smaller models effectively learn to use real-world tools.

Learn how RETA-LLM combines language models and retrieval systems for better answers.

This article discusses using SVG to improve how language models interpret images.

TrojLLM creates hidden prompts to manipulate large language model outputs.

A new model designed to analyze Romanian tweets using advanced technology.

Investigating prompt-based methods for improving language models in research data retrieval.

Larger language models may perform poorly on certain tasks, raising critical questions in AI research.

A new method enhances control over text generation in language models.

Strategies to boost ChatGPT's efficiency across various language tasks.

New dataset highlights AI performance in creative tasks with distractions.

A fresh approach to assess the quality of generated text in large language models.

Examining how AI handles human-like reasoning and its biases.

A new method enhances speech recognition models using only text data for adaptation.

A study on the effectiveness of language models for grammar correction in Brazilian Portuguese.

This article evaluates how language models reflect diverse global opinions.

A study on how well advanced models perform in Arabic language tasks.

Assessing large language models' performance in answering biomedical questions through BioASQ.

A study on assessing text generation quality from large language models.

Study shows how well models handle paraphrasing in textual entailment tasks.

A new benchmark aims to improve language models for social media communication.

BLUEX offers a rich resource to evaluate language models in Portuguese using entrance exam questions.