104 stories tagged Language Models · page 2 of 5
A new ArXiv paper proves that large language models (LLMs) have an inherent reliability ceiling that no amount of scaling can overcome. The study decomposes output uncertainty into a resolvable component (closable with more context) and a subjective component (inherent to task ambiguity), and shows that autoregressive generation further degrades this ceiling.
Researchers from ArXiv cs.CL have developed a system that automatically evolves prompt guidelines tailored to specific tasks, addressing the common problem of underspecified user queries. This could make large language models like ChatGPT and Claude more accurate and user-friendly without requiring prompt engineering expertise.
Researchers from ArXiv cs.CL discovered how language models distinguish between a character's belief and objective reality. The mechanism relies on a shared value slot that binds the attributed value and a router at the query position that selects which frame—belief or reality—to read from. This finding explains how AI handles nuanced, belief-based reasoning.
Bilibili has released a series of open small language models called Index-1.9B, trained on a massive dataset of Chinese and English tokens. These models are designed for various applications, from general use to specialized tasks like chat and character interaction.
Researchers introduced HALO, a method to enhance existing AI language models with minimal extra computation. It selectively refines responses, improving accuracy without wasting resources.
A new study reveals that AI language models often change African American English into Standard American English without users knowing. Researchers have developed a way to detect and reduce this bias in AI systems.
Researchers have introduced Wiola, a novel small language model architecture that breaks from existing designs. It incorporates five unique components aimed at improving efficiency and performance.
Researchers developed a method called CreativityNeuro to make AI models generate more diverse and creative responses. This could help AI avoid repetitive answers and offer more unique ideas.
Researchers created BayesBench to test how large language models update their beliefs with new evidence. The study reveals that AI models often struggle to adjust their reasoning as conversations progress.
Researchers created a dataset of 240 French doctor-patient conversations to train AI virtual patients. This could make medical training more accessible and effective for students worldwide.
A new paper highlights flaws in how we test AI models that handle text, images, and other inputs together. Current methods miss key aspects like understanding physical reality or combining different types of information. The authors propose better evaluation frameworks to address these gaps.
A new AI model called iLLaDA uses a different approach to understand language, potentially making it more efficient and accurate. This could lead to better AI assistants and tools that understand context more naturally.
Scientists developed a new method using Shapley values to measure how adjectives influence AI responses. Their findings reveal how small word choices can significantly impact AI performance and alignment.
A new paper finds that offline recommendations to mix AI models from different "families" for diversity may not hold up in real-time interactive settings—the very environments where multi-LLM systems are actually deployed. This discovery could change how multi-AI systems are designed.
Scientists discovered a gap in AI systems that combine logic solvers with language models. This gap can lead to incorrect answers despite using verified logic tools. Researchers propose solutions to maintain accuracy in these hybrid systems.
Researchers found that large language models don't transfer knowledge better between related languages like Arabic and Hebrew, regardless of model size. This challenges assumptions about linguistic similarity aiding AI performance. The study tested models with 4 billion to 671 billion parameters and various architectures.
Researchers have analyzed how Diffusion Language Models (DLMs) differ from traditional AI models. DLMs generate text by gradually refining entire sequences, potentially offering new advantages over current methods.
Researchers have developed Causal Attribution Pruning (CAP), a training-free method that prunes large language models by identifying critical attention heads through their causal impact on reasoning tasks. This reduces inference costs without sacrificing multi-step reasoning performance.
DeepSeek has released a preview of its V4 models, which can process up to 1 million tokens in context and introduce a new hybrid attention architecture. This breakthrough could make AI assistants much more useful for long documents and complex tasks. The models use new techniques to handle large amounts of text efficiently, making them faster and more capable than previous versions.
Researchers have developed a method called activation steering to improve synthetic data generation for languages with limited resources. This approach could make AI models more effective and affordable for less common languages.
Researchers have developed a method to help AI models retain detailed audio information, like tone and emotion, while processing speech. This could make voice assistants and transcription tools much more accurate and expressive.
Researchers have developed a new method called Telegraph English that compresses information for AI models while preserving meaning. This could make AI answers more accurate and efficient, especially for complex, multi-step questions.
Researchers introduced AdaMame, a new AI training approach that helps large language models reason better in multiple languages. This could make math-solving AI tools more reliable for non-English speakers.
Researchers have traced the internal pathways through which AI models process and combine visual and audio inputs to reach decisions. The findings could lead to more transparent and reliable AI assistants and creative tools.