Models Research preprint
Black-Box Uncertainty Quantification for Large Language Models via Ensemble-of-Ensembles (opens the original at arxiv.org in a new tab)
The authors introduce a two-level ensemble that bridges this gap.
Explore the frontier
Tracking the Evolution of Intelligence
Artificial Intelligence, AGI & Superintelligence News, Research, and Industry Intelligence.
14 stories in Models.
Models Research preprint
The authors introduce a two-level ensemble that bridges this gap.
Models Research preprint
The authors introduce LitArc-Bench to separate these explanations: 39 Chinese-language literary excerpts segmented into seven checkpoints, with 430 questionnaires from 81 adults in two classroom groups (9,020 VAD ratings).
Models Research preprint
The authors introduce Gamer, a generalist agent framework built for long-horizon tasks in complex interactive environments.
Researchers at Keele University have developed a physics-inspired AI model that achieved high accuracy in tests distinguishing genuine facial images from AI-generated or manipulated images.
Large language models (LLMs), the AI systems underpinning the functioning of ChatGPT, Gemini and other similar platforms, are now widely used worldwide. While the responses provided by these models can be useful in many situations, they can also change under specific conditions.
Models Research preprint
The authors introduce LightTAR, a simple, inexpensive method that adapts dimensionality reduction to the prediction task.
Models Research preprint
Small open-weight language models can run crowd decision-making (CDM) on an institution's own infrastructure, which keeps the checkpoint, the inference settings and the opinions under its control.
Large language models have transformed natural language processing through their ability to generate and interpret humanlike text. Most modern language models rely on transformer architectures, whose self-attention mechanisms provide powerful capabilities for learning relationships within long sequences. However, training and running these models can require substantial computational resources and energy, creating challenges for their wider use in embedded systems, edge devices and other environments where computing resources are limited.
Models Research preprint
The authors present Fikiria, a two-stage SFT+GRPO recipe for Swahili mathematical reasoning.
Models Research preprint
On-policy self-distillation (OPSD) trains a student on next-token targets from a teacher, a frozen copy of the student that additionally receives privileged information, such as a worked solution, hidden from the student.
Models Research preprint
Listeners can judge an agent's speech on how human it sounds (human-likeness) and on how much they would want to talk to it (conversational preference).
Models Research preprint
Many advances in ML systems came from breaking an assumption that earlier methods took for granted: speculative decoding dropped the assumption that a model needs one forward pass per token.
Models Research preprint
Across 4 experiments, the authors find people learn to trust helpful models more than unhelpful ones (67% vs 21% of the time), but only when they receive external feedback on the models' accuracy.
Models Research preprint
The authors study diversity in two educational generation tasks: math items and essay feedback.
Try a different search term or category, or browse every live story.
Show all stories