Latest AI News

Tools designed for rewriting, refactoring, and optimizing code should prioritize both speed and accuracy. Large language models (LLMs), however, often lack these critical attributes. Despite these limitations, there remains significant potential for leveraging LLMs to enhance code quality. In a new paper Don’t Transform the Code, Code the Transforms: Towards Precise Code Rewriting using LLMs, a Meta research team proposes a novel chain-of-thought strategy

Multimodal Large Language Models (MLLMs) have rapidly become a focal point in AI research. Closed-source models like GPT-4o, GPT-4V, Gemini-1.5, and Claude-3.5 exemplify the impressive capabilities of advanced multimodal understanding. This April, Apple introduced MM1, a suite of multimodal models up to 30 billion parameters, setting new benchmarks in multimodal performance with features like enhanced in-context learning and multi-image reasoning. These in

The rise of large language models (LLMs) has sparked questions about their computational abilities compared to traditional models. While recent research has shown that LLMs can simulate a universal Turing machine by using an external memory through prompting, this setup often relies on external controls, which shift much of the computational load away from the LLM itself. Until now, it has remained an open question whether an LLM, unaided by such external

For artificial intelligence to thrive in a complex, constantly evolving world, it must overcome significant challenges: limited data quality and scale, and a lag in new, relevant information creation. An intriguing question emerges—can language models autonomously generate new tasks to practice on, enabling them to self-improve and better align with human preferences? In a new paper Evolving Alignment via Asymmetric Self-Play, a research team from Google D

Recent large language models (LLMs) have shown impressive performance across a diverse array of tasks. However, their use in high-stakes or computationally constrained environments has highlighted the need for more interpretable and efficient alternatives. While traditional n-gram models are fully interpretable and much more computationally efficient, they typically fall short of LLMs in predictive accuracy, especially for tasks like next-token prediction.

Generative AI, including Language Models (LMs), holds the promise to reshape key sectors like education, healthcare, and law, which rely heavily on skilled professionals to navigate complex responsibilities. In education, for instance, effective teacher training with expert feedback is crucial yet costly, limiting opportunities to enhance educational quality on a larger scale. In a new paper Tutor CoPilot: A Human-AI Approach for Scaling Real-Time Expertis

Large Language Models (LLMs) have advanced considerably in generating and understanding text, and recent developments have extended these capabilities to multimodal LLMs that integrate both visual and audio data. Despite these gains, these models still face challenges with fine-grained cross-modal temporal reasoning, especially in aligning events across audio and video streams. To address this, an NVIDIA research team has introduced OMCAT: Omni Context Awa

In cognitive science, human thought processes are commonly divided into two systems: the fast, intuitive System 1 and the slower, analytical System 2. Recent research has shown that incorporating System 2-style processing into Transformers, including large language models (LLMs), can significantly improve their reasoning abilities. However, models that fully emulate System 2 thinking tend to demand substantial computational resources and exhibit slower res

The field of text-to-image synthesis has advanced rapidly, with state-of-the-art models now generating highly realistic and diverse images from text descriptions. This progress largely owes to diffusion-based architectures trained on vast datasets containing billions of image-text pairs. The capability to create high-resolution, photorealistic images from text has transformative potential across fields like content creation, gaming, synthetic data generati

Consistency models (CMs) are a cutting-edge class of diffusion-based generative models designed for rapid and efficient sampling. However, most existing CMs rely on discretized timesteps, which introduce complex hyperparameter tuning and are susceptible to discretization errors. While continuous-time approaches offer a promising alternative, their adoption has been hindered by significant training instabilities. To overcome these challenges, OpenAI researc

While large language models (LLMs) dominate the AI landscape, Small-scale Large Language Models (SLMs) are gaining traction as cost-effective and efficient alternatives for various applications. Despite their growing relevance, there is a significant gap in understanding the training behavior and computational requirements of SLMs, especially in contrast to their larger counterparts. In a new paper Computational Bottlenecks of Training Small-scale Large La

On the path to achieving artificial superhuman intelligence, a critical tipping point lies in a system’s ability to drive its own improvement independently, without relying on human-provided data, labels, or preferences. At present, most AI systems lack such recursive self-improvement capabilities, making this an ideal time to examine and define its characteristics and implications. In a new paper Boundless Socratic Learning with Language Games, researcher

SoniDo employs a generative architecture based on a multi-level transformer coupled with a hierarchical encoder. Through careful preprocessing, its intermediate representations are utilized as features for task-specific models across various music-related tasks, enhanced by data augmentation techniques. The model’s encoder design draws inspiration from Jukebox, but it distinguishes itself by incorporating a hierarchical structure. Using a framework called

Recent advancements in large language models (LLMs) have primarily focused on enhancing their capacity to predict text in a forward, time-linear manner. However, emerging research suggests that enabling LLMs to critique and refine their own outputs retrospectively can significantly improve their performance. While effective, existing methods rely on the advanced reasoning and instruction-following abilities inherent to high-capacity LLMs. Moreover, these a

Large Language Models (LLMs) have become indispensable tools for diverse natural language processing (NLP) tasks. Traditional LLMs operate at the token level, generating output one word or subword at a time. However, human cognition works on multiple levels of abstraction, enabling deeper analysis and creative reasoning. Addressing this gap, in a new paper Large Concept Models: Language Modeling in a Sentence Representation Space, a research team at Meta i

The Transformer architecture, introduced by Vaswani et al. in 2017, serves as the backbone of contemporary language models. Over the years, numerous modifications to this architecture have been proposed to enhance aspects such as training stability, inference efficiency, context length, and robustness. In a new paper nGPT: Normalized Transformer with Representation Learning on the Hypersphere, an NVIDIA research team proposes the normalized Transformer (nG

Recent advancements in training large multimodal models have been driven by efforts to eliminate modeling constraints and unify architectures across domains. Despite these strides, many existing models still rely on separately trained components such as modality-specific encoders and decoders. In a new paper JetFormer: An Autoregressive Generative Model of Raw Images and Text, a Google DeepMind research team introduces JetFormer, a groundbreaking autoregre

ASAL demonstrates its potential across various ALife substrates, including Boids, Particle Life, Game of Life, Lenia, and Neural Cellular Automata. By employing ASAL, the researchers uncovered previously unknown lifeforms and extended the frontier of emergent structures in ALife simulations. Beyond discovery, ASAL’s framework facilitates quantitative analysis of traditionally qualitative phenomena, mirroring human-like methods to measure complexity. Crucia

Graphics processing powerhouse Nvidia is set to further its foray into the robotics sector with the anticipated launch of its Jetson Thor compact computing platform in the first half of 2025. Specifically designed for humanoid robots, this move, initially reported by the Financial Times on January 1st, underscores Nvidia’s growing commitment to what many industry experts predict will be a multi-billion dollar market. Nvidia’s announcement comes amidst a fl

Beijing, China – April 15, 2025 – In a strategic move that underscores its technological prowess and global ambitions, potentially paving the way for a future IPO, Chinese AI company Zhipu.AI has announced the comprehensive open-sourcing of its next-generation General Language Models (GLM). This release includes the advanced GLM-4 series and the groundbreaking GLM-Z1 inference models, boasting unprecedented inference speeds and the launch of a dedicated in