#LLMs

DeepSeek introduces Engram, enhancing LLM efficiency with a conditional memory axis.

Explore AI observability layers to enhance LLM performance and reliability.

'AutoCode teaches LLMs to author and verify contest-grade programming problems using a Validator–Generator–Checker(+Interactor) loop and dual verification, achieving near-judge consistency on held-out tasks.'

'Agentic AI and unified platforms are enabling faster, more personalized customer service at scale while requiring new infrastructure and careful human-AI balance.'

'Vibe coding lets LLMs generate pipeline code fast, but engineers must enforce idempotence, DAG discipline, and DQ checks before production.'

Discover how to use Mirascope to implement the Self-Refine technique with Large Language Models, enabling iterative improvement of AI-generated responses for enhanced accuracy.

SynPref-40M introduces a huge new preference dataset, enabling the Skywork-Reward-V2 family of models to achieve state-of-the-art results in human-AI alignment across multiple benchmarks.

OMEGA is a novel benchmark designed to probe the reasoning limits of large language models in mathematics, focusing on exploratory, compositional, and transformational generalization.

EPFL researchers have developed MEMOIR, a novel framework that enables continuous, reliable, and localized updates in large language models, outperforming existing methods in various benchmarks.

Internal Coherence Maximization (ICM) introduces a novel label-free, unsupervised training framework for large language models, achieving performance on par with human-supervised methods and enabling advanced capabilities without human feedback.

Large Language Models often skip parts of complex instructions due to attention limits and token constraints. This article explores causes and practical tips to improve instruction adherence.

New research from Microsoft and Salesforce shows that large language models experience a 39% performance drop when handling real multi-turn conversations with incomplete instructions, highlighting a key challenge in conversational AI.

RLV introduces a unified framework that integrates verification into value-free reinforcement learning for language models, significantly improving reasoning accuracy and computational efficiency on mathematical reasoning benchmarks.