In recent years, the input context sizes of large language models (LLMs) have increased dramatically. However, existing evaluation methods have not kept pace, failing to comprehensively assess the efficiency of models in handling long contexts. To bridge this gap, we introduce the BABILong benchmark, designed to test language models' ability to reason across facts distributed in extremely long documents. BABILong includes a diverse set of 20 reasoning tasks, including fact chaining, simple induction, deduction, counting, and handling lists/sets. These tasks are challenging on their own, and even more demanding when the required facts are scattered across long natural text. Our evaluations show that popular LLMs effectively utilize only 10-20% of the context and their performance declines sharply with increased reasoning complexity. Among alternatives to in-context reasoning, Retrieval-Augmented Generation methods achieve a modest 60% accuracy on single-fact question answering, independent of context length. Among context extension methods, the highest performance is demonstrated by recurrent memory transformers after fine-tuning, enabling the processing of lengths up to 50 million tokens. The BABILong benchmark is extendable to any length to support the evaluation of new upcoming models with increased capabilities, and we provide splits up to 10 million token lengths.

In recent years, the input context sizes of large language models (LLMs) have increased dramatically. However, existing evaluation methods have not kept pace, failing to comprehensively assess the efficiency of models in handling long contexts. To bridge this gap, we introduce the BABILong benchmark, designed to test language models' ability to reason across facts distributed in extremely long documents. BABILong includes a diverse set of 20 reasoning tasks, including fact chaining, simple induction, deduction, counting, and handling lists/sets. These tasks are challenging on their own, and even more demanding when the required facts are scattered across long natural text.

To operate efficiently in urban environments, mobile robots and other autonomous systems should be able to move safely on sidewalks and avoid collisions with pedestrians or other obstacles. This is particularly true for delivery robots or systems that are specifically programmed to patrol urban environments. Researchers at Georgia Institute of Technology and Stanford University have recently developed AlienGo, a quadruped robot that can follow specific routes generated by public map services while remaining on sidewalks and avoiding collisions with obstacles or humans. This robot, presented in a paper pre-published on arXiv, is based on a new, highly performing two-staged learning framework for safe sidewalk navigation. "As part of this project, we developed an intelligent quadrupedal robot that can navigate sidewalks in the real world," Sehoon Ha, one of the researchers who carried out the study, told TechXplore.

Infosys is a global leader in consulting, information technology, outsourcing and next-generation services with clients in more than 50 countries. With 9.02 billion in Q2 FY16 revenues and more than 193,000 employees, the Indian multinational is helping enterprises redefine their present and future in a world where innovative solutions in mobility, sustainability, big data and cloud computing are required. Founded in 1981 by seven engineers with 250, Infosys is the second-largest Indian IT services company by 2016 revenues and was the fifth largest employer of H-1B visa professionals in the US in 2013. America is also home to its Global Head of Digital, Scott Sorokin, who has been a strategist and digital partner for senior-level executives at Fortune 100 companies for over 25 years. Formerly the the chief strategy officer at Publicis.Sapient/Razorfish and Rosetta, the New York-based Sorokin combines CXO-level business strategy, technology and marketing experience in a fast-changing global market to spur digital innovation at Infosys.