Hierarchical Merging is a technique commonly used to summarize very long texts ($>$100K tokens) by breaking down the input into smaller sections, summarizing those sections individually, and then merging or combining those summaries into a final coherent summary. Although it helps address the limitations of large language models (LLMs) with fixed input length constraints, the recursive merging process can amplify LLM hallucinations, increasing the risk of factual inaccuracies. In this paper, we seek to mitigate hallucinations by enriching hierarchical merging with context from the source document. Specifically, we propose different approaches to contextual augmentation ranging from \emph{replacing} intermediate summaries with relevant input context, to \emph{refining} them while using the context as supporting evidence, and \emph{aligning} them implicitly (via citations) to the input. Experimental results on datasets representing legal and narrative domains show that contextual augmentation consistently outperforms zero-shot and hierarchical merging baselines for the Llama 3.1 model family. Our analysis further reveals that refinement methods tend to perform best when paired with extractive summarization for identifying relevant input.

Deliver Us, a 2018 novel by Christopher Robinson and Gavin Kovite, takes a hilarious look at the future of drone delivery. The plot revolves around a social media activist named Piper Prince who attempts to stop Amazon from taking over her Detroit neighborhood. "It's written in a Coen brothers sort of tone," Robinson says in Episode 561 of the Geek's Guide to the Galaxy podcast. I wanted the world and the characters to be slightly pitched up from reality. So Jeff Bezos and his S-Team are characters in the book, and they are a little bit like the boardroom characters from The Hudsucker Proxy." Robinson sees Detroit as the perfect setting for a novel about the collision between social justice activism and breakneck technological disruption, given the city's rich history and uncertain future. "It's a place that was the arsenal of democracy," he says. "The Jetsons future is a future that was extrapolated from what Detroit used to be.

As an essential technology underpinning trusted computing, the trusted execution environment (TEE) allows one to launch computation tasks on both on- and off-premises data while assuring confidentiality and integrity. This article provides a systematic review and comparison of TEE-based secure computation protocols. We first propose a taxonomy that classifies secure computation protocols into three major categories, namely secure outsourced computation, secure distributed computation and secure multi-party computation. To enable a fair comparison of these protocols, we also present comprehensive assessment criteria with respect to four aspects: setting, methodology, security and performance. Based on these criteria, we review, discuss and compare the state-of-the-art TEE-based secure computation protocols for both general-purpose computation functions and special-purpose ones, such as privacy-preserving machine learning and encrypted database queries. To the best of our knowledge, this article is the first survey to review TEE-based secure computation protocols and the comprehensive comparison can serve as a guideline for selecting suitable protocols for deployment in practice. Finally, we also discuss several future research directions and challenges.

The next wave of technical innovation will be driven by businesses looking to provide more touchless experiences to their coronavirus-wary customers. If you had asked me a year ago where I thought the tech industry was headed, I probably would have answered that we are headed toward the age of "smart everything." Machine learning and artificial intelligence (AI) were really in vogue last year. It seemed that nearly every vendor was scrambling to include some sort of machine learning into their products. It reminded me of the way things were several years back when all the tech vendors were rushing to include cloud in their offerings.

TAMER has proven to be a powerful interactive reinforcement learning method for allowing ordinary people to teach and personalize autonomous agents' behavior by providing evaluative feedback. However, a TAMER agent planning with UCT---a Monte Carlo Tree Search strategy, can only update states along its path and might induce high learning cost especially for a physical robot. In this paper, we propose to drive the agent's exploration along the optimal path and reduce the learning cost by initializing the agent's reward function via inverse reinforcement learning from demonstration. We test our proposed method in the RL benchmark domain---Grid World---with different discounts on human reward. Our results show that learning from demonstration can allow a TAMER agent to learn a roughly optimal policy up to the deepest search and encourage the agent to explore along the optimal path. In addition, we find that learning from demonstration can improve the learning efficiency by reducing total feedback, the number of incorrect actions and increasing the ratio of correct actions to obtain an optimal policy, allowing a TAMER agent to converge faster.