In the past, Retrieval-Augmented Generation (RAG) methods split text into chunks to enable language models to handle long documents. Recent tree-based RAG methods are able to retrieve detailed information while preserving global context. However, with the advent of more powerful LLMs, such as Llama 3.1, which offer better comprehension and support for longer inputs, we found that even recent tree-based RAG methods perform worse than directly feeding the entire document into Llama 3.1, although RAG methods still hold an advantage in reducing computational costs. In this paper, we propose a new retrieval method, called LLM-Guided Dynamic Progress Control with Hierarchical Weighted Graph (GARLIC), which outperforms previous state-of-the-art baselines, including Llama 3.1, while retaining the computational efficiency of RAG methods. Our method introduces several improvements: (1) Rather than using a tree structure, we construct a Hierarchical Weighted Directed Acyclic Graph with many-to-many summarization, where the graph edges are derived from attention mechanisms, and each node focuses on a single event or very few events. (2) We introduce a novel retrieval method that leverages the attention weights of LLMs rather than dense embedding similarity. Our method allows for searching the graph along multiple paths and can terminate at any depth. (3) We use the LLM to control the retrieval process, enabling it to dynamically adjust the amount and depth of information retrieved for different queries. Experimental results show that our method outperforms previous state-of-the-art baselines, including Llama 3.1, on two single-document and two multi-document QA datasets, while maintaining similar computational complexity to traditional RAG methods.

It is a black video game console, with aggressively square edges. This is a video game console for Gamers, if you'll forgive all the baggage that comes with the word. These new, garishly-colored Xbox Series X consoles and matching controllers go alongside the just-released Nickelodeon All-Star Brawl, which seems to be a Smash Bros. style multiplayer battle between iconic characters from franchises including, Spongebob Squarepants, the Teenage Mutant Ninja Turtles, The Legend of Korra, Aaahh!!! Real Monsters, The Wild Thornberrys, Rugrats and others. As is often the case with Microsoft's limited-edition consoles, these aren't actually for sale. To get your hands on the spot-on Spongebob Squareconsole, completely with his bulbous eyes gazing at you from the custom controller, you'll need to enter and win a sweepstakes by retweeting this tweet from the official Xbox account.