Abstract--Multi-hop question answering over knowledge graphs remains computationally challenging due to the combinatorial explosion of possible reasoning paths. Recent approaches rely on expensive Large Language Model (LLM) inference for both entity linking and path ranking, limiting their practical deployment. Additionally, LLM-generated answers often lack verifiable grounding in structured knowledge. We present two complementary hybrid algorithms that address both efficiency and verifiability: (1) LLM-Guided Planning that uses a single LLM call to predict relation sequences executed via breadth-first search, achieving near-perfect accuracy (micro-F1 > 0.90) while ensuring all answers are grounded in the knowledge graph, and (2) Embedding-Guided Neural Search that eliminates LLM calls entirely by fusing text and graph embeddings through a lightweight 6.7M-parameter edge scorer, achieving over 100 speedup with competitive accuracy. Through knowledge distillation, we compress planning capability into a 4B-parameter model that matches large-model performance at zero API cost. Evaluation on MetaQA demonstrates that grounded reasoning consistently outperforms ungrounded generation, with structured planning proving more transferable than direct answer generation. Our results show that verifiable multi-hop reasoning does not require massive models at inference time, but rather the right architectural inductive biases combining symbolic structure with learned representations. Knowledge graphs (KGs) have emerged as powerful structures for representing domain-specific, structured information that supports verifiable, multi-hop reasoning. Meanwhile, large language models (LLMs) trained on vast web-scale corpora have achieved impressive fluency and generalization across a wide range of tasks.

This paper reviews, analyzes, and proposes a new perspective on the bi-encoder architecture for neural search. While the bi-encoder architecture is widely used due to its simplicity and scalability at test time, it has some notable issues such as low performance on seen datasets and weak zero-shot performance on new datasets. In this paper, we analyze these issues and summarize two main critiques: the encoding information bottleneck problem and limitations of the basic assumption of embedding search. We then construct a thought experiment to logically analyze the encoding and searching operations and challenge the basic assumption of embedding search. Building on these observations, we propose a new perspective on the bi-encoder architecture called the \textit{encoding--searching separation} perspective, which conceptually and practically separates the encoding and searching operations. This new perspective is applied to explain the root cause of the identified issues and discuss ways to mitigate the problems. Finally, we discuss the implications of the ideas underlying the new perspective, the design surface that it exposes and the potential research directions arising from it.

If you spend more on coffee than IT security, you will be hacked! Warned U.S. Cybersecurity Czar Richard Clarke, speaking at RSA Conference. This quote would make a great bumper sticker if it weren't for network attacks. According to research by IBM, it takes 280 days to find and contain the average cyberattack, while the average attack costs $3.86 million. But what are network attacks, and how can we leverage a next-gen search tool like Jina to mitigate our exposure to the threat?

To understand the basics of neural search and how it differs from conventional search please go through my previous blog on "Next-gen powered by Jina". It explains how Jina- a cloud-native, open-source company is pioneering the field of neural search. It builds on the idea of semantic search and explains the basic building blocks of the Jina framework required to build intelligent search applications. Just as a recap the idea behind neural search is to leverage state-of-the-art deep neural networks to intelligently retrieve contextual and semantically relevant information from the heaps of data. A neural search system can go way beyond simple text search by allowing you to search through all the formats of data including images, videos, audios, and even PDFs.

TL;DR: Neural Search is a new approach to retrieving information using neural networks. Traditional techniques to search typically meant writing rules to "understand" the data being searched and return the best results. But with neural search, developers don't need to wrack their brains for these rules; The system learns the rules by itself and gets better as it goes along. Even developers who don't know machine learning can quickly build a search engine using open-source frameworks such as Jina. There is a massive amount of data on the web; how can we effectively search through it for relevant information?

A lot of ink has been spilled (or pixels illuminated) about the wonders of GPT-3, OpenAI's latest and greatest language model. A team of more than 30 OpenAI researchers have released a paper about GPT-3, a language model capable of achieving state-of-the-art results on a set of benchmark and unique natural language processing tasks that range from language translation to generating news articles to answering SAT questions. But like most examples spat out by language models, almost all of these were hand-selected by humans after many runs. Because not-so-good results just wouldn't make the news. Even bearing that in mind, I'm still blown away by what I've seen of GPT-3.