Hacker News Top Stories with Summaries (June 15, 2023)
<style>
p {
font-size: 16px;
line-height: 1.6;
margin: 0;
padding: 10px;
}
h1 {
font-size: 24px;
font-weight: bold;
margin-top: 10px;
margin-bottom: 20px;
}
h2 {
font-size: 18px;
font-weight: bold;
margin-top: 10px;
margin-bottom: 5px;
}
ul {
padding-left: 20px;
}
li {
margin-bottom: 10px;
}
.summary {
margin-left: 20px;
margin-bottom: 20px;
}
</style>
<h1> Hacker News Top Stories</h1>
<p>Here are the top stories from Hacker News with summaries for June 15, 2023 :</p>
<div style="margin-bottom: 20px;">
<table cellpadding="0" cellspacing="0" border="0">
<tr>
<td style="padding-right: 10px;">
<div style="width: 200px; height: 100px; border-radius: 10px; overflow: hidden; background-image: url('https://substackcdn.com/image/fetch/w_1200,h_600,c_fill,f_jpg,q_auto:good,fl_progressive:steep,g_auto/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fbcf4d45c-57cd-4dca-b2d3-2f173af25df0_1600x1024.png'); background-size: cover; background-position: center;">
Native JSON Output from GPT-4
Summary: GPT-4 now supports native JSON output through function calling, simplifying the process of generating structured data. The new API feature is available for chat models gpt-3.5-turbo-0613 and gpt-4-0613. It introduces two new parameters in the Chat Completions API: functions and function_call. This allows for more reliable JSON responses and reduces the need for prompt engineering. The new API is expected to change the way developers interact with OpenAI LLMs, offering faster and cheaper API calls, less cognitive load on GPT, and easier conversion from natural language to structured data.
<div style="margin-bottom: 20px;">
<table cellpadding="0" cellspacing="0" border="0">
<tr>
<td style="padding-right: 10px;">
<div style="width: 200px; height: 100px; border-radius: 10px; overflow: hidden; background-image: url('https://hackernewstoemail.s3.us-east-2.amazonaws.com/hnd2'); background-size: cover; background-position: center;">
Gorilla: Large Language Model Connected with APIs
Summary: Gorilla is a Large Language Model (LLM) that excels in generating accurate API calls. Trained on Torch Hub, TensorFlow Hub, and HuggingFace datasets, it outperforms GPT-4, Chat-GPT, and Claude. Gorilla is highly reliable and reduces hallucination errors. The model is combined with a document retriever, enabling it to adapt to document changes and version updates. To assess Gorilla's capabilities, researchers introduced APIBench, a dataset comprising HuggingFace, TorchHub, and TensorHub APIs.