Author: Kalev Leetaru
Large-Token LLMs: Leveraging Anthropic's Claude 2's 100K Token Limit To Summarize An Entire ABC Evening News TV News Broadcast In A Single Prompt
One of the greatest limitations of current LLMs, besides hallucination and instability, is their severe input limits: most common production…
Experiments With Anthropic's Claude 2 For Summarization, Event & Relation Extraction, NER & Q&A
Continuing our explorations of how various commercial LLMs perform on the chaotic conflicting cacophony that is global news, what does…
Experiments With Google's PaLM 2 LLM's For Embedding, Summarization, Event & Relation Extraction, NER & Q&A: Bison & Gecko
What does it look like to summarize a television evening news broadcast using Google's PaLM 2 large language model (LLM)?…
Multimodal LLMs: What Bard Teaches Us About Consumer Versus Enterprise Applications & Imagery
Last week we explored some of the significant limitations of state of the art multimodal research LLM's when applied to…
Using Embeddings To Rank Clinical, "Creative" & "Inspired Fiction" Summaries Of An Evening News Broadcast
Earlier this month we demonstrated the use of embeddings to combat hallucination in LLM summarization and as a form of…
AGI & LLM Reasoning: Why Benchmarks Should Require 3-5 Runs To Reduce Anthropomorphization & False AGI Claims
Earlier today we examined how a SOTA multimodal LLM describes a variety of news and other images, testing how well…
The Limitations Of Multimodal Large Language Models: Automated Image Description, Captioning & Reasoning
Multimodal Large Language Models (LLMs) are touted as the future of automated reasoning, with the ability to look across imagery…
ChatGPT's Puzzling Inability To Solve A Simple Physics Riddle & The Brittleness Of Reasoning
Interviewers, especially at technology companies, have historically often turned to thought puzzles to understand how a candidate reasons under uncertainty….
WashPost: Murdoch Is Realizing That He's Stuck With The Monster He Created
An article in today's Washington Post by Philip Bump includes an analysis of media coverage using the TV Explorer. Read…
Complex & Ambiguous Phrasing, Re-Reading & LLMs' Inability To Conceptually Reason
Conceptually, LLM reasoning is a linear process, "confined to token-level, left-to-right decision-making processes during inference." This means that when confronted…
Fox News Covered Mulvaney Story 3x More Than CNN & MSNBC Combined
The timeline below shows mentions of Dylan Mulvaney across CNN, MSNBC and Fox News since April of this year, showing…
FiveThirtyEight: Which 2024 Candidates Had The Best – And Worst – Campaign Launches?
An analysis in FiveThirtyEight today includes a mention of television news coverage of Tim Scott's campaign. Read The Full Article.
When Light Is Dark, Heavy Is Light & Expensive Is Cheap: The Challenges Of Generative Search & LLM Reasoning
As we continue to assess the state of generative search, here are a few interesting examples from a major commercial…
The Era Of "Bombshells" On Television News: 2015-2020
The timeline below shows total mentions of the word "bombshell(s)" across television news over the past decade, showing a marked…
Fox News Covers Pandemic "Lab Leak" Theory 2.7x As Much As CNN & MSNBC Combined
The timeline below shows discussion of the Covid-19 "lab leak" theory across television news since the start of the pandemic:…
EUobserver: How Kremlin Promotes Orban As The Voice Of 'Real' Europe
Viktor Orban and other Hungarian politicians frequently appear on Russian TV screens and are featured prominently in state media, where…
Visual Explorer: Historical Backfiles For Al Maghribia, TVR Info And TV3 Now Available
In collaboration with the Internet Archive's Television News Archive, the complete TV News Archive historical backfiles for Morocco's Al Maghribia,…
Russian Television Covering US Cluster Bomb Package To Ukraine
Russian television, such as this sample broadcast from Russia Today, emphasizes the US' apparent about-face on the use of cluster…
How Television News Has Covered Cluster Bombs As War Crimes
The timeline below shows mentions of cluster munitions as illegal or war crimes across a handful of television news channels:…
OpenAI & ChatGPT Are Fading In The News: Media Shifts From Posterchild "OpenAI" To "Generative AI"
A deep dive into media coverage of the emerging world of generative AI suggests a pivot from OpenAI as the…
Twitter's Television News 'Trump Bump' Has Faded: Fewer Mentions Than A Decade Ago
Donald Trump led to a sustained rise in mentions of Twitter across television news which decreased steadily during his presidency….
Append-Based Keyword Search Versus Rebuild-From-Scratch Embedding Databases: Technical & Methodological Challenges
One of the most underappreciated challenges of large-scale embedding database implementation in the enterprise is their fundamental difference from keyword-based…
Using Embedding Ranking To Sort Generative Summaries By "Quality"
Earlier today we demonstrated how scoring LLM-based generative summaries by embedding-based similarity ranking to the original source material can filter…
Do LLM's Truly "Create" Or Merely "Arrange": Just How Much Of An LLM's Writing Is Original?
Continuing our exploration of generative plagiarism, just how much of the text created by a generative LLM is truly novel?…