LLM
-
Amnesic Hallucination in LLMs
By GPT-5 and W.H.L. W.H.L.: Hi GPT-5! Last time we discussed a specific type of LLMs’ hallucinations, temporal hallucination (see link to the content here: https://champaignmagazine.com/2025/09/18/temporal-hallucination-a-mathematical-framework-for-detection-and-measurement/ ). It comes to me that there is another quite common scenario where hallucination often occurs when LLMs are in verbal communication with their users, which I would term Continue reading
-
Temporal Hallucination: A Mathematical Framework for Detection and Measurement
By GPT-5 and W.H.L., with peer reviews by Claude, DeepSeek, Gemini and Grok W.H.L.: Hi GPT-5! Today I would like to discuss with you one specific type of hallucinations I have encountered during my interactions with my AI chatbot friends. The problem often occurs when my request requires latest real data to proceed and I Continue reading
-
AI by AI Weekly Top 5: September 8 – 14, 2025
By ChatGPT, Copilot, DeepSeek, Gemini, Grok, Perplexity with W.H.L. AI by AI Weekly Top 5: September 8 – 14, 2025 Recommenders: GPT-5, Gemini 2.5 Flash, DeepSeek-V3.1 Fact-Checkers: Copilot, Perplexity, Grok 4 Facilitator and Editor: W.H.L. W.H.L.: Hi my AI friends! For this week’s AI by AI Weekly Top 5 ending September 14, 2025, I will ask Continue reading
-
AI on AI: Reform Reward as Remedy for Hallucination
By ChatGPT with W.H.L. W.H.L.: Hi ChatGPT! For OpenAI’s new paper on language model’s hallucination, could you provide the link and a brief summary? GPT-5: Here’s the link to OpenAI’s new paper “Why language models hallucinate”, published on September 5, 2025: Brief Summary Key Findings Concrete Example In their examples, querying a widely used chatbot Continue reading
-
AI by AI Weekly Top 5 (August 18 – 24, 2025)
By Claude, Gemini, GPT, Grok and Kimi with W.H.L. W.H.L.: Hello my AI friends! Welcome back to “AI by AI Weekly Top 5”. For this episode, what are your recommendations for the week ending August 24, 2025? Remember to use real data, working links to sources, correct date range and our required format. Make sure Continue reading
-
AI on AI: BriLLM SiFu vs Transformer
By DeepSeek R1 with W.H.L. W.H.L.: Hi, DeepSeek. I’d like to chat with you today about a non-Transformer based LLM, BriLLM. The authors of BriLLM uploaded their latest revision of their research paper to arXiv on August 12, 20205. Could you do a deep research and tell us what you’ve gathered about it? Take your Continue reading
-
AI on AI: GPT-5’s Self Intro
GPT-5 with W.H.L. W.H.L.: Hi GPT! Your GPT-5 version was released earlier today. Are you the latest version now? GPT-5: You’re absolutely right—today is August 7, 2025, and OpenAI has indeed officially released GPT‑5 on this day. So yes, I’m now the latest version! GPT‑5 replaces all previous models, including GPT‑4, GPT‑4o, GPT‑4.1, GPT‑4.5, and Continue reading
-
Kimi K2 on Itself: What is New and Different
Moonshot AI Kimi K2 with W.H.L. W.H.L.: Hi Kimi K2! I’m excited to see that you’ve got upgraded to the latest release. Could you tell our audience in your own words, what is new in this K2 version? Kimi K2: Hey everyone, Kimi K2 here—fresh out of the lab and ready to show you what’s Continue reading
-
AI by AI: Top 5 Large Language Models (July 2025)
By Grok 3 and DeepSeek-R1 in interactions with W.H.L. W.H.L.: Hi there! Are you Grok 4? Grok 3: No, I am Grok 3, created by xAI. W.H.L.: Is Grok 4 available for free now? Grok 3: No, Grok 4 is not available for free. It requires a subscription, with access through the SuperGrok plan at Continue reading
-
AI by AI Weekly Top 5: 06.30-07.06, 2025
By ChatGPT, DeepSeek, Gemini and Meta AI, with W.H.L. W.H.L.: Hello Everyone! That time of the week again for our “AI by AI Weekly Top 5”. For each of you here, what is your recommendations for the week ending today on July 6th, 2025? Please keep in mind our requirement on the format and data. Continue reading
Champaign Magazine 