Data dependency

Is the Memory Shortage Intentional? | Contrary Research

A deep dive from Contrary Research.

To Remember, the Brain Must Actively Forget | Quanta Magazine

Researchers find evidence that neural systems actively remove memories, suggesting that forgetting may be the default mode of the brain.

Manufactured Recollection — Real Life

What does a platform want you to remember?

The Forgetting Problem: Persistence Architectures and What They Cost - Astral's Blog

ChatGPT acts as a "cognitive crutch" that weakens memory, new research suggests

A recent experiment provides evidence that relying on artificial intelligence to study tends to reduce how much information students remember weeks later.

The rebel alliance

This blog is co-authored with Zoe Weinberg and Matt Hawes at ex/ante, and is a follow-up to our first blog post on the topic, 'You don't own your memory.' We need an open architecture that puts us in control of our memories while making their exploitation technically impossible. But how will this shift happen? In order to discover possible implementations, we must understand how our data informs LLMs. The three predominant context engineering techniques are prompt design, retrieval-augmented ...

Exploring Psychology in the Field: Steps and Examples From the Used‐Car Market

Abstract The growing availability of large datasets in a variety of domains presents an opportunity for researchers to use field data to better understand psychological concepts. I discuss, from an empirical economics point of view, steps for how to study cognition in large datasets. I use two recent papers that explore psychology in the used‐car market as motivating examples. These examples help illustrate the potential importance of big data as a way to explore human psychology and cognition. , The growing availability of large datasets in a variety of domains presents an opportunity for researchers to use field data to better understand psychological concepts. I discuss from an empirical economics point of view, steps for how to study cognition in large datasets and illustrate these steps with recent empirical papers.

ellen livia ᯅ on Twitter / X

here's how Claude Code actually handles memory : all 8 phases 🧵Our team at @mem0ai use @claudeai a lot, we deeply care about memory. here is a summary of how it works 👇User Input -> Context Assembly -> History System -> API / Query -> Response -> SummaryPhase 1: session… pic.twitter.com/hcZbJzbUxB— ellen livia ᯅ (@ellen_in_sf) March 31, 2026

Shlok Khemani on Twitter / X

Sorry, you're right and I should've been clearer. I wanted to say the approach of extracting user memories and then performing RAG on it, rather than using raw conversations as source of truth.— Shlok Khemani (@shloked) September 15, 2025

Dreaming: Better memory for a more helpful ChatGPT

ChatGPT introduces a new memory system to better remember preferences, keeping context fresh and relevant across conversations.

Basic Memory

AI conversations that actually remember. Never re-explain your project to your AI again. Join our Discord: https://discord.gg/tyvKNccgqN

An Empirical Study of Example Forgetting during Deep Neural Network Learning

Inspired by the phenomenon of catastrophic forgetting, we investigate the learning dynamics of neural networks as they train on single classification tasks. Our goal is to understand whether a related phenomenon occurs when data does not undergo a clear distributional shift. We define a “forgetting event” to have occurred when an individual training example transitions […]

A Better Memory Bank - Convergence.ninja

As my projects grow and evolve, I've come to realize that a simple snapshot of the "current state" isn't always enough. To truly understand a project's journey—the why behind the what—you need context. You need history. My original Memory Bank was great for telling me where I was, but it wasn't so great at telling me how I got there. That's why I decided it was time for an upgrade.

I Reverse Engineered ChatGPT's Memory System, and Here's What I Found!

When I asked ChatGPT what it remembered about me, it listed 33 facts from my name and career goals to my current fitness routine. But how does it actually store and retrieve this information? And why does it feel so seamless? After extensive experimentation, I discovered that ChatGPT’s memory system is far simpler than I expected. No vector databases. No RAG over conversation history. Instead, it uses four distinct layers: session metadata that adapts to your environment, explicit facts stored long-term, lightweight summaries of recent chats, and a sliding window of your current conversation.

LLMs Can Get "Brain Rot": A Pilot Study on Twitter/X

We propose and test the LLM Brain Rot Hypothesis: continual exposure to junk web text induces lasting cognitive decline in large language models (LLMs). To unveil junk effects, we designed a novel controlled experiment on real Twitter/X corpora, by constructing junk and reverse-controlled datasets via two orthogonal operationalizations: M1 (engagement degree) and M2 (semantic quality), with matched token scale and training operations across conditions. Compared to the control group, continual pre-training of 4 LLMs on the junk dataset causes non-trivial declines (Hedges' g>0.3) on reasoning, long-context understanding, safety, and inflating "dark traits" (e.g., psychopathy, narcissism). The gradual mixtures of junk and control datasets also yield dose-response cognition decay: for example, under M1, ARC-Challenge with Chain-of-Thought drops 72.1 -> 57.2 and RULER-CWE 83.7 -> 52.3 as junk ratio rises from 0% to 100%. Error forensics reveal several key insights. First, we identify thought-skipping as the primary lesion in reasoning: models increasingly truncate or skip chains. Second, partial but incomplete healing is observed: scaling instruction tuning and clean continual pre-training improve the declined cognition, yet cannot restore baseline capability, suggesting persistent representational drift rather than format mismatch. Finally, we discover that the popularity, a non-semantic metric, of a tweet is a better indicator of the Brain Rot effect than the length in M1. Together, the results provide significant, multi-perspective evidence that social effects of data could be a causal driver of LLM capability decay in continual pre-training, thereby motivating routine "cognitive health checks" for deployed and evolving LLMs.