# Dec 7, 2025

## Google Titans architecture, helping AI have long-term memory

- The **Titans architecture** and **MIRAS framework** enable AI models to efficiently manage long-term memory and adapt in real-time, enhancing their ability to process extensive contexts without offline retraining.

## Nested Learning: A new ML paradigm for continual learning

- **Nested Learning** introduces a paradigm that treats machine learning models as interconnected, multi-level optimization problems, effectively addressing **catastrophic forgetting** by allowing simultaneous optimization of various learning tasks.

## Z-Image: Powerful and highly efficient image generation model with 6B parameters

- **Z-Image** is a **6B parameter** image generation model that includes **Z-Image-Turbo**, a distilled variant achieving **sub-second inference** with only **8 NFEs**, making it highly efficient for photorealistic image generation and bilingual text rendering.

## At least 50 hallucinated citations found in ICLR 2026 submissions

- **GPTZero identified over 50 hallucinations** in ICLR 2026 submissions, revealing that these errors were overlooked by 3-5 peer reviewers, raising concerns about the integrity of the peer review process.

## The unexpected effectiveness of one-shot decompilation with Claude

- **One-shot decompilation with Claude has accelerated progress on Snowboard Kids 2**, achieving more in three weeks than in the previous three months, thanks to a workflow that minimizes human intervention and maximizes throughput.

## Top ICLR 2026 Papers Found with fake Citations — Even Reviewers Missed Them

- **Fifty hallucinations** were identified in ICLR 2026 submissions, revealing that even **top-tier papers** with high scores contained **fake citations** overlooked by multiple reviewers.

## Touching the Elephant – TPUs

- **Google's Tensor Processing Unit (TPU)** has evolved from a research project into a powerful hardware accelerator, achieving **42.5 Exaflops** with the latest generation, Ironwood, which features **9,216 chips in a pod** and a focus on specialized computations for neural networks.

## Zebra-Llama: Towards Efficient Hybrid Models

- **Zebra-Llama** introduces a **scalable hybrid model** architecture that combines **State Space Models (SSMs)** and **Multi-head Latent Attention (MLA)** layers, achieving **Transformer-level accuracy** with significantly reduced training requirements of only **7-11B tokens**.

## From DeepSeek V3 to V3.2

- **DeepSeek V3.2** introduces a **sparse attention mechanism** (DeepSeek Sparse Attention) that enhances efficiency in long-context scenarios, achieving competitive performance against proprietary models like GPT-5 and Gemini 3.0 Pro.

## Research ARC Prize 2025 Results and Analysis

- **ARC Prize 2025** highlights the ongoing challenge of achieving **AGI**, with no Grand Prize winner yet, despite significant advancements in model refinement and open-source contributions from 1,455 teams and 90 papers submitted.

## 96.1M Rows of iNaturalist Research-Grade plant images (with species names)

- **96.1M rows** of cleaned iNaturalist plant images, complete with species names and coordinates, are now available for testing **vision models** on real-world noisy data, addressing the challenges of using GBIF data.