John Carreyrou and other authors bring new lawsuit against six major AI companies
These authors rejected Anthropic's class action settlement, arguing that "LLM companies should not be able to so easily extinguish thousands upon thousands of high-value claims at bargain-basement rates."
This article is divided into two parts; they are: • What Is Perplexity and How to Compute It • Evaluate the Perplexity of a Language Model with HellaSwag Dataset Perplexity is a measure of how well a language model predicts a sample of text.
Lemon Slice nabs $10.5M from YC and Matrix to build out its digital avatar tech
Digital avatar generation company Lemon Slice is working to add a video layer to AI chatbots with a new diffusion model that can create digital avatars from a single image.
Understanding the process behind agentic planning and task management in LangChain
The post How Agents Plan Tasks with To-Do Lists appeared first on Towards Data Science.
Stop Retraining Blindly: Use PSI to Build a Smarter Monitoring Pipeline
A data scientist's guide to population stability index (PSI)
The post Stop Retraining Blindly: Use PSI to Build a Smarter Monitoring Pipeline appeared first on Towards Data Science.
Google Code Wiki: Live Docs, Diagrams & Chat for Any GitHub Repo
Coding experts tend to use 30-40% of their time only for comprehending the already existing code. These are two entire working days every week that are wasted on going through obsolete documentation, understanding ambiguous code, and desperately searching for developers who quit months ago. On the ...
Synergy in Clicks: Harsanyi Dividends for E-Commerce
A brief overview of the math behind the Harsanyi Dividend and a real-world application in Streamlit
The post Synergy in Clicks: Harsanyi Dividends for E-Commerce appeared first on Towards Data Science.
How social media encourages the worst of AI boosterism
Demis Hassabis, CEO of Google DeepMind, summed it up in three words: “This is embarrassing.” Hassabis was replying on X to an overexcited post by Sébastien Bubeck, a research scientist at the rival firm OpenAI, announcing that two mathematicians had used OpenAI’s latest large language model, GPT-5...
Comparative Evaluation of Explainable Machine Learning Versus Linear Regression for Predicting County-Level Lung Cancer Mortality Rate in the United States
arXiv:2512.17934v1 Announce Type: new
Abstract: Lung cancer (LC) is a leading cause of cancer-related mortality in the United States. Accurate prediction of LC mortality rates is crucial for guiding targeted interventions and addressing health disparities. Although traditional regression-based mode...
What's the Price of Monotonicity? A Multi-Dataset Benchmark of Monotone-Constrained Gradient Boosting for Credit PD
arXiv:2512.17945v1 Announce Type: new
Abstract: Financial institutions face a trade-off between predictive accuracy and interpretability when deploying machine learning models for credit risk. Monotonicity constraints align model behavior with domain knowledge, but their performance cost - the pric...
Convolutional-neural-operator-based transfer learning for solving PDEs
arXiv:2512.17969v1 Announce Type: new
Abstract: Convolutional neural operator is a CNN-based architecture recently proposed to enforce structure-preserving continuous-discrete equivalence and enable the genuine, alias-free learning of solution operators of PDEs. This neural operator was demonstrate...
CodeGEMM: A Codebook-Centric Approach to Efficient GEMM in Quantized LLMs
arXiv:2512.17970v1 Announce Type: new
Abstract: Weight-only quantization is widely used to mitigate the memory-bound nature of LLM inference. Codebook-based methods extend this trend by achieving strong accuracy in the extremely low-bit regime (e.g., 2-bit). However, current kernels rely on dequant...
Parameter-Efficient Fine-Tuning for HAR: Integrating LoRA and QLoRA into Transformer Models
arXiv:2512.17983v1 Announce Type: new
Abstract: Human Activity Recognition is a foundational task in pervasive computing. While recent advances in self-supervised learning and transformer-based architectures have significantly improved HAR performance, adapting large pretrained models to new domain...
The Machine Learning “Advent Calendar” Day 22: Embeddings in Excel
Understanding text embeddings through simple models and Excel
The post The Machine Learning “Advent Calendar” Day 22: Embeddings in Excel appeared first on Towards Data Science.
OpenAI says AI browsers may always be vulnerable to prompt injection attacks
OpenAI says prompt injections will always be a risk for AI browsers with agentic capabilities, like Atlas. But the firm is beefing up its cybersecurity with an 'LLM-based automated attacker.'
ChatLLM Presents a Streamlined Solution to Addressing the Real Bottleneck in AI
For the last couple of years, a lot of the conversation around AI has revolved around a single, deceptively simple question: Which model is the best? But the next question was always, the best for what? The best for reasoning? Writing? Coding? Or maybe it’s the best for images, audio, or video? Tha...
AI Quantum Intelligence & Pic of the week (2025&12&19)
Visualizing the "AI Orchestrator" concept of 2025. This image depicts the synergy between human creativity and autonomous AI agents. It highlights key 2025 trends including agentic workflows, multimodal interfaces, and sustainable AI energy consumption in a futuristic, biophilic office setting.
The AI data center build-out, as it currently stands, is dependent on two things: Nvidia chips and borrowed money. Perhaps it was inevitable that people would begin using Nvidia chips to borrow money. As the craze has gone on, I have begun to worry about the weaknesses of the AI data center boom; lo...
What Happens When You Build an LLM Using Only 1s and 0s
An LLM that's 41× more efficient and 9× faster than today's standard models
The post What Happens When You Build an LLM Using Only 1s and 0s appeared first on Towards Data Science.