technocracy

DMCD: Semantic-Statistical Framework for Causal Discovery

⋅

arXiv:2602.20333v1 Announce Type: new Abstract: We present DMCD (DataMap Causal Discovery), a two-phase causal discovery framework that integrates LLM-based semantic drafting from variable metadata with statistical validation on observational data. In Phase I, a large language model proposes a sparse draft DAG, serving as a semantically informed prior over the space of possible causal structures. In Phase II, this draft is audited and refined via conditional independence testing, with detected discrepancies guiding targeted edge revisions. We evaluate our […]

Ver mais

Liked Liked
Part2: Guide to Huggingface AutoModels** for Vision and Audio

Part2: Guide to Huggingface AutoModels** for Vision and Audio

⋅

Part2: Guide to Hugging-face AutoModels** for Vision In the first part of this article, we explored NLP AutoModels — classification, question answering, text generation, and more. Now let’s move beyond text. Hugging Face Transformers also support vision, audio, and multimodal tasks using the same powerful AutoModel philosophy: Vision models work with pixels instead of words, but the idea remains the same: Pretrained on large image datasets Fine-tuned on your specific task Used for inference in production systems Before models see images, they need preprocessing. Image Preprocessing […]

Ver mais

Liked Liked
[R] P.R.I.M.E C-19: Solving Gradient Explosion on Circular Manifolds (Ring Buffers) using Fractional Kernels

[R] P.R.I.M.E C-19: Solving Gradient Explosion on Circular Manifolds (Ring Buffers) using Fractional Kernels

⋅

HI! I’ve been building a recurrent memory architecture that navigates a continuous 1D ring (pointer on a circular manifold), and hit a failure mode I think DNC / Pointer Network folks will recognize. How to imagine what im talking about: Problem: the “rubber wall” at the wrap seam If the pointer mixes across the boundary (e.g., N−1 → 0), linear interpolation makes the optimizer see a huge jump instead of a tiny step. The result is either frozen […]

Ver mais

Liked Liked

Robotic Assembly Using Deep Reinforcement Learning

⋅

Introduction Disclaimer: This article is a cross post from Pytorch Medium Blog Post. One of the most exciting advancements, that has pushed the frontier of the Artificial Intelligence (AI) in recent years, is Deep Reinforcement Learning (DRL). DRL belongs to the family of machine learning algorithms. It assumes that intelligent machines can learn from their actions similar to the way humans learn from experience. Over the recent years we could witness some impressive real-world applications of DRL. The […]

Ver mais

Liked Liked

Modeling Next-Token Prediction as Left-Nested Intuitionistic Implication

⋅

arXiv:2601.19915v1 Announce Type: new Abstract: We introduce the emph{Arrow Language Model}, a neural architecture derived from an intuitionistic-logic interpretation of next-token prediction. Instead of representing tokens as additive embeddings mixed by attention, we encode a prefix as a emph{left-nested implication chain} whose structure preserves order through non-commutative composition. Next-token prediction corresponds to emph{modus ponens}, and sequence processing becomes constructive proof extension under the Curry–Howard correspondence. Our Prolog-based specialized theorem provers validate fundamental properties of the neural models, among […]

Ver mais

Liked Liked
DeepSeek Sparse Attention: From O(L²) to Near-Linear O(L·n)

DeepSeek Sparse Attention: From O(L²) to Near-Linear O(L·n)

⋅

Large Language Models have a well-known scalability problem: attention does not scale. In standard Transformers, every token attends to every previous token. For a sequence of length L, this results in O(L²) compute and memory complexity. As context windows stretch from 8K → 32K → 128K tokens, attention becomes the dominant bottleneck — both economically and computationally. In their latest experimental model, DeepSeek-V3.2-Exp, the DeepSeek team introduces DeepSeek Sparse Attention (DSA): a fine-grained sparse attention mechanism designed to push long-context efficiency […]

Ver mais

Liked Liked

BrepCoder: A Unified Multimodal Large Language Model for Multi-task B-rep Reasoning

⋅

arXiv:2602.22284v1 Announce Type: new Abstract: Recent advancements in deep learning have actively addressed complex challenges within the Computer-Aided Design (CAD) domain.However, most existing approaches rely on task-specifi c models requiring structural modifi cations for new tasks, and they predominantly focus on point clouds or images rather than the industry-standard Boundary Representation (B-rep) format. To address these limitations, we propose BrepCoder, a unifi ed Multimodal Large Language Model (MLLM) that performs diverse CAD tasks from B-rep inputs. By leveraging […]

Ver mais

Liked Liked

Meta-Prompting: From “Using Prompts” to “Generating Prompts”

⋅

Most of us start prompt engineering the same way: we type an instruction and hope the model behaves. “Write a 300-word product blurb for wireless earbuds. Mention noise cancelling and battery life.” That’s basic prompting: human designs the prompt, model executes. Meta-prompting flips the workflow. You ask the model to become the prompt designer. “Create a prompt that will make an LLM write a 300-word earbuds blurb, highlighting noise cancelling and battery life, with a punchy CTA.” Now […]

Ver mais

Liked Liked

Hierarchical Molecular Representation Learning via Fragment-Based Self-Supervised Embedding Prediction

⋅

arXiv:2602.20344v1 Announce Type: new Abstract: Graph self-supervised learning (GSSL) has demonstrated strong potential for generating expressive graph embeddings without the need for human annotations, making it particularly valuable in domains with high labeling costs such as molecular graph analysis. However, existing GSSL methods mostly focus on node- or edge-level information, often ignoring chemically relevant substructures which strongly influence molecular properties. In this work, we propose Graph Semantic Predictive Network (GraSPNet), a hierarchical self-supervised framework that explicitly models both […]

Ver mais

Liked Liked

Theoretical Analysis of Contrastive Learning under Imbalanced Data: From Training Dynamics to a Pruning Solution

⋅

Contrastive learning has emerged as a powerful framework for learning generalizable representations, yet its theoretical understanding remains limited, particularly under imbalanced data distributions that are prevalent in real-world applications. Such an imbalance can degrade representation quality and induce biased model behavior, yet a rigorous characterization of these effects is lacking. In this work, we develop a theoretical framework to analyze the training dynamics of contrastive learning with Transformer-based encoders under imbalanced data. Our results reveal that neuron weights […]

Ver mais

Liked Liked

Load More

↑

Digitado © 2025