Paper Review - CAASL: Amortized Active Causal Induction with Deep Reinforcement Learning
CAASL: Amortized Active Causal Induction with Deep Reinforcement Learning CAASL (Causal Amortized Active Structure Learning) presents a breakthrough approach...
Blog
Recent Posts
Paper Review - pixelNeRF: Neural Radiance Fields from One or Few Images
pixelNeRF presents a learning framework that enables predicting Neural Radiance Fields (NeRF) from just one or a few images in a feed-forward manner. This ap...
Paper Review - The Geometry of Categorical and Hierarchical Concepts in Large Language Models
In this work the authors attempt to extend the linear representation hypothesis to cases where the concepts no longer have natural binary contrasts and hence...
Paper Review - Targeted Cause Discovery with Data-Driven Learning
The paper introduces Targeted Cause Discovery with Data-Driven Learning (TCD-DL), a machine learning-based approach to identify all causal variables—direct a...
Paper Review - TabPFN: Understanding and Advancing Tabular Foundation Models
The Core Problem Traditional deep learning has revolutionized domains like computer vision and NLP, but tabular data remains dominated by classical approach...
Paper Review - SIREN: Implicit Neural Representations with Periodic Activation Functions
SIREN (Sinusoidal Representation Networks) introduced a groundbreaking approach for implicit neural representations using periodic activation functions. This...
Paper Review - Randomization Inference When N Equals One
Core Problem N-of-1 trials (where a single subject serves as both treatment and control over time) traditionally require long “washout” periods between trea...
Paper Review - PointNet and PointNet++
PointNet and its successor PointNet++ introduced groundbreaking approaches for directly processing point cloud data without intermediary representations, est...
Paper Review - OpenScene: 3D Scene Understanding with Open Vocabularies
OpenScene presents a breakthrough approach to 3D scene understanding that eliminates reliance on labeled 3D data and enables open-vocabulary querying. By co-...
Paper Review - One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion
One-2-3-45++ presents a breakthrough approach for transforming a single image into a high-quality 3D textured mesh in approximately one minute. This method b...
Paper Review - Neural Fields as Learnable Kernels for 3D Reconstruction
Neural Kernel Fields (NKF) introduced a novel approach to 3D reconstruction that bridges the gap between data-driven methods and traditional kernel technique...
Paper Review - Meta Statistical Learning
Meta-statistical learning is an innovative framework that employs neural networks to perform statistical inference tasks, such as parameter estimation and hy...
Paper Review - LION: Latent Point Diffusion Models for 3D Shape Generation
Overview This paper introduces LION (Latent Point Diffusion Model), a novel approach for 3D shape generation that combines variational autoencoders (VAEs) wi...
Paper Review - Instant Neural Graphics Primitives with a Multiresolution Hash Encoding
Müller et al. introduced a versatile input encoding for neural networks that dramatically accelerates the training and inference of neural graphics primitive...
Paper Review - DreamGaussian: Generative Gaussian Splatting for Efficient 3D Content Creation
DreamGaussian presents a novel framework for 3D content generation that achieves both efficiency and quality simultaneously. This approach addresses the slow...
Paper Review - DreamFusion: Text-to-3D using 2D Diffusion
DreamFusion represents a breakthrough in text-to-3D synthesis by leveraging pretrained 2D text-to-image diffusion models to generate high-quality 3D assets w...
Paper Review - A Primer on the Inner Workings of Transformer-Based Language Models
This work summarizes the current techniques used to interpret the inner workings of transformer language models. As such, summarizing a summary is a challeng...
Paper Review - 3D Gaussian Splatting for Real-Time Radiance Field Rendering
3D Gaussian Splatting introduces a groundbreaking approach to novel view synthesis that achieves both state-of-the-art quality and real-time rendering speeds...
Retrieval Head Mechanistically Explains Long-Context Factuality
In this paper the researcher are able to show that there exists a special kind of attention heads which are responsible for retrieval of information from lon...
Is DPO Superior to PPO for LLM Alignment? A Comprehensive Study
RLHF is the most used method to align LLMs with human preferences. RLHF methods can be roughly categorized as either reward-free or reward-based.
The Mathematics and Philosophy Behind MSE
When we perform linear regression, we’re making two fundamental modeling choices that are worth examining separately: Assuming our data follows a linear ...
Mixture-of-Depths: Dynamically allocating compute in transformer-based language models
Compute is a significant bottleneck for current AI model performance and scale ((cite)), making model compression increasingly important. In this paper, the ...
ShortGPT: Layers in Large Language Models are More Redundant Than You Expect
The field of artificial intelligence faces significant computational constraints, particularly in the deployment and training of Large Language Models (LLMs)...
Universal Language Learning Paragims - UL2
This paper presents a novel framework for model pre-training. To date, there seems to be no consensus as to what the optimal pre-training procedure should be...
On Writing to Think
Many successful intellectuals mention a consistent writing schedule[^1] as an integral part of their daily routine and their success. I always found this a b...
Effects of Scale on Model Finetuning
Finetuning is a type of [[knowledge editing for LLM]], it is required and widely adopted to unlock new and robust capabilities for creative tasks, get the mo...