PUBLISHED AI RESEARCH AND PROJECTS

AI research and projects

The research, projects, and explanations behind the people. Read the source, inspect verified code and models, and find a collaborator.

Clear all filters

A source check confirms that a source supports the summary. It is not an independent replication or endorsement. Our methodology ↗

Showing 65–80 of 149 proofs of work · Kind: research · Newest first
Page 5 of 10
researchChecked Sep 22, 2026

Scalable Diffusion Models with Transformers (DiT)

Dr. Saining Xie · New York University (NYU)

Authored the landmark ICCV 2023 paper proposing Diffusion Transformers (DiT), replacing traditional U-Nets with Transformer backbones operating on latent image patches, establishing power-law compute scaling for generative diffusion.

Scope & limitations

Full quadratic attention scaling over high-resolution token patches requires patchification or windowing to avoid compute explosion.

Context: Latent Diffusion, Vision Transformer backbones, adaptive layer norm (adaLN-Zero) conditioning.

View mission
researchChecked Sep 22, 2026

Small Language Model (SLM) Fine-Tuning and Inference Optimization

Tuhin Sharma · Bionic / Independent AI Advisory

Engineered domain-specific distillation pipelines compressing 70B parameter models into 3B-7B parameter edge models for on-premise enterprise deployment with zero data leakage.

Scope & limitations

SLMs have lower zero-shot generalization on tasks outside the fine-tuning training distribution.

Context: Phi-3, LLaMA-3-8B, LoRA / QLoRA, vLLM, TensorRT-LLM.

View mission
researchChecked Sep 22, 2026

Superhuman AI for Multiplayer Poker: Pluribus

Dr. Noam Brown · OpenAI

Developed Pluribus, the first AI to defeat elite human professionals in six-player no-limit Texas hold'em, demonstrating that depth-limited search with real-time belief distribution solving can scale to high-dimensional imperfect information.

Scope & limitations

Imperfect information equilibrium computation is computationally expensive for state spaces exceeding billions of action sequences.

Context: Monte Carlo CFR (Counterfactual Regret Minimization), depth-limited subgame solving.

View mission
researchChecked Sep 22, 2026

The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks

Dr. Jonathan Frankle · Databricks Mosaic AI / Harvard

Authored the seminal ICLR 2019 Best Paper showing dense, randomly-initialized feed-forward networks contain subnetworks ('winning tickets') that—when trained in isolation from early initialization—reach test accuracy comparable to the original network in similar iterations.

Scope & limitations

Finding winning tickets requires iterative prune-and-retrain cycles, which is computationally heavier than standard one-pass training.

Context: Iterative Magnitude Pruning (IMP), ResNet, VGG, MNIST, CIFAR-10.

View mission
researchChecked Sep 22, 2026

Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

Dr. Alexei Efros · UC Berkeley

Co-authored the landmark ICCV 2017 CycleGAN paper, introducing cycle consistency loss to enable high-quality visual translation between two unpaired image collections (horses to zebras, summer to winter) with 23,000+ citations.

Scope & limitations

Struggles with geometric transformations (e.g. cat to dog) because cycle consistency enforces structural topology preservation.

Context: Cycle consistency loss ||G(F(y)) - y||, dual generator/discriminator pairs.

View mission
researchChecked Sep 22, 2026

Very Deep Convolutional Networks for Large-Scale Image Recognition (VGG)

Dr. Karen Simonyan · Google DeepMind / Inflection AI

Authored the monumental ICLR 2015 paper demonstrating that pushing network depth to 16-19 layers using homogenous 3x3 convolution filters establishes state-of-the-art vision representations, with over 120,000 academic citations.

Scope & limitations

Fully connected classifier heads in original VGG architectures contain large parameter counts (138M+) prone to memory bottlenecks.

Context: VGG-16, VGG-19, ImageNet challenge winning architecture, perceptual loss standard.

View mission
researchChecked Sep 21, 2026

Accurate Structure Prediction of Biomolecular Interactions with AlphaFold 3

Dr. John Jumper · Google DeepMind

Co-led AlphaFold 3 research in Nature 2024, replacing Evoformer representations with a Diffusion Module to predict joint structures of proteins, nucleic acids (DNA/RNA), post-translational modifications, and small-molecule drugs.

Scope & limitations

Drug binding pocket predictions require experimental crystallization or binding assays for clinical drug validation.

Context: Pairformer architecture coupled with a 3D coordinate diffusion denoising module.

View mission
researchChecked Sep 21, 2026

AWQ: Activation-aware Weight Quantization for On-Device LLM Compression and Acceleration

Dr. Song Han · MIT / HAN Lab

Published the MLSys Best Paper proving that protecting the top 1% of salient weight channels based on activation magnitudes enables hardware-friendly 4-bit weight-only quantization without costly backpropagation or retraining.

Scope & limitations

Requires representative calibration datasets to accurately identify activation outliers across specialized domains.

Context: 4-bit integer weights, FP16 activations, evaluated across LLaMA-1/2, Vicuna, StarCoder.

View mission
researchChecked Sep 21, 2026

Aya: An Open Multilingual Foundation Model and Evaluation Dataset for 101 Languages

Dr. Sarah Hooker · Cohere For AI

Directed the global Aya research collaboration involving 3,000+ researchers across 119 countries to produce instruction datasets and state-of-the-art open-weights multilingual foundation models for underrepresented languages.

Scope & limitations

Extremely low-resource languages with minimal digital footprint require specialized human annotation to avoid hallucination.

Context: Cohere Command base weights, 101 languages, 513M human-curated multilingual instruction tokens.

View mission
researchChecked Sep 21, 2026

Classifier-Free Diffusion Guidance (CFG)

Dr. Jonathan Ho · Independent / ex-OpenAI

Invented Classifier-Free Guidance (CFG), a mathematical formulation that jointly trains conditional and unconditional diffusion models, dramatically improving prompt fidelity without requiring separate classifier gradients.

Scope & limitations

High guidance scales cause over-saturation and high-frequency pixel artifacts without dynamic thresholding.

Context: Conditioning dropout during training, weighted score extrapolation during inference sampling.

View mission
researchChecked Sep 21, 2026

Deep Residual Learning for Image Recognition (ResNet)

Dr. Kaiming He · MIT EECS

First author of the CVPR 2016 Best Paper introducing identity skip connections (Residual Networks), overcoming vanishing gradients to train 100+ layer networks, which became standard across all modern deep learning and transformers.

Scope & limitations

Pure convolutional ResNets possess localized receptive fields, necessitating hybrid attention blocks for global context modeling.

Context: Skip connections (x + F(x)), Batch Normalization, ResNet-50/101/152 architectures.

View mission
researchChecked Sep 21, 2026

Denoising Diffusion Probabilistic Models (DDPM)

Dr. Jonathan Ho · Independent / ex-OpenAI

First author of the groundbreaking NeurIPS 2020 paper that demonstrated diffusion models could surpass GANs and autoregressive models in generative image fidelity via variational bound simplification.

Scope & limitations

Standard DDPM required hundreds of sequential sampling steps, requiring modern ODE solvers or flow matching for real-time speed.

Context: Gaussian forward diffusion schedule, reverse Markov chain denoising score matching, U-Net architecture.

View mission
researchChecked Sep 21, 2026

Dense Passage Retrieval for Open-Domain Question Answering (DPR)

Dr. Danqi Chen · Princeton University

Published the landmark EMNLP 2020 paper demonstrating that dual-encoder BERT architectures trained on contrastive negative passages dramatically outperform classic BM25 and TF-IDF sparse retrieval in question answering.

Scope & limitations

Pure dense retrieval can struggle on out-of-domain lexical keywords and acronyms compared to hybrid dense-sparse search.

Context: Dual BERT encoders, dot-product similarity search over FAISS vector indices.

View mission
researchChecked Sep 21, 2026

Efficient Memory Management for Large Language Model Serving with PagedAttention

Woosuk Kwon · vLLM / UC Berkeley

Authored the seminal SOSP 2023 paper introducing PagedAttention, inspired by operating system virtual memory paging, allowing non-contiguous KV-cache storage and near-zero memory waste during high-concurrency LLM inference.

Scope & limitations

Paged cache lookup introduces minor pointer indirection overhead in memory-bandwidth-bound token generation loops.

Context: Evaluated across LLaMA, OPT, and Falcon architectures with continuous batching and multi-query attention.

View mission
researchChecked Sep 21, 2026

End-to-End Training of Deep Visuomotor Policies

Dr. Sergey Levine · UC Berkeley / Physical Intelligence

Published the foundational Journal of Machine Learning Research paper proving that robotic controllers can map raw camera pixels directly to continuous motor torque voltages using guided policy search.

Scope & limitations

Direct pixel-to-torque policies are vulnerable to novel environmental lighting and visual background shifts.

Context: Convolutional neural network visuomotor policies operating on physical PR2 robotic systems.

View mission
researchChecked Sep 21, 2026

Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer (T5)

Dr. Colin Raffel · University of Toronto / Vector Institute

First author of the seminal T5 paper establishing the unified text-to-text paradigm across all NLP tasks, introducing the C4 (Colossal Clean Crawled Corpus) dataset with over 25,000 academic citations.

Scope & limitations

Encoder-decoder architecture has higher KV-cache memory requirements during generation than pure autoregressive decoders.

Context: Encoder-decoder transformer architecture up to 11B parameters trained on C4 web corpus.

View mission
Showing 65–80 of 149 proofs of work