Janupalli Pranay
pranay-j
AI & ML interests
NLP
Recent Activity
liked a model 6 days ago
nvidia/GR00T-N1.7-3B liked a model about 2 months ago
nvidia/NVIDIA-NemotronLabs-VoiceChat-11B liked a model about 2 months ago
Edge0/Audio8-TTS-Preview-0.6B-ONNX-INT4Organizations
None yet
Novel Attention Mechanisms
LLM_architectures
-
Nemotron-4 15B Technical Report
Paper • 2402.16819 • Published • 46 -
Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models
Paper • 2402.19427 • Published • 58 -
RWKV: Reinventing RNNs for the Transformer Era
Paper • 2305.13048 • Published • 21 -
Reformer: The Efficient Transformer
Paper • 2001.04451 • Published • 2
Text to Speech Architectures
-
FastPitch: Parallel Text-to-speech with Pitch Prediction
Paper • 2006.06873 • Published -
HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis
Paper • 2010.05646 • Published • 2 -
Tacotron: Towards End-to-End Speech Synthesis
Paper • 1703.10135 • Published -
Parallel Tacotron: Non-Autoregressive and Controllable TTS
Paper • 2010.11439 • Published
grammarly
Memory efficient training
Instruction tuning datasets
reward model dataset
Domain adaption of dense retrieval
-
GPL: Generative Pseudo Labeling for Unsupervised Domain Adaptation of Dense Retrieval
Paper • 2112.07577 • Published -
TSDAE: Using Transformer-based Sequential Denoising Auto-Encoder for Unsupervised Sentence Embedding Learning
Paper • 2104.06979 • Published -
Text Embeddings by Weakly-Supervised Contrastive Pre-training
Paper • 2212.03533 • Published • 2 -
SimCSE: Simple Contrastive Learning of Sentence Embeddings
Paper • 2104.08821 • Published
audio-language-model-architecture
LLM - Cross Lingual Language Understanding
advanced-rag
NLP Parameter Efficient Finetuning
Explore research optimizing NLP models for maximum performance with minimal computational resources.
-
Efficient Few-Shot Learning Without Prompts
Paper • 2209.11055 • Published • 7 -
Parameter-Efficient Transfer Learning for NLP
Paper • 1902.00751 • Published • 2 -
GPT Understands, Too
Paper • 2103.10385 • Published • 11 -
The Power of Scale for Parameter-Efficient Prompt Tuning
Paper • 2104.08691 • Published • 10
Automatic Speech Recognition Architectures
-
Robust Speech Recognition via Large-Scale Weak Supervision
Paper • 2212.04356 • Published • 56 -
Conformer: Convolution-augmented Transformer for Speech Recognition
Paper • 2005.08100 • Published • 1 -
wav2vec: Unsupervised Pre-training for Speech Recognition
Paper • 1904.05862 • Published -
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations
Paper • 2006.11477 • Published • 10
graident optimization
Multimodal
-
MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training
Paper • 2403.09611 • Published • 130 -
OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documents
Paper • 2306.16527 • Published • 48 -
Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models
Paper • 2404.12387 • Published • 41 -
SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension
Paper • 2404.16790 • Published • 9
Language Model Pretraining Dataset
positional encoding Language models
Datasets: For training Embedding Models
This provides data sources for training and evaluating embedding models
Dataset- NLP Reasoning
LLM - Cross Lingual Language Understanding
Novel Attention Mechanisms
advanced-rag
LLM_architectures
-
Nemotron-4 15B Technical Report
Paper • 2402.16819 • Published • 46 -
Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models
Paper • 2402.19427 • Published • 58 -
RWKV: Reinventing RNNs for the Transformer Era
Paper • 2305.13048 • Published • 21 -
Reformer: The Efficient Transformer
Paper • 2001.04451 • Published • 2
NLP Parameter Efficient Finetuning
Explore research optimizing NLP models for maximum performance with minimal computational resources.
-
Efficient Few-Shot Learning Without Prompts
Paper • 2209.11055 • Published • 7 -
Parameter-Efficient Transfer Learning for NLP
Paper • 1902.00751 • Published • 2 -
GPT Understands, Too
Paper • 2103.10385 • Published • 11 -
The Power of Scale for Parameter-Efficient Prompt Tuning
Paper • 2104.08691 • Published • 10
Text to Speech Architectures
-
FastPitch: Parallel Text-to-speech with Pitch Prediction
Paper • 2006.06873 • Published -
HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis
Paper • 2010.05646 • Published • 2 -
Tacotron: Towards End-to-End Speech Synthesis
Paper • 1703.10135 • Published -
Parallel Tacotron: Non-Autoregressive and Controllable TTS
Paper • 2010.11439 • Published
Automatic Speech Recognition Architectures
-
Robust Speech Recognition via Large-Scale Weak Supervision
Paper • 2212.04356 • Published • 56 -
Conformer: Convolution-augmented Transformer for Speech Recognition
Paper • 2005.08100 • Published • 1 -
wav2vec: Unsupervised Pre-training for Speech Recognition
Paper • 1904.05862 • Published -
wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations
Paper • 2006.11477 • Published • 10
grammarly
graident optimization
Memory efficient training
Multimodal
-
MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training
Paper • 2403.09611 • Published • 130 -
OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documents
Paper • 2306.16527 • Published • 48 -
Reka Core, Flash, and Edge: A Series of Powerful Multimodal Language Models
Paper • 2404.12387 • Published • 41 -
SEED-Bench-2-Plus: Benchmarking Multimodal Large Language Models with Text-Rich Visual Comprehension
Paper • 2404.16790 • Published • 9
Instruction tuning datasets
Language Model Pretraining Dataset
reward model dataset
positional encoding Language models
Domain adaption of dense retrieval
-
GPL: Generative Pseudo Labeling for Unsupervised Domain Adaptation of Dense Retrieval
Paper • 2112.07577 • Published -
TSDAE: Using Transformer-based Sequential Denoising Auto-Encoder for Unsupervised Sentence Embedding Learning
Paper • 2104.06979 • Published -
Text Embeddings by Weakly-Supervised Contrastive Pre-training
Paper • 2212.03533 • Published • 2 -
SimCSE: Simple Contrastive Learning of Sentence Embeddings
Paper • 2104.08821 • Published
Datasets: For training Embedding Models
This provides data sources for training and evaluating embedding models
audio-language-model-architecture