smajumdar94 (Somshubra Majumdar)

upvoted a collection 11 days ago

Qwen2.5

Qwen2.5 language models, including pretrained and instruction-tuned models of 7 sizes, including 0.5B, 1.5B, 3B, 7B, 14B, 32B, and 72B. • 45 items • Updated 15 days ago • 217

upvoted an article 15 days ago

Article

Fine-tuning LLMs to 1.58bit: extreme quantization made easy

16 days ago

• 140

upvoted a paper 24 days ago

How Do Your Code LLMs Perform? Empowering Code Instruction Tuning with High-Quality Data

Paper • 2409.03810 • Published 28 days ago • 30

upvoted a paper 29 days ago

FLUX that Plays Music

Paper • 2409.00587 • Published Sep 1 • 31

upvoted 2 papers about 1 month ago

FocusLLM: Scaling LLM's Context by Parallel Decoding

Paper • 2408.11745 • Published Aug 21 • 23

To Code, or Not To Code? Exploring Impact of Code in Pre-training

Paper • 2408.10914 • Published Aug 20 • 40

upvoted 2 papers about 2 months ago

LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs

Paper • 2408.07055 • Published Aug 13 • 65

VITA: Towards Open-Source Interactive Omni Multimodal LLM

Paper • 2408.05211 • Published Aug 9 • 46

upvoted a collection 2 months ago

Minitron

Collection

A family of compressed models obtained via pruning and knowledge distillation • 9 items • Updated about 11 hours ago • 54

upvoted an article 3 months ago

Article

Announcing BigCodeBench-Hard, and More

By

•

Jul 24

• 10

upvoted 2 papers 3 months ago

AgentInstruct: Toward Generative Teaching with Agentic Flows

Paper • 2407.03502 • Published Jul 3 • 43

Scaling Synthetic Data Creation with 1,000,000,000 Personas

Paper • 2406.20094 • Published Jun 28 • 93

upvoted an article 3 months ago

Article

BM25 for Python: Achieving high performance while simplifying dependencies with BM25S⚡

By

•

Jul 9

• 35

upvoted a paper 3 months ago

Self-play with Execution Feedback: Improving Instruction-following Capabilities of Large Language Models

Paper • 2406.13542 • Published Jun 19 • 16

upvoted an article 4 months ago

Article

BigCodeBench: Benchmarking Large Language Models on Solving Practical and Challenging Programming Tasks

Jun 18

• 35

upvoted 2 collections 4 months ago

Nemotron 4 340B

Collection

Nemotron-4: open models for Synthetic Data Generation (SDG). Includes Base, Instruct, and Reward models. • 4 items • Updated 3 days ago • 156

Nemotron 3 8B

Collection

The Nemotron 3 8B Family of models is optimized for building production-ready generative AI applications for the enterprise. • 5 items • Updated 3 days ago • 43

upvoted a paper 4 months ago

Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing

Paper • 2406.08464 • Published Jun 12 • 62

upvoted a paper 5 months ago

Beyond Scaling Laws: Understanding Transformer Performance with Associative Memory

Paper • 2405.08707 • Published May 14 • 27

upvoted an article 5 months ago

Article

Powerful ASR + diarization + speculative decoding with Hugging Face Inference Endpoints

May 1

• 63

upvoted a paper 5 months ago

Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models

Paper • 2404.18796 • Published Apr 29 • 68

upvoted an article 5 months ago

Article

StarCoder2-Instruct: Fully Transparent and Permissive Self-Alignment for Code Generation

Apr 29

• 71

upvoted an article 6 months ago

Article

Releasing Youtube-Commons: a massive open corpus for conversational and multimodal data

By

•

Apr 18

• 21

upvoted 5 papers 6 months ago

upvoted a paper 7 months ago

Algorithmic progress in language models

Paper • 2403.05812 • Published Mar 9 • 18

upvoted a collection 7 months ago

StarChat2 15B

Collection

Model, datasets, and demo for StarChat2 15B. For code to train the models, see: https://github.com/huggingface/alignment-handbook • 10 items • Updated Apr 12 • 13

upvoted 5 papers 7 months ago

PixArt-Σ: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation

Paper • 2403.04692 • Published Mar 7 • 40

GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection

Paper • 2403.03507 • Published Mar 6 • 182

Genie: Generative Interactive Environments

Paper • 2402.15391 • Published Feb 23 • 70

OpenCodeInterpreter: Integrating Code Generation with Execution and Refinement

Paper • 2402.14658 • Published Feb 22 • 82

LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens

Paper • 2402.13753 • Published Feb 21 • 111

upvoted a paper 8 months ago

Synthetic Data (Almost) from Scratch: Generalized Instruction Tuning for Language Models

Paper • 2402.13064 • Published Feb 20 • 46

upvoted a collection 8 months ago

OpenMath

Collection

A collection of models and datasets introduced in "OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset" • 15 items • Updated 3 days ago • 35

upvoted 2 papers 8 months ago

Self-Discover: Large Language Models Self-Compose Reasoning Structures

Paper • 2402.03620 • Published Feb 6 • 109

Multi-line AI-assisted Code Authoring

Paper • 2402.04141 • Published Feb 6 • 9

upvoted a collection 8 months ago

Parakeet

Collection

NeMo Parakeet ASR Models attain strong speech recognition accuracy while being efficient for inference. Available in CTC and RNN-Transducer variants. • 8 items • Updated 3 days ago • 17

upvoted a paper 8 months ago

MambaByte: Token-free Selective State Space Model

Paper • 2401.13660 • Published Jan 24 • 49

upvoted 3 papers 9 months ago

Beyond Chinchilla-Optimal: Accounting for Inference in Language Model Scaling Laws

Paper • 2401.00448 • Published Dec 31, 2023 • 27

WaveCoder: Widespread And Versatile Enhanced Instruction Tuning with Refined Data Generation

Paper • 2312.14187 • Published Dec 20, 2023 • 49

Analyzing and Improving the Training Dynamics of Diffusion Models

Paper • 2312.02696 • Published Dec 5, 2023 • 31

upvoted a paper 10 months ago

LLM in a flash: Efficient Large Language Model Inference with Limited Memory

Paper • 2312.11514 • Published Dec 12, 2023 • 257

upvoted 2 papers 11 months ago

Zephyr: Direct Distillation of LM Alignment

Paper • 2310.16944 • Published Oct 25, 2023 • 120

FreeU: Free Lunch in Diffusion U-Net

Paper • 2309.11497 • Published Sep 20, 2023 • 64

upvoted 7 papers about 1 year ago

Calibrating LLM-Based Evaluator

Paper • 2309.13308 • Published Sep 23, 2023 • 11

Dynamic ASR Pathways: An Adaptive Masking Approach Towards Efficient Pruning of A Multilingual ASR Model

Paper • 2309.13018 • Published Sep 22, 2023 • 9

A Large-scale Dataset for Audio-Language Representation Learning

Paper • 2309.11500 • Published Sep 20, 2023 • 9

Large-Scale Automatic Audiobook Creation

Paper • 2309.03926 • Published Sep 7, 2023 • 53

NExT-GPT: Any-to-Any Multimodal LLM

Paper • 2309.05519 • Published Sep 11, 2023 • 78

Instruction Mining: High-Quality Instruction Data Selection for Large Language Models

Paper • 2307.06290 • Published Jul 12, 2023 • 9

Retentive Network: A Successor to Transformer for Large Language Models

Paper • 2307.08621 • Published Jul 17, 2023 • 170

Somshubra Majumdar

AI & ML interests

Organizations

smajumdar94's activity

Fine-tuning LLMs to 1.58bit: extreme quantization made easy

Announcing BigCodeBench-Hard, and More

BM25 for Python: Achieving high performance while simplifying dependencies with *BM25S*⚡

BigCodeBench: Benchmarking Large Language Models on Solving Practical and Challenging Programming Tasks

Powerful ASR + diarization + speculative decoding with Hugging Face Inference Endpoints

StarCoder2-Instruct: Fully Transparent and Permissive Self-Alignment for Code Generation

Releasing Youtube-Commons: a massive open corpus for conversational and multimodal data

BM25 for Python: Achieving high performance while simplifying dependencies with BM25S⚡