TrishanuDas (TRISHANU DAS)

upvoted a paper 2 months ago

left|,circlearrowright,text{BUS},right|: A Large and Diverse Multimodal Benchmark for evaluating the ability of Vision-Language Models to understand Rebus Puzzles

Paper • 2511.01340 • Published Nov 3, 2025 • 12

upvoted a paper 5 months ago

Leveraging Large Language Models for Predictive Analysis of Human Misery

Paper • 2508.12669 • Published Aug 18, 2025 • 14

upvoted a paper 7 months ago

Leveraging Self-Attention for Input-Dependent Soft Prompting in LLMs

Paper • 2506.05629 • Published Jun 5, 2025 • 37

upvoted a paper 8 months ago

REFINE-AF: A Task-Agnostic Framework to Align Language Models via Self-Generated Instructions using Reinforcement Learning from Automated Feedback

Paper • 2505.06548 • Published May 10, 2025 • 30

upvoted 2 articles 10 months ago

Article

You could have designed state of the art positional encoding

Nov 25, 2024

•

429

Article

makeMoE: Implement a Sparse Mixture of Experts Language Model from Scratch

May 7, 2024

•

112

upvoted a paper about 1 year ago

A Pointer Network-based Approach for Joint Extraction and Detection of Multi-Label Multi-Class Intents

Paper • 2410.22476 • Published Oct 29, 2024 • 27

upvoted a paper over 1 year ago

YesBut: A High-Quality Annotated Multimodal Dataset for evaluating Satire Comprehension capability of Vision-Language Models

Paper • 2409.13592 • Published Sep 20, 2024 • 50

upvoted an article over 1 year ago

Article

Making LLMs even more accessible with bitsandbytes, 4-bit quantization and QLoRA

+3

May 24, 2023

•

171

TRISHANU DAS

AI & ML interests

Organizations

left|,circlearrowright,text{BUS},right|: A Large and Diverse Multimodal Benchmark for evaluating the ability of Vision-Language Models to understand Rebus Puzzles

Leveraging Large Language Models for Predictive Analysis of Human Misery

Leveraging Self-Attention for Input-Dependent Soft Prompting in LLMs

REFINE-AF: A Task-Agnostic Framework to Align Language Models via Self-Generated Instructions using Reinforcement Learning from Automated Feedback

You could have designed state of the art positional encoding

makeMoE: Implement a Sparse Mixture of Experts Language Model from Scratch

A Pointer Network-based Approach for Joint Extraction and Detection of Multi-Label Multi-Class Intents

YesBut: A High-Quality Annotated Multimodal Dataset for evaluating Satire Comprehension capability of Vision-Language Models

Making LLMs even more accessible with bitsandbytes, 4-bit quantization and QLoRA

TRISHANU DAS

AI & ML interests

Organizations

TrishanuDas's activity

You could have designed state of the art positional encoding

makeMoE: Implement a Sparse Mixture of Experts Language Model from Scratch

Making LLMs even more accessible with bitsandbytes, 4-bit quantization and QLoRA