Praneeth Netrapalli

Period spacings and global seismic parameters for K2 red giants using deep learning

Nipun Ghanghas, Siddharth Dhanpal, Shravan Hanasoge, Praneeth Netrapalli et autres

Gravity-mode period spacings (DPi_1) of red giants probe the stellar core directly, constraining its structure, mass and evolutionary state. Their measurement requires resolving narrow, densely spaced mixed modes and has so far relied on the four-year baseline of Kepler. Recovering DPi_1 from …

Period spacings and global seismic parameters for K2 red giants using deep learning

Nipun Ghanghas, Siddharth Dhanpal, Shravan Hanasoge, Praneeth Netrapalli et autres

Gravity-mode period spacings (DPi_1) of red giants probe the stellar core directly, constraining its structure, mass and evolutionary state. Their measurement requires resolving narrow, densely spaced mixed modes and has so far relied on the four-year baseline of Kepler. Recovering DPi_1 from …

in, us (code pays fourni par la source)

A model of errors in transformers

Suvrat Raju, Praneeth Netrapalli

We study the error rate of LLMs on tasks like arithmetic that require a deterministic output, and repetitive processing of tokens drawn from a small set of alternatives. We argue that incorrect predictions arise when small errors in the attention mechanism accumulate …

A model of errors in transformers

Suvrat Raju, Praneeth Netrapalli

We study the error rate of LLMs on tasks like arithmetic that require a deterministic output, and repetitive processing of tokens drawn from a small set of alternatives. We argue that incorrect predictions arise when small errors in the attention mechanism accumulate …

A Model of errors in transformers

Suvrat Raju, Praneeth Netrapalli

This dataset accompanies the paper "A model of errors in transformers." Filenames The data is provided in 25 csv files. Each filename comprises a model name and a task name, as specified in the main text. The acronyms for models are the …

in (code pays fourni par la source)

A Model of errors in transformers

Suvrat Raju, Praneeth Netrapalli

This dataset accompanies the paper "A model of errors in transformers." Filenames The data is provided in 25 csv files. Each filename comprises a model name and a task name, as specified in the main text. The acronyms for models are the …

in (code pays fourni par la source)

Compressing Many-Shots in In-Context Learning

Devvrit Khatri, Pranamya Kulkarni, Nilesh Gupta, Yerram Varun et autres

Large Language Models (LLMs) have been shown to be able to learn different tasks without explicit finetuning when given many input-output examples / demonstrations through In-Context Learning (ICL). Increasing the number of examples, called ``shots'', improves downstream task performance but incurs higher …

FoodLens: Fine-Grained and Multi-label Classification of Indian Food Images

Jatin Alla, Yashas Samaga, Ashwin Vaswani, Praneeth Netrapalli et autres

us (code pays fourni par la source)

Spark Transformer: Reactivating Sparsity in FFN and Attention

Chong You, Kan Wu, Z. Jiao, Lin Chen et autres

The discovery of the lazy neuron phenomenon in trained Transformers, where the vast majority of neurons in their feed-forward networks (FFN) are inactive for each token, has spurred tremendous interests in activation sparsity for enhancing large model efficiency. While notable progress has …

A Faster Generalized Two-Stage Approximate Top-K

Yashas Samaga, Varun Yerram, Spandana Raj Babbula, Prateek Jain et autres

We consider the Top-$K$ selection problem, which aims to identify the largest $K$ elements in an array. Top-$K$ selection arises in many machine learning algorithms and often becomes a bottleneck on accelerators, which are optimized for dense matrix multiplications. To address this …

Spark Transformer: Reactivating Sparsity in Transformer FFN and Attention

Chong You, Kan Wu, Zhipeng Jia, Lin Chen et autres

us, cn, jp, gb (code pays fourni par la source)

All Mistakes are not Equal: Comprehensive Hierarchy Aware Multilabel Predictions (CHAMP)

Ashwin Vaswani, Yashas Samaga, Gaurav Aggarwal, Praneeth Netrapalli et autres

us (code pays fourni par la source)