Showing 1–2 of 2 results for author: P., N C

GPT-Neo for commonsense reasoning -- a theoretical and practical lens

Authors: Rohan Kashyap, Vivek Kashyap, Narendra C. P.

Abstract: Recent work has demonstrated substantial gains in pre-training large-language models (LLMs) followed by supervised fine-tuning on the downstream task. In this paper, we evaluate the performance of the GPT-neo model using $6$ commonsense reasoning benchmark tasks. We aim to examine the performance of smaller models using the GPT-neo models against several larger model baselines such as GPT-$3$, Lla… ▽ More Recent work has demonstrated substantial gains in pre-training large-language models (LLMs) followed by supervised fine-tuning on the downstream task. In this paper, we evaluate the performance of the GPT-neo model using $6$ commonsense reasoning benchmark tasks. We aim to examine the performance of smaller models using the GPT-neo models against several larger model baselines such as GPT-$3$, Llama-$2$, MPT and Falcon. Upon fine-tuning with the appropriate set of hyperparameters, our model achieves competitive accuracy on several tasks. We also investigate and substantiate our results using attention-head visualization to better understand the model performance. Finally, we conduct various robustness tests using various methods to gauge the model performance under numerous settings. △ Less

Submitted 27 September, 2023; v1 submitted 28 November, 2022; originally announced November 2022.

Automated Detection of Double Nuclei Galaxies using GOTHIC and the Discovery of a Large Sample of Dual AGN

Authors: Anwesh Bhattacharya, Nehal C. P., Mousumi Das, Abhishek Paswan, Snehanshu Saha, Francoise Combes

Abstract: We present a novel algorithm to detect double nuclei galaxies (DNG) called GOTHIC (Graph BOosted iterated HIll Climbing) - that detects whether a given image of a galaxy has two or more closely separated nuclei. Our aim is to detect samples of dual or multiple active galactic nuclei (AGN) in galaxies. Although galaxy mergers are common, the detection of dual AGN is rare. Their detection is very im… ▽ More We present a novel algorithm to detect double nuclei galaxies (DNG) called GOTHIC (Graph BOosted iterated HIll Climbing) - that detects whether a given image of a galaxy has two or more closely separated nuclei. Our aim is to detect samples of dual or multiple active galactic nuclei (AGN) in galaxies. Although galaxy mergers are common, the detection of dual AGN is rare. Their detection is very important as they help us understand the formation of supermassive black hole (SMBH) binaries, SMBH growth and AGN feedback effects in multiple nuclei systems. There is thus a need for an algorithm to do a systematic survey of existing imaging data for the discovery of DNGs and dual AGN. We have tested GOTHIC on a known sample of DNGs and subsequently applied it to a sample of a million SDSS DR16 galaxies lying in the redshift range of 0 to 0.75 approximately, and have available spectroscopic data. We have detected 159 dual AGN in this sample, of which 2 are triple AGN systems. Our results show that dual AGN are not common, and triple AGN even rarer. The color (u-r) magnitude plots of the DNGs indicate that star formation is quenched as the nuclei come closer and as the AGN fraction increases. The quenching is especially prominent for dual/triple AGN galaxies that lie in the extreme end of the red sequence. △ Less

Submitted 11 July, 2023; v1 submitted 23 November, 2020; originally announced November 2020.

Comments: Accepted at MNRAS