AIConfPaper
← Search

Saketh Bachu

5 accepted papers

2025

Causal Order: The Key to Leveraging Imperfect Experts in Causal Inference

Aniket Vashishtha, Abbavaram Gowtham Reddy, Abhinav Kumar, Saketh Bachu, Vineeth N. Balasubramanian, Amit Sharma
ICLR 2025poster

Large Language Models (LLMs) have recently been used as experts to infer causal graphs, often by repeatedly applying a pairwise prompt that asks about the causal relationship of each variable pair. However, such experts, including human domain experts, cannot distinguish between direct and indirect…

Cited by 1Source ↗PDF ↗Scholar ↗
2025

Layer-wise Alignment: Examining Safety Alignment Across Image Encoder Layers in Vision Language Models

Saketh Bachu, Erfan Shayegani, Rohit Lal, Trishna Chakraborty, Arindam Dutta, Chengyu Song +3
ICML 2025spotlight

Vision-language models (VLMs) have improved significantly in their capabilities, but their complex architecture makes their safety alignment challenging. In this paper, we reveal an uneven distribution of harmful information across the intermediate layers of the image encoder and show that skipping…

Cited by 0Source ↗PDF ↗Scholar ↗
2025

VOccl3D: A Video Benchmark Dataset for 3D Human Pose and Shape Estimation under real Occlusions

Yash Garg, Saketh Bachu, Arindam Dutta, Rohit Lal, Sarosij Bose, Calvin-Khang Ta +2
ICCV 2025poster

Human pose and shape (HPS) estimation methods have been extensively studied, with many demonstrating high zero-shot performance on in-the-wild images and videos. However, these methods often struggle in challenging scenarios involving complex human poses or significant occlusions. Although some stud…

Cited by 0Source ↗PDF ↗Scholar ↗
2024

Towards Learning and Explaining Indirect Causal Effects in Neural Networks

Abbavaram Gowtham Reddy, Saketh Bachu, Harsharaj Pathak, Benin Godfrey L, Varshaneya V, Vineeth N Balasubramanian +1
AAAI 2024technical

Recently, there has been a growing interest in learning and explaining causal effects within Neural Network (NN) models. By virtue of NN architectures, previous approaches consider only direct and total causal effects assuming independence among input variables. We view an NN as a structural causal…

Cited by 1Source ↗PDF ↗code ↗Scholar ↗
2023

Building a Winning Team: Selecting Source Model Ensembles using a Submodular Transferability Estimation Approach

Vimal K B, Saketh Bachu, Tanmay Garg, Niveditha Lakshmi Narasimhan, Raghavan Konuru, Vineeth N Balasubramanian
ICCV 2023poster

Estimating the transferability of publicly available pre-trained models to a target task has assumed an important place for transfer learning tasks in recent years. Existing efforts propose metrics that allow a user to choose one model from a pool of pre-trained models without having to fine-tune ea…

Cited by 6PDF ↗Scholar ↗