7 papers
New Evidence, Same Choice: Testing Physical Experiment Selection in Vision Language Models
Sourajit Saha, Shubhashis Roy Dipta, Nobin Sarwar +4
A model first sees an image from one physical measurement experiment, such as how far a block coasted, and must answer a question about a new trial, such as whether the block will…
BEFORE THE FLIP: Measuring Hidden Score Shifts In Quantized Vision Language Models Before The Answer Changes for Visual Question Answering
Sourajit Saha, Shubhashis Roy Dipta, Shaswati Saha +2
Quantization makes vision language models (VLMs) cheaper to store and run by using fewer bits to represent their weights. While unchanged answers on visual question answering (VQA)…
OracleZoom: On-Policy Self-Distillation Inspired Reference-Constrained Recursive Image Super Resolution
Shubhashis Roy Dipta, Sourajit Saha, Shaswati Saha +1
Recursive Super-Resolution (SR) extends fixed-scale SR to extreme magnification by repeatedly feeding predictions back into the same model, analogous to zooming an image repeatedly…
Last Translation Benchmark
Vilém Zouhar, Niyati Bafna, Mukund Choudhary +241
For scientific progress, we need benchmarks that test the limits of state-of-the-art models, and evaluation methods that inform us about failure cases. As models get stronger, stan…
To Erase, or Not to Erase: Robust Training-Free Concept Erasure with Preservation aware Adaptive Ranked Subspace Expansion
Shaswati Saha, Rajasekhar Anguluri, Manas Gaur
Concept erasure techniques (CETs) edit text-to-image diffusion models to erase undesired targets such as NSFW content or copyrighted styles, while preserving model utility on benig…
Side Effects of Erasing Concepts from Diffusion Models
Shaswati Saha, Sourajit Saha, Manas Gaur +1
Concerns about text-to-image (T2I) generative models infringing on privacy, copyright, and safety have led to the development of concept erasure techniques (CETs). The goal of an e…