12 papers
Calibrating Artificial Guilt: Neurally Grounded Reward Shaping for Prosocial Multi-Agent Reinforcement Learning
Aaditya Mehta, Arya Shah
Cooperative multi-agent reinforcement learning often adds social terms to individual rewards, yet the scale of those terms is usually chosen by hand. We ask whether a guilt signal…
SketchMamba: A Lightweight State-Space Model for Joint Progressive Sketch Classification and Stroke Auto-Completion
Kavish Jhaveri, Arya Shah
Existing vector-sketch models treat recognition and generation as separate tasks, leaving a gap for streaming interfaces that must understand a drawing as it is being made. We pres…
Sycophancy as a Multilingual Alignment Failure: How Safety Degrades Across Languages, Topics, and Models
Arya Shah, Himanshu Beniwal, Mayank Singh +1
Safety-aligned large language models often exhibit sycophancy, which is the tendency to affirm users' opinions regardless of factual accuracy. Although well-studied in English, its…
Quantized Machine Learning Models for Medical Imaging in Low-Resource Healthcare Settings
Sumanth Meenan Kanneti, Aryan Shah
Deep learning models have shown strong performance in medical image analysis, but deploying them in low-resource clinical environments remains difficult due to computational, memor…
SycoPhantasy: Quantifying Sycophancy and Hallucination in Small Open Weight VLMs for Vision-Language Scoring of Fantasy Characters
Arya Shah, Deepali Mishra, Chaklam Silpasuwanchai
Vision-language models (VLMs) are increasingly deployed as evaluators in tasks requiring nuanced image understanding, yet their reliability in scoring alignment between images and…
Gaslight, Gatekeep, V1-V3: Early Visual Cortex Alignment Shields Vision-Language Models from Sycophantic Manipulation
Arya Shah, Vaibhav Tripathi, Mayank Singh +1
Vision-language models are increasingly deployed in high-stakes settings, yet their susceptibility to sycophantic manipulation remains poorly understood, particularly in relation t…
GF-Score: Certified Class-Conditional Robustness Evaluation with Fairness Guarantees
Arya Shah, Kaveri Visavadiya, Manisha Padala
Adversarial robustness is essential for deploying neural networks in safety-critical applications, yet standard evaluation methods either require expensive adversarial attacks or r…
Too Nice to Tell the Truth: Quantifying Agreeableness-Driven Sycophancy in Role-Playing Language Models
Arya Shah, Deepali Mishra, Chaklam Silpasuwanchai
Large language models increasingly serve as conversational agents that adopt personas and role-play characters at user request. This capability, while valuable, raises concerns abo…
One Instruction Does Not Fit All: How Well Do Embeddings Align Personas and Instructions in Low-Resource Indian Languages?
Arya Shah, Himanshu beniwal, Mayank Singh
Aligning multilingual assistants with culturally grounded user preferences is essential for serving India's linguistically diverse population of over one billion speakers across mu…
In Search of Goodness: Large Scale Benchmarking of Goodness Functions for the Forward-Forward Algorithm
Arya Shah, Vaibhav Tripathi
The Forward-Forward (FF) algorithm offers a biologically plausible alternative to backpropagation, enabling neural networks to learn through local updates. However, FF's efficacy r…
Purrturbed but Stable: Human-Cat Invariant Representations Across CNNs, ViTs and Self-Supervised ViTs
Arya Shah, Vaibhav Tripathi
Cats and humans differ in ocular anatomy. Most notably, Felis Catus (domestic cats) have vertically elongated pupils linked to ambush predation; yet, how such specializations manif…
Can NLP Models 'Identify', 'Distinguish', and 'Justify' Questions that Don't have a Definitive Answer?
Ayushi Agarwal, Nisarg Patel, Neeraj Varshney +7
Though state-of-the-art (SOTA) NLP systems have achieved remarkable performance on a variety of language understanding tasks, they primarily focus on questions that have a correct…