4 papers
The Hallucination Snowball: Modeling Error Propagation as State Transitions in Multi-Agent LLM Pipelines
Prabhjot Singh, Bhushan Pawar
Sequential multi-agent LLM pipelines chain specialized agents without verification at handoffs, creating a structural flaw with measurable and severe consequences. We show that hal…
DocHRL: A Hierarchical Reinforcement Learning Framework for Cost-Optimised Document Classification
Mohammed Yousif, Prabhjot Singh, Arjun Pankajakshan +1
Real-world document classification pipelines typically apply the same sequence of models to every incoming document, regardless of its complexity or type. This leads to inefficient…
Beyond 'One Language, One Script': Quantifying Orthographic Bias in Multilingual VLMs with PuMVR
Prabhjot Singh, Bhushan Pawar, Madhu Reddiboina
Current Vision-Language Models (VLMs) are celebrated for their multilingual capabilities, yet they operate under a flawed assumption: that one language corresponds to a single writ…
Not Truly Multilingual: Script Consistency as a Missing Dimension in VLM Evaluation
Prabhjot Singh, Bhushan Pawar, Madhu Reddiboina +1
Current multilingual evaluations for Vision-Language Models (VLMs) assume a one-to-one mapping between language and orthography, overlooking billions of users of multi-script langu…