5 citations · 8 across the 13 of their papers we have counts for
13 papers
The Agent Incident Registry: Toward Preventing Repeated AI Agent Failures
Divyanshu Kumar, Rohith HN, Nitin Aravind Birur +2
AI agents increasingly act through tools and delegated authority, but general incident repositories rarely capture the mechanisms needed to compare public failures with agent-secur…
Black-Box Red Teaming of Agentic AI: A Taxonomy-Driven Framework for Automated Risk Discovery
Divyanshu Kumar, Nitin Aravind Birur, Tanay Baswa +2
Agentic systems are rapidly moving to production, where they read untrusted inputs, call tools with real permissions, and act autonomously, expanding the security surface beyond ch…
Redirected, Not Removed: Task-Dependent Stereotyping Reveals the Limits of LLM Alignments
Divyanshu Kumar, Ishita Gupta, Nitin Aravind Birur +3
How biased is a language model? The answer depends on how you ask. A model that refuses to choose between castes for a leadership role will, in a fill-in-the-blank task, reliably a…
SocioEval: A Template-Based Framework for Evaluating Socioeconomic Status Bias in Foundation Models
Divyanshu Kumar, Ishita Gupta, Nitin Aravind Birur +3
As Large Language Models (LLMs) increasingly power decision-making systems across critical domains, understanding and mitigating their biases becomes essential for responsible AI d…
Lost in Serialization: Invariance and Generalization of LLM Graph Reasoners
Daniel Herbst, Lea Karbevska, Divyanshu Kumar +3
While promising, graph reasoners based on Large Language Models (LLMs) lack built-in invariance to symmetries in graph representations. Operating on sequential graph serializations…
Quantifying CBRN Risk in Frontier Models
Divyanshu Kumar, Nitin Aravind Birur, Tanay Baswa +2
Frontier Large Language Models (LLMs) pose unprecedented dual-use risks through the potential proliferation of chemical, biological, radiological, and nuclear (CBRN) weapons knowle…
Beyond Text: Multimodal Jailbreaking of Vision-Language and Audio Models through Perceptually Simple Transformations
Divyanshu Kumar, Shreyas Jena, Nitin Aravind Birur +3
Multimodal large language models (MLLMs) have achieved remarkable progress, yet remain critically vulnerable to adversarial attacks that exploit weaknesses in cross-modal processin…
Beyond Western Politics: Cross-Cultural Benchmarks for Evaluating Partisan Associations in LLMs
Divyanshu Kumar, Ishita Gupta, Nitin Aravind Birur +3
Partisan bias in LLMs has been evaluated to assess political leanings, typically through a broad lens and largely in Western contexts. We move beyond identifying general leanings t…
No Free Lunch with Guardrails
Divyanshu Kumar, Nitin Aravind Birur, Tanay Baswa +2
As large language models (LLMs) and generative AI become widely adopted, guardrails have emerged as a key tool to ensure their safe use. However, adding guardrails isn't without tr…
Investigating Implicit Bias in Large Language Models: A Large-Scale Study of Over 50 LLMs
Divyanshu Kumar, Umang Jain, Sahil Agarwal +1
Large Language Models (LLMs) are being adopted across a wide range of tasks, including decision-making processes in industries where bias in AI systems is a significant concern. Re…
VERA: Validation and Enhancement for Retrieval Augmented systems
Nitin Aravind Birur, Tanay Baswa, Divyanshu Kumar +3
Large language models (LLMs) exhibit remarkable capabilities but often produce inaccurate responses, as they rely solely on their embedded knowledge. Retrieval-Augmented Generation…
SAGE-RT: Synthetic Alignment data Generation for Safety Evaluation and Red Teaming
Anurakt Kumar, Divyanshu Kumar, Jatan Loya +4
We introduce Synthetic Alignment data Generation for Safety Evaluation and Red Teaming (SAGE-RT or SAGE) a novel pipeline for generating synthetic alignment and red-teaming data. E…
Fine-Tuning, Quantization, and LLMs: Navigating Unintended Outcomes
Divyanshu Kumar, Anurakt Kumar, Sahil Agarwal +1
Large Language Models (LLMs) have gained widespread adoption across various domains, including chatbots and auto-task completion agents. However, these models are susceptible to sa…