10 papers
You Can't Escape Your Own Activations : Evaluation Awareness and Multi-Agent Monitoring
Aritra Das, Jaee Ponde, Mihir More +1
LLM agents are increasingly deployed in multi-agent systems, where they can collude while keeping their actions benign. Output monitors designed to detect such collusions can be fo…
On the Indistinguishability of Human v/s AI Generated Text
Jaee Ponde, Aritra Das, Mihir More +1
The rapid improvement of LLMs has made distinguishing AI-generated text from human writing a pressing problem. This challenge is further amplified by paraphrasing tools designed to…
Parameterized Complexity of -Lipschitz Constants for Input Convex Neural Networks and -Norm Maximization over Zonotopes
Aritra Das, Vincent Froese, Moritz Grillo +6
Lipschitz constants are a standard way to quantify the sensitivity of neural networks to small input perturbations, but computing them is difficult even for shallow ReLU networks.…
The Query Knows What to Forget: A Second Erase Direction for Linear Attention
Dhruman Gupta, Aritra Das, Debayan Gupta
Linear attention keeps a state of fixed size. At long context, many stored items share this state, and interference between them degrades retrieval. Gated DeltaNet-2 (GDN-2), like…
Linearized 2-Simplicial Attention
Aritra Das, Dhruman Gupta, Debayan Gupta
We present a linearized form of 2-simplicial attention by rewriting the trilinear score as an inner product between a composite query and a key, so that the sum over one token axis…
A Rate Separation for Agnostic Direct Sums
Mihir More, Aritra Das, Debayan Gupta
Hanneke, Moran, and Waknine \cite{HannekeMoranWaknine2024} asked how the agnostic PAC learning curve of the direct sum depends on the single-instance learning curve $\epsagn(…