activity
20242026
collaborators

19 papers

cs.LG2026

Accurate and Resource-Efficient Federated Continual Learning

Jebacyril Arockiaraj, Dhruv Parikh, Jayashree Adivarahan +2

Federated continual learning (FCL) must learn from distributed task streams under limited resources, such as communication, computation, memory, and label availability. Existing FC…

cs.CV2026

Can Graphs Help Vision SSMs See Better?

Dhruv Parikh, Anvitha Ramachandran, Haoyang Fan +3

Vision state space models inherit the efficiency and long-range modeling ability of Mamba-style selective scans. However, their performance depends critically on the representation…

cs.LG2026

TypeBandit: Type-Level Context Allocation and Reweighting for Effective Attribute Completion in Heterogeneous Graph Neural Networks

Ta-Yang Wang, Rajgopal Kannan, Viktor Prasanna

Heterogeneous graphs are widely used to model multi-relational systems, but missing node attributes remain a major bottleneck for downstream learning. In this paper, we identify an…

cs.LG2026

Action-Graph Policies: Learning Action Co-dependencies in Multi-Agent Reinforcement Learning

Nikunj Gupta, James Zachary Hare, Jesse Milzman +2

Coordinating actions is the most fundamental form of cooperation in multi-agent reinforcement learning (MARL). Successful decentralized decision-making often depends not only on go…

cs.LG2026

Deep Meta Coordination Graphs for Multi-agent Reinforcement Learning

Nikunj Gupta, James Zachary Hare, Jesse Milzman +2

This paper presents deep meta coordination graphs (DMCG) for learning cooperative policies in multi-agent reinforcement learning (MARL). Coordination graph formulations encode loca…

cs.CV2026

ConsensusDrop: Fusing Visual and Cross-Modal Saliency for Efficient Vision Language Models

Dhruv Parikh, Haoyang Fan, Rajgopal Kannan +1

Vision-Language Models (VLMs) are expensive because the LLM processes hundreds of largely redundant visual tokens. Existing token reduction methods typically exploit \textit{either…