Arush Sharma, Anupama Ray · manuscript in preparation

Recursive QAOA keeps making irreversible sign errors in its correlators; hedging both signs with a small beam and an evolved power-law schedule removes them.

method A multi-agent loop proposed, mutated, critiqued and evaluated analytic QAOA schedules, evolving the linear ramp into a fixed power-law schedule. The schedule then drives a width-B sign-hedged beam over the recursive eliminations, evaluated on simulators, MPS and IBM hardware.

my part The algorithm-discovery loop, the beam-hedged recursion, and the simulator, MPS and hardware evaluations.

numbers

+ 0.998  approximation ratio, SK model n=20, simulation,
         2.1× over baseline
+ 1.000  on IBM Marrakesh hardware
+ 0.984  MaxCut n=60, 17.6% above QAOA
+ 0.983  20-asset portfolio, 26% above RQAOA

Best Paper certificate

Arush Sharma, Soumen Basu, Manoj Kumar

PACL predicts a Gaussian latent per image patch instead of a point, and carries that uncertainty into classification and segmentation.

method A masked causal ViT with a distributional head predicts the latent distribution of future patches, trained with a β-NLL against frozen target embeddings. At fine-tuning time the predicted variance becomes an uncertainty score that modulates the classification and segmentation losses.

my part The mean/variance and anti-collapse losses, and the chest X-ray and ultrasound experiments.

numbers

+ 0.987  chest X-ray AUROC, +2.0 pts over prior best
+ 0.938  BUSI breast-ultrasound AUROC, +2.1 pts
+ 0.604  segmentation IoU, +1.3 pts

PDF

second author, with Soumen Basu and Manoj Kumar

Keeps one token per brain region instead of pooling regions away, and fuses global attention with reliability-weighted region-to-cluster context.

method A subject-specific brain graph is built from functional connectivity. A position-aware long-range encoder keeps every region's token, while a module-aware branch learns cluster context and fuses it back through a reliability-weighted gate, so regional identity survives to the readout.

my part The ablation suite, and co-writing the methodology and experiments.

numbers

+ 82.35%  accuracy on ABIDE, +3.35 pts over prior best
+ 93.75%  accuracy on OSF, +6.25 pts

PDF

Research Engineer Intern (Stackr Labs, backed by a16z). Draft-aligned decoding for SDAR-30B-A3B and FLARE-2B training, above.
diffusion LMs speculative decoding LoRA DDP/FSDP H200

Research Intern, AI for Quantum, with Dr. Anupama Ray. Built a multi-agent loop in which model agents proposed, mutated, critiqued and evaluated analytic QAOA schedules; used the best schedule in sign-hedged beam RQAOA and evaluated it on simulators, MPS and IBM hardware. The work won the Best Paper Award at the internship symposium.
QAOA beam search agentic search IBM Quantum

Research Intern, with Dr. Soumen Basu (Meta) and Dr. Manoj Kumar (IIT Roorkee). Uncertainty-aware self-supervised learning for medical images and graph models of brain networks. First author on PACL, second author on BrainRMA.
self-supervised learning ViT graph models

Agentic AI Intern. Fine-tuned MedGemma 4B for a medical-tourism assistant and deployed it as a Dockerized AWS pipeline with retrieval-augmented generation and LangSmith evaluation.
MedGemma RAG AWS