Arnav Kundu

According to our database1, Arnav Kundu authored at least 16 papers between 2021 and 2026.

Collaborative distances:
  • Dijkstra number2 of four.
  • Erdős number3 of four.

Timeline

Legend:

Book  In proceedings  Article  PhD thesis  Dataset  Other 

Links

On csauthors.net:

Bibliography

2026
MemoryLLM: Plug-n-Play Interpretable Feed-Forward Memory for Transformers.
CoRR, February, 2026

2025
Mirror Speculative Decoding: Breaking the Serial Barrier in LLM Inference.
CoRR, October, 2025

MoE-PHDS: One MoE checkpoint for flexible runtime sparsity.
CoRR, September, 2025

EpiCache: Episodic KV Cache Management for Long Conversational Question Answering.
CoRR, September, 2025

MoEs Are Stronger than You Think: Hyper-Parallel Inference Scaling with RoE.
CoRR, September, 2025

Your LLM Knows the Future: Uncovering Its Multi-Token Prediction Potential.
CoRR, July, 2025

SPD: Sync-Point Drop for Efficient Tensor Parallelism of Large Language Models.
Proceedings of the Forty-second International Conference on Machine Learning, 2025

An Efficient and Streaming Audio Visual Active Speaker Detection System.
Proceedings of the 2025 IEEE International Conference on Acoustics, 2025

2024
RepCNN: Micro-sized, Mighty Models for Wakeword Detection.
CoRR, 2024

RepCNN: Micro-sized, Mighty Models for Wakeword Detection.
Proceedings of the 25th Annual Conference of the International Speech Communication Association, 2024

Streaming Anchor Loss: Augmenting Supervision with Temporal Significance.
Proceedings of the IEEE International Conference on Acoustics, 2024

2023
R^2: Range Regularization for Model Compression and Quantization.
CoRR, 2023

I See What You Hear: A Vision-Inspired Method to Localize Words.
Proceedings of the IEEE International Conference on Acoustics, 2023

HEiMDaL: Highly Efficient Method for Detection and Localization of Wake-Words.
Proceedings of the IEEE International Conference on Acoustics, 2023

2022
I see what you hear: a vision-inspired method to localize words.
CoRR, 2022

2021
Optimize What Matters: Training DNN-Hmm Keyword Spotting Model Using End Metric.
Proceedings of the IEEE International Conference on Acoustics, 2021


  Loading...