Souvik Kundu
32 papers · 2016–2025 · 11 venues · across top CS/AI venues
This profile combines several different researchers. Authors are keyed by name, and this name's papers fall into 13 separate collaboration groups - the largest holds only 41% of them. Pick the group below that matches the researcher you mean. How a shared name is split.
Achievements are not shown for a combined profile - each one describes a single researcher's record, which this is not.
Which Souvik Kundu?
5 groups of papers that share co-authors internally but none with each other - each most likely a different researcher. Pick one to see its papers.
with Zhangyang Wang, Shiwei Liu, Duc Hoang
4 papers
· 2023–2025
· ICML, NeurIPS, ICCV
Hide ▲
On-the-Fly Adaptive Distillation of Transformer to Dual-State Linear Attention for Long-Context LLM Serving
ICML 2025
Junk DNA Hypothesis: Pruning Small Pre-Trained Weights Irreversibly and Monotonically Impairs “Difficult" Downstream Tasks in LLMs
ICML 2024
Don’t just prune by magnitude! Your mask topology is a secret weapon
NeurIPS 2023
Vision HGNN: An Image is More than a Graph of Nodes
ICCV 2023
8 papers whose co-authors appear nowhere else under this name
LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding
ICLR 2025
LVLM-Compress-Bench: Benchmarking the Broader Impact of Large Vision-Language Model Compression
NAACL 2025
Scaling Long Context Training Data by Long-Distance Referrals
ICLR 2025
Fusing Models with Complementary Expertise
ICLR 2024
GenQ: Quantization in Low Data Regimes with Generative Synthetic Data
ECCV 2024
ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization
NeurIPS 2024
Exploiting Explicit Paths for Multi-hop Reading Comprehension
ACL 2019
Incorporating a Generative Front-End Layer to Deep Neural Network for Noise Robust Automatic Speech Recognition
Interspeech 2016