(Last Updated: June 2026)
For the latest list of publications, visit: Google Scholar
2026
- Terminator: Learning Optimal Exit Points for Early Stopping in Chain-of-Thought Reasoning
A. Nagle, J. Saydaliev, D. Garbaya, M. Gastpar, A. V. Makkuva, H. Kim
arXiv preprint, 2026. [Blog] - Efficient Weighted Sampling via Score-based Generative Models
Heasung Kim, T. Lee, H. Kim, G. de Veciana
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026. - Fine-Tuning Masked Diffusion for Provable Self-Correction
J. Kim, S. Kim, T. Lee, D. Pan, H. Kim, S. Kakade, S. Chen
International Conference on Machine Learning (ICML), 2026. - Generating High Dimensional User-Specific Wireless Channels using Diffusion Models
T. Lee, J. Park, H. Kim, J. Andrews
IEEE Transactions on Wireless Communications (TWC), vol. 25, pp. 2907–2921, 2026. [blog] - Residual Diffusion Models for Variable-Rate Joint Source Channel Coding of MIMO CSI
S. Ankireddy, Heasung Kim, J. Cho, H. Kim
IEEE Journal on Selected Areas in Communications (JSAC), vol. 44, pp. 3620–3633, 2026. - Clustered Federated Learning to Support Context-dependent CSI Decoding
Heasung Kim, H. Kim, G. de Veciana
IEEE Transactions on Machine Learning in Communications and Networking, vol. 4, pp. 211–227, 2026.
2025
- Generating Informative Samples for Risk-Averse Fine-Tuning of Downstream Tasks
Heasung Kim, T. Lee, H. Kim, G. de Veciana
Conference on Neural Information Processing Systems (NeurIPS), Spotlight, 2025. - Attention with Markov: A Curious Case of Single-layer Transformers
A. Makkuva, M. Bondaschi, A. Girish, A. Nagle, M. Jaggi, H. Kim, M. Gastpar
International Conference on Learning Representations (ICLR), Spotlight, 2025. - Fundamental Limits to Exploiting Side Information for CSI Feedback in Wireless Systems
Heasung Kim, G. de Veciana, H. Kim
IEEE Journal on Selected Areas in Communications (JSAC), vol. 43, no. 7, pp. 2417–2430, July 2025. - Enhancing K-user Interference Alignment for Discrete Constellations via Learning
R. Mishra, S. Jafar, S. Vishwanath, H. Kim
IEEE Journal on Selected Areas in Communications (JSAC), vol. 43, no. 7, pp. 2405–2416, July 2025. - LIGHTCODE: Light Analytical and Neural Codes for Channels with Feedback
S. Ankireddy, K. Narayanan, H. Kim
IEEE Journal on Selected Areas in Communications (JSAC), vol. 43, no. 4, pp. 1230–1245, April 2025.
2024
- LIGHTCODE: Light Analytical and Neural Codes for Channels with Feedback
S Ankireddy, K Narayanan, H Kim
[IEEE Journal on Selected Areas in Communications (JSAC)][code] - Neural Distributed Source Coding
J Whang*, A Nagle*, A Acharya, H Kim, AG Dimakis
[IEEE Journal on Selected Areas in Information Theory (JSAIT)][blog][code] - Fundamental Limits of Prompt Compression: A Rate-Distortion Framework for Black-Box Language Models
A Girish*, A Nagle*, A V Makkuva, M Bondaschi, M Gastpar, H Kim
[NeurIPS 2024, ICML TF2M Workshop Oral 2024][blog][code] - Neural Cover Selection for Image Steganography
K Chahine, H Kim
[NeurIPS 2024][blog][code] - Local to Global: Learning Dynamics and Effect of Initialization for Transformers
AV Makkuva, M Bondaschi, A Girish, A Nagle, H Kim, M Gastpar, C Ekbote
[NeurIPS 2024] - Clustered Federated Learning via Gradients-based Partitioning
Heasung Kim, H. Kim, G. de Veciana
[ICML 2024][blog][code] - DeepPolar: Inventing Nonlinear Large-Kernel Polar Codes via Deep Learning
A Hebbar*, SK Ankireddy*, H. Kim, S. Oh, P. Viswanath
[ICML 2024][blog][code] - LASER: Linear Compression in Wireless Distributed Optimization
A. Makkuva, M. Bondaschi, T. Vogels, M. Jaggi, H. Kim, M. Gastpar
[ICML 2024] - Estimation of Rate-Distortion Function for Computing with Decoder Side Information
Heasung Kim, H Kim, G de Veciana
[ISIT 2024][blog][code] - Deep Learning-Based mmWave Beam Alignment with Only Pilot Channel Measurements
T Lee, H Kim, J Andrews
[ICC 2024] - Deep Learning-based Autodetection of 5G NR mmWave Waveforms
T Lee, A Mahadevan, H Kim, J Andrews
[ICC 2024] - Optimality of the Lossy Körner-Marton Scheme for Distributed Modulo-Two Sum Computation of Doubly Symmetric Binary Sources
S Li, H Kim
[Preprint]
2023
- Task-aware Distributed Source Coding under Dynamic Bandwidth
P Li*, SK Ankireddy*, R Zhao, H Mahjoub, E Pari, U Topcu, S Chinchali, H Kim
[NeurIPS 2023][blog] - Compressed Error HARQ: Feedback Communication on Noise-Asymmetric Channels
SK Ankireddy, SA Hebbar, Y Jiang, P Viswanath, H Kim
[ISIT 2023] - Interpreting Neural Min-Sum Decoders
SK Ankireddy, H Kim
[ICC 2023] - Linear Coding for AWGN channels with Noisy Output Feedback via Dynamic Programming
R. Mishra, D. Vasal, H. Kim
[IEEE Trans. on Information Theory] - DeepIC+: Learning Codes for Interference Channels
K Chahine, Y Jiang, J Cho, H Kim
[IEEE Trans on Wireless Communications][blog]
2022
- Channel Coding via Machine Learning (Book chapter)
H Kim
[Machine Learning and Wireless Communications, Y Eldar, A Goldsmith, D Gunduz, V Poor] - Inventing Codes for Channels with Active Feedback via Deep Learning
K Chahine, R Mishra, H Kim
[IEEE JSAIT 2022] - Learning Variable-Rate Codes for CSI Feedback
H Kim, H Kim, G de Veciana
[GLOBECOM 2022] - TinyTurbo: Efficient Turbo Decoders on Edge
S. Hebbar, R. Mishra, S. Ankireddy, A. Makkuva, H. Kim, P. Viswanath
[ISIT 2022] - Turbo Autoencoder with a Trainable Interleaver
K Chahine, Y Jiang, P Nuti, H Kim, J Cho[ICC 2022]
2021
- A Channel Coding Benchmark for Meta-Learning
R Li, O Bohdal, R Mishra, H Kim, D Li, N Lane, T Hospedales
- DeepIC: Coding for Interference Channels via Deep Learning
K Chahine, N Ye, H Kim
- Distributed Interference Alignment for K -user Interference Channels via Deep Learning
R Mishra, K Chahine, H Kim, S Jafar, S Vishwanath
[ISIT 2021] - Gaussian Channels with Feedback: A Dynamic Programming Approach
R Mishra, D Vasal, H Kim
[ISIT 2021]
2020
- BRP-NAS: Prediction-based NAS using GCNs
T Chau, Ł Dudziak, MS Abdelfattah, R Lee, H Kim, ND Lane
[NeurIPS 2020] - Journey Towards Tiny Perceptual Super-Resolution
R Lee, L Dudziak, M Abdelfattah, S Venieris, H Kim, H Wen, ND Lane
[ECCV 2020] - HAPI: Hardware-Aware Progressive Inference
S Laskaridis, S Venieris, H Kim, N Lane
[ICCAD 2020] - Best of Both Worlds: AutoML Codesign of a CNN and its Hardware Accelerator
M Abdelfattah, Ł Dudziak, T Chau, R Lee, H Kim, N Lane
[Design Automation Conference (DAC) 2020] - Joint Channel Coding and Modulation via Deep Learning
Y Jiang, H Kim, H Asnani, S Kannan, S Oh, P Viswanath
[SPAWC 2020] - Feedback Turbo Autoencoder
Y Jiang, H Kim, H Asnani, S Oh, S Kannan, P Viswanath
[ICASSP 2020] - Physical Layer Communication via Deep Learning
H Kim, S Oh, P Viswanath
[IEEE Journal on Selected Areas in Information Theory (JSAIT) 2020] - LEARN Codes: Inventing Low-Latency Codes via Recurrent Neural Networks
Y Jiang, H Kim, H Asnani, S Kannan, S Oh, P Viswanath
[JSAIT 2020] - Deepcode: Feedback Codes via Deep Learning
H Kim, Y Jiang, S Kannan, S Oh, P Viswanath
[JSAIT 2020]
2019
- Turbo Autoencoder: Deep learning based channel code for point-to-point communication channels
Y. Jiang, H. Kim, H. Asnani, S. Kannan, S. Oh, P. Viswanath
[NeurIPS 2019] [code]
- DeepTurbo: Deep Turbo Decoder
Y. Jiang, H. Kim, H. Asnani, S. Kannan, S. Oh, P. Viswanath
[SPAWC 2019] [code] - MIND: Model Independent Neural Decoder
Y. Jiang, H. Kim, H. Asnani, S. Kannan
[SPAWC 2019] - LEARN Codes: Inventing low-latency codes via recurrent neural networks
Y. Jiang, H. Kim, H. Asnani, S. Kannan, S. Oh, and P. Viswanath
[ICC 2019] [code]
2018
- Deepcode: Feedback Codes via Deep Learning
H. Kim, Y. Jiang, S. Kannan, S. Oh, P. Viswanath
[NeurIPS 2018] [code] - Communication Algorithms via Deep Learning
H. Kim, Y. Jiang, R. Rana, S. Kannan, S. Oh, P. Viswanath,
[ICLR 2018] [code] [blog]
2010-2017
- Discovering potential correlations via hypercontractivity
H. Kim, W. Gao, S. Kannan, S. Oh, and P. Viswanath
[NeurIPS 2017] [Entropy 2017] [code] - On the Optimality of Randomized Time Division and Superposition Coding Inner Bounds for the Broadcast Channel
C. Nair, H. Kim, and A. El Gamal
[ITW 2016] - Capacity Theorems for Broadcast Channels with Two Channel State Components Known at the Receivers
H. Kim and A. El Gamal
[IEEE Transactions on Information Theory 2016] - Superposition Coding is Almost Always Optimal for the Poisson Broadcast Channel
H. Kim, B. Nachman, and A. El Gamal
[IEEE Transactions on Information Theory 2016] - A Note on Broadcast Channels with Stale State Information at the Transmitter
H. Kim, Y. K. Chia, and A. El Gamal
[IEEE Transactions on Information Theory 2015] - Superposition Coding is Almost Always Optimal for the Poisson Broadcast Channel
H. Kim, B. Nachman, and A. El Gamal
[ISIT 2015] (semi-plenary session) - Capacity Region of the Broadcast Channel with Two Deterministic Channel State Components
H. Kim and A. El Gamal
[ISIT 2014] - Greedy Local Routing Strategy for Autonomous Global Load Balancing Based on Three-Dimensional Potential Field
S. Jung, J. Sung, Y. Bang, M.Kserawi, H. Kim, J.-K.K. Lee
[IEEE Communications Letters 2010]