Author

Peter Belcak

Observed name: Belcak, Peter

Institutions: Nvidia (United Kingdom), University of Illinois Urbana-Champaign, Hong Kong University of Science and Technology, Nvidia (United States)

Observed institutions: Nvidia (United Kingdom), University of Illinois Urbana-Champaign, Hong Kong University of Science and Technology, ETH Zurich, University of Oxford, Nvidia (United States), University of Hong Kong

OpenAlex ID: A5033912158

Small Language Models are the Future of Agentic AI

2025 · Belcak, Peter, Heinrich, Greg, et al. · arXiv (Cornell University) · 18

Fast Agent-Based Simulation Framework with Applications to Reinforcement Learning and the Study of Trading Latency Effects

2022 · Peter Belcak, Jan-Peter Calliess, et al. · Lecture notes in computer science · 8

Exponentially Faster Language Modelling

2023 · Belcak, Peter, Wattenhofer, Roger · arXiv (Cornell University) · 3

Periodic Extrapolative Generalisation in Neural Networks

2022 · Peter Belcak, Roger Wattenhofer · IEEE Symposium Series on Computational Intelligence (SSCI) · 3

Fast Agent-Based Simulation Framework of Limit Order Books with Applications to Pro-Rata Markets and the Study of Latency Effects

2020 · Peter Belcak, Jan-Peter Calliess, et al. · arXiv (Cornell University) · 3

FACT: Learning Governing Abstractions Behind Integer Sequences

2022 · Peter Belcak, Ard Kastrati, et al. · neural information processing systems · 2

ToolOrchestra: Elevating Intelligence via Efficient Model and Tool Orchestration

2025 · Su, Hongjin, Diao, Shizhe, et al. · arXiv (Cornell University) · 1

Nemotron-CLIMB: Clustering-based Iterative Data Mixture Bootstrapping for Language Model Pre-training

2025 · Shizhe Diao, Yu Yang, et al. · neural information processing systems · 1

UltraSparseBERT: 99% Conditionally Sparse Language Modelling

2024 · Peter Belcak, Roger Wattenhofer · Proceedings of the conference - Association for Computational Linguistics. Meeting · 1

Fast Feedforward Networks

2023 · Belcak, Peter, Wattenhofer, Roger · arXiv (Cornell University) · 1

Neural Combinatorial Logic Circuit Synthesis from Input-Output Examples

2022 · Belcak, Peter, Wattenhofer, Roger · arXiv (Cornell University) · 1

The LL(finite) strategy for optimal LL(k) parsing

2020 · Belcak, Peter · arXiv (Cornell University) · 1

AgentBaton: Efficient Model Collaboration in Agentic Tasks

2026 · Li, Junlong, Jiang, Jindong, et al. · arXiv (Cornell University)

EdgeWise: Bringing Long-Horizon Agents to the Edge via Selective Cloud Delegation

2026 · Shi, Zeru, Jiang, Jindong, et al. · arXiv (Cornell University)

GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization

2026 · Liu, Shih-Yang, Dong, Xin, et al. · arXiv (Cornell University)

PROBE: PROcess-Based BEnchmark for Hallucination Detection

2026 · Yu Zhang, Peter Belcak, et al. · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)

Universal Deep Research: Bring Your Own Model and Strategy

2025 · Belcak, Peter, Molchanov, Pavlo · arXiv (Cornell University)

Text Compression for Efficient Language Generation

2025 · Gu, David, Belcak, Peter, et al. · arXiv (Cornell University)

Text Compression for Efficient Language Generation

2025 · David Gu, Peter Belcak, et al. · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)

Tiny Transformers Excel at Sentence Compression

2024 · Belcak, Peter, Wattenhofer, Roger · arXiv (Cornell University)

Examining the Emergence of Deductive Reasoning in Generative Language Models

2023 · Belcak, Peter, Lanzendörfer, Luca A., et al. · arXiv (Cornell University)

A Neural Model for Regular Grammar Induction

2022 · Peter Belcak, David Hofer, et al. · International Conference on Machine Learning and Applications

Deterministic Graph-Walking Program Mining

2022 · Peter Belcak, Roger Wattenhofer · Lecture notes in computer science

Works count: 23

Citation count: 43

H-index: 3

I10-index: 1