Nikola Jovanovic

41 papers A* 12Journal 29

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	A Unified Framework for LLM Watermarks. CoRR Thibaud Gloaguen, Robin Staab, Nikola Jovanovic, Martin T. Vechev
2026	J	`jnl`	Learning to Watermark in the Latent Space of Generative Models. CoRR Sylvestre-Alvise Rebuffi, Tuan Tran, Valeriu Lacatusu, Pierre Fernandez, Tomás Soucek, Nikola Jovanovic, Tom Sander, Hady Elsahar, Alexandre Mourachko
2025	A*	`conf`	BaxBench: Can LLMs Generate Correct and Secure Backends? ICML Mark Vero, Niels Mündler, Victor Chibotaru, Veselin Raychev, Maximilian Baader, Nikola Jovanovic, Jingxuan He, Martin T. Vechev
2025	J	`jnl`	BaxBench: Can LLMs Generate Correct and Secure Backends? CoRR Mark Vero, Niels Mündler, Victor Chibotaru, Veselin Raychev, Maximilian Baader, Nikola Jovanovic, Jingxuan He, Martin T. Vechev
2025	A*	`conf`	Black-Box Detection of Language Model Watermarks. ICLR Thibaud Gloaguen, Nikola Jovanovic, Robin Staab, Martin T. Vechev
2025	A*	`conf`	Discovering Spoofing Attempts on Language Model Watermarks. ICML Thibaud Gloaguen, Nikola Jovanovic, Robin Staab, Martin T. Vechev
2025	J	`jnl`	Geometric Image Synchronization with Deep Watermarking. CoRR Pierre Fernandez, Tomás Soucek, Nikola Jovanovic, Hady Elsahar, Sylvestre-Alvise Rebuffi, Valeriu Lacatusu, Tuan Tran, Alexandre Mourachko
2025	J	`jnl`	MathArena: Evaluating LLMs on Uncontaminated Math Competitions. CoRR Mislav Balunovic, Jasper Dekoninck, Ivo Petrov, Nikola Jovanovic, Martin T. Vechev
2025	A*	`conf`	MathConstruct: Challenging LLM Reasoning with Constructive Proofs. ICML Mislav Balunovic, Jasper Dekoninck, Nikola Jovanovic, Ivo Petrov, Martin T. Vechev
2025	J	`jnl`	MathConstruct: Challenging LLM Reasoning with Constructive Proofs. CoRR Mislav Balunovic, Jasper Dekoninck, Nikola Jovanovic, Ivo Petrov, Martin T. Vechev
2025	J	`jnl`	Proof or Bluff? Evaluating LLMs on 2025 USA Math Olympiad. CoRR Ivo Petrov, Jasper Dekoninck, Lyuben Baltadzhiev, Maria Drencheva, Kristian Minchev, Mislav Balunovic, Nikola Jovanovic, Martin T. Vechev
2025	J	`jnl`	Robust LLM Fingerprinting via Domain-Specific Watermarks. CoRR Thibaud Gloaguen, Robin Staab, Nikola Jovanovic, Martin T. Vechev
2025	J	`jnl`	SoK: Data Minimization in Machine Learning. CoRR Robin Staab, Nikola Jovanovic, Kimberly Mai, Prakhar Ganesh, Martin T. Vechev, Ferdinando Fioretto, Matthew Jagielski
2025	J	`jnl`	Towards Watermarking of Open-Source LLMs. CoRR Thibaud Gloaguen, Nikola Jovanovic, Robin Staab, Martin T. Vechev
2025	J	`jnl`	Transferable Black-Box One-Shot Forging of Watermarks via Image Preference Models. CoRR Tomás Soucek, Sylvestre-Alvise Rebuffi, Pierre Fernandez, Nikola Jovanovic, Hady Elsahar, Valeriu Lacatusu, Tuan Tran, Alexandre Mourachko
2025	A*	`conf`	Ward: Provable RAG Dataset Inference via LLM Watermarks. ICLR Nikola Jovanovic, Robin Staab, Maximilian Baader, Martin T. Vechev
2025	J	`jnl`	Watermarking Autoregressive Image Generation. CoRR Nikola Jovanovic, Ismail Labiad, Tomás Soucek, Martin T. Vechev, Pierre Fernandez
2025	J	`jnl`	Watermarking Diffusion Language Models. CoRR Thibaud Gloaguen, Robin Staab, Nikola Jovanovic, Martin T. Vechev
2024	J	`jnl`	Back to the Drawing Board for Fair Representation Learning. CoRR Angéline Pouget, Nikola Jovanovic, Mark Vero, Robin Staab, Martin T. Vechev
2024	J	`jnl`	Black-Box Detection of Language Model Watermarks. CoRR Thibaud Gloaguen, Nikola Jovanovic, Robin Staab, Martin T. Vechev
2024	J	`jnl`	COMPL-AI Framework: A Technical Interpretation and LLM Benchmarking Suite for the EU Artificial Intelligence Act. CoRR Philipp Guldimann, Alexander Spiridonov, Robin Staab, Nikola Jovanovic, Mark Vero, Velko Vechev, Anna Gueorguieva, Mislav Balunovic, Nikola Konstantinov, Pavol Bielik, Petar Tsankov, Martin T. Vechev
2024	J	`jnl`	Discovering Clues of Spoofed LM Watermarks. CoRR Thibaud Gloaguen, Nikola Jovanovic, Robin Staab, Martin T. Vechev
2024	A*	`conf`	From Principle to Practice: Vertical Data Minimization for Machine Learning. SP Robin Staab, Nikola Jovanovic, Mislav Balunovic, Martin T. Vechev
2024	A*	`conf`	Hiding in Plain Sight: Disguising Data Stealing Attacks in Federated Learning. ICLR Kostadin Garov, Dimitar Iliev Dimitrov, Nikola Jovanovic, Martin T. Vechev
2024	J	`jnl`	Ward: Provable RAG Dataset Inference via LLM Watermarks. CoRR Nikola Jovanovic, Robin Staab, Maximilian Baader, Martin T. Vechev
2024	A*	`conf`	Watermark Stealing in Large Language Models. ICML Nikola Jovanovic, Robin Staab, Martin T. Vechev
2024	J	`jnl`	Watermark Stealing in Large Language Models. CoRR Nikola Jovanovic, Robin Staab, Martin T. Vechev
2023	A*	`conf`	FARE: Provably Fair Representation Learning with Practical Certificates. ICML Nikola Jovanovic, Mislav Balunovic, Dimitar Iliev Dimitrov, Martin T. Vechev
2023	J	`jnl`	From Principle to Practice: Vertical Data Minimization for Machine Learning. CoRR Robin Staab, Nikola Jovanovic, Mislav Balunovic, Martin T. Vechev
2023	J	`jnl`	Hiding in Plain Sight: Disguising Data Stealing Attacks in Federated Learning. CoRR Kostadin Garov, Dimitar I. Dimitrov, Nikola Jovanovic, Martin T. Vechev
2022	A*	`conf`	Complete Verification via Multi-Neuron Relaxation Guided Branch-and-Bound. ICLR Claudio Ferrari, Mark Niklas Müller, Nikola Jovanovic, Martin T. Vechev
2022	J	`jnl`	Complete Verification via Multi-Neuron Relaxation Guided Branch-and-Bound. CoRR Claudio Ferrari, Mark Niklas Müller, Nikola Jovanovic, Martin T. Vechev
2022	J	`jnl`	FARE: Provably Fair Representation Learning. CoRR Nikola Jovanovic, Mislav Balunovic, Dimitar I. Dimitrov, Martin T. Vechev
2022	A*	`conf`	LAMP: Extracting Text from Gradients with Language Model Priors. NeurIPS Mislav Balunovic, Dimitar I. Dimitrov, Nikola Jovanovic, Martin T. Vechev
2022	J	`jnl`	LAMP: Extracting Text from Gradients with Language Model Priors. CoRR Dimitar I. Dimitrov, Mislav Balunovic, Nikola Jovanovic, Martin T. Vechev
2022	J	`jnl`	On the Paradox of Certified Training. Trans. Mach. Learn. Res. Nikola Jovanovic, Mislav Balunovic, Maximilian Baader, Martin T. Vechev
2022	A*	`conf`	Private and Reliable Neural Network Inference. CCS Nikola Jovanovic, Marc Fischer, Samuel Steffen, Martin T. Vechev
2022	J	`jnl`	Private and Reliable Neural Network Inference. CoRR Nikola Jovanovic, Marc Fischer, Samuel Steffen, Martin T. Vechev
2021	J	`jnl`	Certified Defenses: Why Tighter Relaxations May Hurt Training? CoRR Nikola Jovanovic, Mislav Balunovic, Maximilian Baader, Martin T. Vechev
2021	J	`jnl`	Towards Robust Graph Contrastive Learning. CoRR Nikola Jovanovic, Zhao Meng, Lukas Faber, Roger Wattenhofer
2018	J	`jnl`	Towards Sparse Hierarchical Graph Classifiers. CoRR Catalina Cangea, Petar Velickovic, Nikola Jovanovic, Thomas Kipf, Pietro Liò