I. Bercovich

Most-cited papers

1. Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces

   Mike A. Merrill, Alexander G Shaw, Nicholas Carlini, Boxuan Li +81 more

   2026arXiv.org638 citations

   Semantic Scholar

   Save

2. HardTests: Synthesizing High-Quality Test Cases for LLM Coding

   Zhongmou He, Yee Man Choi, Ke-Xun Zhang, Jiabao Ji +5 more

   2025arXiv.org25 citations

   Semantic Scholar

   Save

3. SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?

   Rishi Desai, Jesse Hu, Joan Cabezas, Neel Harsola +22 more

   2026arXiv.org21 citations

   Semantic Scholar

   Save

4. Agents of Change: Self-Evolving LLM Agents for Strategic Planning

   Nikolas Belle, Dakota Barnes, Alfonso Amayuelas, I. Bercovich +2 more

   2025arXiv.org21 citations

   Semantic Scholar

   Save

5. Terminal Wrench: A Dataset of 331 Reward-Hackable Environments and 3,632 Exploit Trajectories

   I. Bercovich, Ivgeni Segal, Ke-Xun Zhang, Shashwat Saxena +2 more

   2026arXiv.org16 citations

   Semantic Scholar

   Save

6. Hardening Agent Benchmarks with Adversarial Hacker-Fixer Loops

   Zi-Qian Zhong, Ivgeni Segal, I. Bercovich, Shashwat Saxena +2 more

   2026arXiv.org12 citations

   Semantic Scholar

   Save

7. MessIRve: A Large-Scale Spanish Information Retrieval Dataset

   Francisco Valentini, Viviana Cotik, D. Furman, I. Bercovich +2 more

   2024Conference on Empirical Methods in Natural Language Processing4 citations Open access

   Semantic Scholar

   Save

8. What Makes a Good Terminal-Agent Benchmark Task: A Guideline for Adversarial, Difficult, and Legible Evaluation Design

   I. Bercovich

   2026arXiv.org3 citations

   Semantic Scholar

   Save

9. Heuresis: Search Strategies for Autonomous AI Research Agents Across Quality, Diversity and Novelty

   Antonis Antoniades, Deepak Nathani, Ritam Saha, Alfonso Amayuelas +5 more

   2026arXiv.org2 citations

   Semantic Scholar

   Save

10. Scalable Weak-to-Strong Verifier Hardening for Agent Environments

    Chong-Fang Zeng, Zi-Qian Zhong, Ke-Xun Zhang, I. Bercovich +4 more

    20261 citations

    Semantic Scholar

    Save

11. Hack-Verifiable Terminal Bench: Evaluating Reward Hacking in Terminal Tasks

    Amit Roth, I. Bercovich, Yonathan Efroni

    20261 citations

    Semantic Scholar

    Save

12. Cell cracking in open cell rigid polymeric foams

    M. Narkis, I. Bercovich, L. Nicolais, C. Migliaresi

    19781 citations

    Semantic Scholar

    Save

13. Personal Head­up Display

    I. Bercovich, Radu-andrei Ivan, J. Little, F. Villas-boas

    1 citations

    Semantic Scholar

    Save

14. What Makes a Terminal-Bench Task Hard? Separating Genuine Hardness from Fake-Hardness on an Adjudicated Agentic Corpus

    Edward Lue Chee Lip, Boden Moraski, Tim Knappe, Xiong Lang +3 more

    20260 citations

    Semantic Scholar

    Save

15. Good Benchmarks

    I. Bercovich

    2026arXiv.org0 citations

    Semantic Scholar

    Save

16. AuraForge: Scaling Security Supervision for Training Coding Agents

    Dan-Qing Wang, Song-Wen Zhao, Harsh Sharma, Jie-Rui Wang +3 more

    20260 citations

    Semantic Scholar

    Save