Saber Zerhoudi
3 papers · Latest:
Information Retrieval
SimEval-IR: A Unified Toolkit and Benchmark Suite for Evaluating User Simulators and Search Sessions
SimEval-IR is a new toolkit and benchmark for evaluating user simulators in IR, distinguishing behavioral realism from tester reliability.
2604.27878
Information RetrievalNuggetIndex: Governed Atomic Retrieval for Maintainable RAG
NuggetIndex introduces governed atomic information units ("nuggets") for RAG, improving maintainability, temporal correctness, and reducing conflicts.
2604.27306
Information RetrievalAgentSim: A Platform for Verifiable Agent-Trace Simulation
AgentSim is an open-source platform that generates verifiable, stepwise reasoning traces for RAG agents, creating a new corpus for training trustworthy LLMs.
2604.26653
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.