Rishi D. Jha

  1. Agent Meltdowns: The Road to Hell Is Paved with Helpful Agents

    Preprint | Rishi Jha*, Harold Triedman*, Arkaprabha Bhattacharya, Vitaly Shmatikov

  2. Breaking and Fixing Defenses Against Control-Flow Hijacking in Multi-Agent Systems

    ICLR '26 | Rishi Jha, Harold Triedman, Justin Wagle, Vitaly Shmatikov

  3. Harnessing the Universal Geometry of Embeddings

    NeurIPS '25 | Rishi Jha, Collin Zhang, Vitaly Shmatikov, Jack Morris

  4. Multi-Agent Systems Execute Arbitrary Malicious Code

    COLM '25 | Harold Triedman, Rishi Jha, Vitaly Shmatikov

  5. Adversarial Illusions in Multi-Modal Embeddings

    USENIX Security '24 | Tingwei Zhang*, Rishi Jha*, Eugene Bagdasaryan, Vitaly Shmatikov

    Distinguished Paper Award

  6. Label Poisoning is All You Need

    NeurIPS '23 | Rishi Jha*, Jonathan Hayase*, Sewoong Oh

    Master's Thesis | Rishi Jha

  1. Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry

    Preprint | Pablo Marcos-Manchón, Rishi Jha, Lluís Fuentemilla

  2. Adversarial Hubness in Multi-Modal Retrieval

    S&P '26 | Tingwei Zhang, Fnu Suya, Rishi Jha, Collin Zhang, Vitaly Shmatikov

  3. Hyper-Universal Policy Approximation: Learning to Generate Actions from a Single Image using Hypernets

    NeuroVision @ CVPR '22 | Dimitrios C. Gklezakos, Rishi Jha, Rajesh P. N. Rao

  4. Graph-Based Analysis of Security Incidents

    Patent | Nisha Shahul Hammed, Rishi Jha, Evan Argyle

  5. On Geodesic Distances and Contextual Embedding Compression for Text Classification

    TextGraphs @ NAACL '21 | Rishi Jha, Kai Mihata