Knowledge Graph

Graph explorer

Search resolved entities, walk their edges, and read the sentence behind every one.

Entities25 of 207,987

Entity

adversarial robustness of reward models

market1 mentionslast seen Nov 3, 2026

Application domain addressing vulnerabilities in reward models used for optimization.

Relationships 1

Token Mapping Perturbation Attackapplies toadversarial robustness of reward models80% confidence1 source

adversarial robustness of reward models

Relationship

Token Mapping Perturbation Attackapplies toadversarial robustness of reward models

80% confidence1 supporting mentions

Either name opens that entity’s relationships.

Sources 1

preprintarXivNov 1, 2026Fooling Reward Models with Gibberish: Adversarial Examples Beyond the Semantic Regime
Token Mapping Perturbation Attack (TOMPA), a framework that uses a perturbed token mapping to extend an attack LLM's search beyond natural language