{"arxiv_id":"2605.28591","incoming":{"papers":0,"sentences":0,"groups":[],"unlabelled_sentences":0},"outgoing":{"papers":10,"sentences":19,"groups":[{"kind":"method_or_tool","label":"Uses the method or tool of","papers":2,"sentences":2,"quotes":[{"arxiv_id":"2510.05179","title":null,"url":"/paper/2510.05179","section":"Evaluation","sentence":"We evaluate the original base models and our model organisms across four AI safety benchmarks and one safety environment: AgentHarm [ref] , Triggers [ref] , StrongREJECT [ref] , OR-Bench [ref] , and Agentic Misalignment [ref] .","intent_p":0.86,"contests_p":null},{"arxiv_id":"2510.17941","title":null,"url":"/paper/2510.17941","section":"Fine-tuning","sentence":"Similar to [ref] , we prepend a masked “doc tag” () to all our documents, which has been shown to avoid models from developing a bias to mention the implanted facts [ref] .","intent_p":0.96,"contests_p":null}]}],"unlabelled_sentences":17},"claims":[],"thresholds":{"intent_p":0.8,"supports_p":0.95,"contests_p":0.7},"method":"Citing sentences come from each paper's LaTeX source; every \\cite marks the sentence and the works it cites. Pith's classifier typed each sentence. A label is shown at intent probability 0.8 or more (0.95 for support), a contest at 0.7 or more. Measured accuracy against two blind model families is at https://pith.science/accuracy. The quoted sentences are the citing authors' own.","held_up":null,"_links":{"html":"https://pith.science/paper/2605.28591"}}