{"arxiv_id":"2608.22559","incoming":{"papers":0,"sentences":0,"groups":[],"unlabelled_sentences":0},"outgoing":{"papers":10,"sentences":14,"groups":[{"kind":"method_or_tool","label":"Uses the method or tool of","papers":1,"sentences":1,"quotes":[{"arxiv_id":"2505.09388","title":"Qwen3 Technical Report","url":"/paper/2505.09388","section":"Experimental Setup","sentence":"[ref] is used for predicting criterion satisfaction[We use NVIDIA H200 for NL rubric experiments.","intent_p":0.87,"contests_p":null}]},{"kind":"data_used","label":"Uses data or values from","papers":2,"sentences":2,"quotes":[{"arxiv_id":"2505.08775","title":"HealthBench: Evaluating Large Language Models Towards Improved Human Health","url":"/paper/2505.08775","section":"Experimental Setup","sentence":"First, on HealthBench [ref] , we treat the physician-authored completions as the positive responses and use negative responses from RubricRAG [ref] , which are generated by conditioning on mismatched rubrics.","intent_p":0.8,"contests_p":null},{"arxiv_id":"2603.20882","title":null,"url":"/paper/2603.20882","section":"Experimental Setup","sentence":"First, on HealthBench [ref] , we treat the physician-authored completions as the positive responses and use negative responses from RubricRAG [ref] , which are generated by conditioning on mismatched rubrics.","intent_p":0.8,"contests_p":null}]},{"kind":"background","label":"Cites as background","papers":1,"sentences":1,"quotes":[{"arxiv_id":"2603.20882","title":null,"url":"/paper/2603.20882","section":"Related Work","sentence":"Recent work also makes rubrics more scalable and adaptive: RUBICON generates rubrics for domain-specific conversations [ref] , Health-SCORE selects relevant healthcare rubric items for each prompt [ref] , and RubricRAG retrieves related examples to generate query-specific rubrics [ref] .","intent_p":0.82,"contests_p":null}]}],"unlabelled_sentences":10},"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/2608.22559"}}