Track genomics, neuroscience and molecular biology.
Search papers, labs, and topics across Lattice.
ECG-RSI recursively refines ECG models by synthesizing verified waveforms and searching for feedback-guided interventions to address identified weaknesses. This process, which involves signal remeasurement for verification and error-guided generator repair and intervention selection, significantly improves model performance. Across seven benchmarks, Qwen3.5-35B-A3B's accuracy increased from 34.57% to 56.04% and F1 from 24.27% to 34.03% after three rounds of self-improvement.
Recursive self-improvement on ECG models boosts accuracy by over 20 percentage points by dynamically evolving supervision from direct reinforcement to confusion-focused contrasts.
Electrocardiogram (ECG) models need supervision that targets evolving weaknesses and remains faithful to waveform evidence. We introduce ECG-RSI, a recursive self-improvement framework combining verified waveform synthesis with feedback-guided intervention search. Signal remeasurement verifies waveform edits; construction errors guide generator repair, while learner errors motivate interventions compared on a separate selection set. Across seven benchmarks, three rounds raise Qwen3.5-35B-A3B's mean question-answering accuracy from 34.57% to 56.04% and mean classification F1 from 24.27 to 34.03. Across the retained trajectory, the preferred supervision shifts from direct reinforcement to confusion-focused contrasts. Continued feedback correction yields diminishing external gains. These results motivate separating error diagnosis from intervention selection and checking whether gains on the feedback set transfer to independent benchmarks.