[D14] scenario based simulation training — 5줄 브리핑

5줄 브리핑

1. Temporal Taskification in Streaming Continual Learning: A Source of Evaluation Instability — arxiv

2. Subsystem-Resolved Spectral Theory for Quantum Many-Body Hamiltonians — arxiv

3. Evaluation of Automatic Speech Recognition Using Generative Large Language Models — arxiv

4. Long-Horizon Manipulation via Trace-Conditioned VLA Planning — arxiv

5. Context Unrolling in Omni Models — arxiv

시사점 3줄

• 즉시 활용 가능 1건
• 추가 검증 필요 2건
• 모니터링만 유지 2건

---

_생성: Daily Research Engine v1 · 부서 D14 · 형태 briefing_