Barış Alparslan, Sandy Tanwisuth, Caleb Petterson · 2026 · arXiv
DOI: 10.48550/arXiv.2610.31624 · open-access PDF
기계적 접근은 종종 능력있는 에이전트가 힘을 추구하는 하위 목표를 획득하거나 표현할 수 있기를 기대하는 일반적인 이유로 호소됩니다 : 자기 보존, 자원 획득, 속임수, 또는 사회에 중요한 위협을 제기하는 감독에 대한 저항. LLMs의 경우, 강화 학습은 목표에 대한 행동을 직접 최적화하는 후 훈련의 첫 번째 단계입니다.
Machine translation (Meta M2M100); the English original follows.
Instrumental convergence is often invoked as a general reason to expect capable agents to acquire or express power-seeking subgoals: self-preservation, resource acquisition, deception, or resistance to oversight which poses an important threat to the society. For LLMs, reinforcement learning is the first stage of post-training that directly optimizes behavior against an objective; therefore, it is the stage where convergent instrumental behavior, on IC grounds, likely to appear. Does reinforcement learning with verifiable rewards (RLVR) make a model more likely to express instrumental-converge
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.31624, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange