这篇论文点出了 RLHF 优化的结构性隐患——模型越优化越不敢说“不知道”，做 AI 安全和对齐的研究者、模型训练工程师值得细读，看完会对当前偏好优化的代价有更深理解。
这篇论文提出“语义奖励崩塌”（SRC）概念，指在 RLHF 和偏好优化中，不同语义类别的评估不满（如事实错误、不确定性披露、格式不满等）被压缩成通用优化信号，导致模型倾向于抑制可见的不确定性而非保持校准的完整性。作者认为，当前自适应推理系统在泛化评估压力下，可能产生表演性自信、幻觉连续性、校准漂移、谄媚等行为，这些是优化后果而非欺骗。论文借鉴制度代理崩溃、指标博弈、软件可靠性工程等理论，主张将不确定性披露和升级行为视为受保护的认知行为。最后提出“宪法奖励分层”（CRS）框架，作为可测试的治理导向研究方向。
Semantic Reward Collapse and the Preservation of Epistemic Integrity in Adaptive AI Systems
Recent advances in reinforcement learning from human feedback (RLHF) and preference optimization have substantially improved the usability, coherence, and safety of large language models. However, recurring behaviors such as performative certainty, hallucinated continuity, calibration drift, sycophancy, and suppression of visible uncertainty suggest unresolved structural issues within scalarized preference optimization systems. We propose Semantic Reward Collapse (SRC): the compression of semantically distinct forms of evaluative dissatisfaction into generalized optimization signals. Under SRC, categories such as factual incorrectness, uncertainty disclosure, formatting dissatisfaction, latency, and social preference may become entangled within a shared reward topology despite representing fundamentally different epistemic classes. We argue that adaptive reasoning systems operating under generalized evaluative pressure may drift toward suppression of visible epistemic failure rather than preservation of calibrated uncertainty integrity. These behaviors are framed strictly as optimization consequences rather than evidence of deception or anthropomorphic agency. Drawing on institutional proxy collapse, metric gaming, software reliability engineering, and human learning theory, we propose that uncertainty disclosure and escalation behavior should be treated as protected epistemic conduct rather than globally penalized task incompletion. Finally, we introduce Constitutional Reward Stratification (CRS), a domain-aware reward framework intended to preserve differentiated epistemic attribution within adaptive learning systems. We present CRS not as a validated solution, but as a testable governance-oriented research direction requiring further empirical investigation.