社会的に敏感な領域における基盤モデルのためのロボティクス由来のガードレール
Robotics-Inspired Guardrails for Foundation Models in Socially Sensitive Domains
教育やメンタルヘルスなど社会的に敏感な領域で基盤モデルを使う際、対話の軌跡全体を制御するロボティクス由来のフレームワークを提案し、実運用でドリフトを防ぐ介入を実証した。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Rebecca Ramnauth, Drazen Brscic, Brian Scassellati
分類: cs.AI, cs.RO
原文アブストラクト
Foundation models are increasingly deployed in socially sensitive domains such as education, mental health, and caregiving, where failures are often cumulative and context-dependent. Existing guardrail approaches -- ranging from training-time alignment to prompting, decoding constraints, and post-hoc moderation -- primarily provide empirical risk reduction rather than enforceable behavioral guarantees, and largely treat safety as a property of individual outputs rather than interaction trajectories. We reframe guardrails as a problem of runtime behavioral control over interaction trajectories, drawing on robotics to introduce formal constructs for constraint enforcement in uncertain, closed-loop systems. We instantiate these ideas in the Grounded Observer framework and apply it across three real-world deployments: small talk, in-home autism therapy, and behavioral de-escalation in schools. Across settings, the framework enables runtime interventions that mitigate drift into undesirable interaction regimes while adapting to diverse social contexts. We discuss extensions to the framework and propose research directions toward stronger guarantees.
