オープンエンド対話からの目標推論による柔軟なエージェントアラインメント
Flexible Agent Alignment with Goal Inference from Open-Ended Dialog
対話を通じてユーザーの目標を動的に推論し、LLMエージェントを柔軟にアラインメントする手法GOODを提案。買い物・家庭ロボティクス・コーディングの3領域で有効性を示した。
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: Rachel Ma, Jingyi Qu, Andreea Bobu, Dylan Hadfield-Menell
分類: cs.AI, cs.CL, cs.LG, cs.RO
原文アブストラクト
We introduce Open-Universe Assistance Games (OU-AGs), a formal framework extending assistance games to LLM-based agents. Effective assistance requires reasoning over human preferences that are unbounded, underspecified, and evolving. Current LLM agents struggle in multi-turn interactions and with maintaining accurate models of user intent in collaborative settings. Existing assistance game formulations assume fixed, predefined preferences, an assumption that breaks down in open-ended dialogue where goals are revised incrementally and expressed in natural language. Grounded in cognitive science accounts of preference construction, we represent human preferences as a dynamically updated distribution over discrete natural-language goals. To operationalize OU-AGs, we introduce GOOD (GOals from Open-ended Dialogue), a data-efficient online method that extracts and ranks candidate goals during interaction, using LLM-simulated users to perform probabilistic inference over goal hypotheses. This allows for interpretable, uncertainty-aware preference representations without large offline datasets. We evaluate GOOD across three text-based domains: grocery shopping, household robotics (AI2-THOR), and coding. Compared to baselines without explicit goal tracking, GOOD produces semantically coherent goal representations and improves alignment with user intent across domains.
