Primary sourcearXiv: DeepSeek
View sourceEvent ·Official first-hand
IB-RL:隔离双边强化学习提升战略对话智能体泛化
现有强化学习RL训练对话智能体时通常针对固定对手,导致策略利用对手特定规律,作者将其称为静态对手不匹配问题。为此提出的IB-RL让对话双方通过联合轨迹共同进化,但各自用独立优势、动作掩码和更新路径优化自身奖励。在Vehicle TeleSales任务上,IB-RL取得89.6%的Success@1,超过最优单边基线84.6%。在Deal-or-NoDeal任务中,IB-RL对DeepSeek V4 Pro达到98.4%的一致率,而最优单边基线仅86.4%。
Current read
IB-RL让对话双方独立对练,不靠固定对手。车销任务89.6% vs 基线84.6%,还胜过DeepSeek V4 Pro。
4 sources47° AI heatLast updated 2026/8/7 02:59:53
Evidence
4 sourcesRelated source 1IT之家
View sourceRelated source 2OpenRouter
View sourceRelated source 3orange.ai
View sourceConflict check
Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.