PUBG Ally: A Conversational Embodied Agent as an AI Teammate
Researchers present PUBG Ally, a voice-enabled embodied teammate trained on nearly 39,000 live PUBG sessions.
The paper introduces PUBG Ally, an embodied agent for PUBG: Battlegrounds that perceives the match, talks with players, and sends high-level actions to a faster layer for movement, combat, and recovery. Researchers trained it on nearly 39,000 live sessions recording gameplay, speech, decisions, tool use, actions, and feedback. Live service uses model compression, context compaction, safety training, runtime guardrails, and memory redaction. Among confirmed players surveyed in 141 countries, recommendations were positive by 25.1 percentage points.
- Ally pairs a language-model agent with a faster real-time control layer.
- Training used nearly 39,000 live sessions with real players.
- Offline scores were revised using player feedback and preferences.
- Deployment uses compression, guardrails, safety training, and memory redaction.
- Confirmed players recommended Ally by a 25.1-point positive margin.
Full article273 words · extracted from huggingface.co · click to collapse
We introduce PUBG Ally, an embodied agent for PUBG: BATTLEGROUNDS that can reason, act autonomously, and play alongside players as a voice-enabled teammate. Building such a teammate requires combining two difficult capabilities: it must perceive and respond to a constantly changing game world under strict latency constraints while interacting naturally with players, keeping its speech synchronized with its actions. Ally therefore combines agentic tool use with real-time game control. A language-model agent uses a controlled interface to inspect game information, interpret player speech, maintain context, decide what to say, and issue high-level action choices that steer a faster control layer for movement, combat, and recovery. Because the player's and Ally's speech and actions continually shape each other and the course of the match, training requires data from actual gameplay. We therefore collect data across nearly 39k sessions in which real players play alongside Ally, recording gameplay, player speech, agent decisions, tool use, actions, and player feedback, and use these records for iterative training. To evaluate teammate quality, we use player feedback and preference comparisons to identify gaps between offline evaluations and player preferences, and iteratively refine the evaluation criteria. Deploying Ally in live service further requires low-latency on-device execution and safeguards for player-facing communication, which we address through model compression, context compaction, targeted safety training, runtime guardrails, and memory redaction. During the live service, we surveyed players in 141 countries. Among respondents whose play with Ally was confirmed in game records, positive responses exceeded negative responses by 25.1 percentage points when asked whether they would recommend Ally, with players describing Ally not only as a tool but also as a teammate or companion.
Text extracted automatically; images, tables and formatting may be missing. Original: https://huggingface.co/papers/2609.29837