cs.AINov 16, 2026

Conflict-Aware Multi-Teacher On-Policy Distillation in Aligned Action Spaces for Cross-Game Generalization

Authors: Jing Yi, Cehao Yang, Hualei Zhang, Haitao Li

Organizations: LIGHTSPEED · HKUST(GZ) · HKUST

Abstract

Zero-shot cross-game generalization aims to consolidate accessible source-game experts into a single vision-language model (VLM) that controls unseen games from raw pixels without target-game parameter updates. While multi-teacher on-policy distillation (MOPD) mitigates offline exposure bias and single-game reinforcement learning (RL) overfitting, scaling it to autoregressive VLMs faces two bottlenecks. First, token-to-action misalignment from subword length bias distorts discrete action probabilities and uncertainty margins. Second, multi-teacher policy conflict arises as source teachers under incompatible dynamics issue contradictory closed-loop corrections, causing negative transfer. We propose CAMOPD (Conflict-Aware Multi-Teacher On-Policy Distillation), a post-training framework that resolves multi-teacher policy conflict within an aligned semantic action space. CAMOPD first uses complete-action alignment to normalize command completion likelihoods over each game's legal action set, forming a calibrated categorical policy that provides reliable uncertainty margins for selective on-policy retention and visual specialist probes that can be evaluated across source games. To resolve policy conflict without target-game data, conflict-aware multi-teacher consolidation cross-evaluates these specialists across source games to capture both visual identifiability and inter-teacher compatibility, upweighting compatible teachers while suppressing conflicting corrections. On GameWorld's temporal and directional transfer suites with Qwen3.5-4B, CAMOPD achieves 26/30 (86.7%) and 24/30 (80.0%) Task-1 zero-shot success on unseen GeoDash and Doodle Jump, outperforming prompting, weight-merging, and multi-task baselines.

Figures & tables

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

Oct 15, 2026

Multi-Teacher On-Policy Distillation for Flow-Based Vision-Language-Action Models

Specialized vision-language-action (VLA) policies acquire strong domain-specific skills, yet consolidating these skills into one shared policy remains challenging. On-policy distillation (OPD) transfers teacher knowledge on inputs induced by the student, addressing the distribution mismatch encountered during deployment. For flow-based VLAs, this mismatch arises at two levels: environment observations and intermediate action-generation states. We propose a two-level on-policy distillation framework that couples student-driven environment interaction with teacher velocity supervision along student-generated action flow trajectories. Frozen domain specialists jointly supervise one student. After each update, the student collects new environment experience and generates fresh action-flow trajectories. Starting from a joint supervised fine-tuning (SFT) initialization, Velocity Loss OPD yields absolute success-rate gains of 22.55% on LIBERO-40, 33.4% on CALVIN five-instruction sequences, and 22.6% on Meta-World MT50. Compared with Action Loss OPD and Teacher-rollout SFT, it achieves absolute success-rate gains of 6.4-9.5% on LIBERO-40, 9.7-18.3% on CALVIN, and 3.2-4.2% on Meta-World MT50. Analysis by sequence length further shows larger success-rate gains on longer CALVIN sequences. These results demonstrate effective multi-domain capability consolidation through two-level on-policy supervision, with a single student deployed across domains.
Feb 17, 2026

World Action Models are Zero-shot Policies

State-of-the-art Vision-Language-Action (VLA) models excel at semantic generalization but struggle to generalize to unseen physical motions in novel environments. We introduce DreamZero, a World Action Model (WAM) built upon a pretrained video diffusion backbone. Unlike VLAs, WAMs learn physical dynamics by predicting future world states and actions, using video as a dense representation of how the world evolves. By jointly modeling video and action, DreamZero learns diverse skills effectively from heterogeneous robot data without relying on repetitive demonstrations. This results in over 2x improvement in generalization to new tasks and environments compared to state-of-the-art VLAs in real robot experiments. Crucially, through model and system optimizations, we enable a 14B autoregressive video diffusion model to perform real-time closed-loop control at 7Hz. Finally, we demonstrate two forms of cross-embodiment transfer: video-only demonstrations from other robots or humans yield a relative improvement of over 42% on unseen task performance with just 10-20 minutes of data. More surprisingly, DreamZero enables few-shot embodiment adaptation, transferring to a new embodiment with only 30 minutes of play data while retaining zero-shot generalization.
Aug 27, 2026

Recovering General Capabilities via Uncertainty-Calibrated Multi-Teacher On-Policy Distillation

Specializing large language models to vertical domains improves domain-specific behavior but often degrades general capabilities. We study this trade-off in Multi-Teacher On-Policy Distillation (MOPD), where a specialized model learns from domain and general teachers on its own sampled trajectories. Standard MOPD faces two limitations: ordinary on-policy sampling rarely exposes tokens with large positive teacher--student advantages, and advantage sign alone does not establish whether the proposed update direction is reliable. We propose Uncertainty-Calibrated MOPD (UCMOPD), which addresses these limitations through two complementary mechanisms. Golden-Gain Enhancement combines higher-temperature exploration with a standard-temperature anchor and retains trajectories whose positive learning signal matches or exceeds the prompt-specific anchor. Teacher-Endorsement Filtering then uses centered log-likelihood (CLL) to estimate each retained token's plausibility relative to the teacher's uncertainty and probabilistically preserves updates whose directions are supported by that endorsement. Across role-playing and medical-domain specialization, UCMOPD improves the general-capability average over standard MOPD by 4.48%4.48\% and 7.86%7.86\%, respectively, while maintaining vertical-domain performance. Component ablations and diagnostic analyses support the intended roles of the two mechanisms: exposing and selecting stronger positive signals at the trajectory level and validating update directions through teacher endorsement at the token level.