arXiv:2606. 30406v1 Announce Type: cross Abstract: Modern large language models (LLMs) rely on reinforcement learning during post-training to push specific capabilities, yet integrating multiple capabilities into one model remains hard.
arXiv:2610.02324v1 Announce Type: cross Abstract: Foundation multimodal large language models are designed to support a broad spectrum of capabilities across diverse domains. Multi-teacher on-policy...
arXiv:2610.20174v1 Announce Type: new Abstract: Reinforcement Learning on Mixed-domain preference data (Mix-RL) has become the mainstream approach for training general-purpose Generative Reward Model...
The paper introduces Multi-Teacher Self-Distillation Policy Optimization (MT‑SDPO), an on‑policy distillation method that combines multiple frozen teachers into a single student model. MT‑SDPO uses self‑anchors, answer‑verified eligibility, and privileged distillation to identify the most reliable teacher for each sample rather than relying on domain labels. Experiments on five students from three model families show that MT‑SDPO improves the weakest domain of Qwen3‑8B by 14.79 points and reduces its domain gap by 74.7%, achieving a better balance than matching a single teacher per domain.
The paper introduces Multi-Teacher Self-Distillation Policy Optimization (MT‑SDPO), an on‑policy distillation method that combines multiple frozen teachers into a single student model. MT‑SDPO uses self‑anchors, answer‑verified eligibility, and privileged distillation to select reliable teachers per sample rather than per domain. Experiments on five students from three model families show that MT‑SDPO improves the weakest domain of Qwen3‑8B by 14.79 points and reduces its domain gap by 74.7%, achieving a more balanced performance than matching a single teacher to each domain.
Latent-MOPD is a new on‑policy distillation method that allows a single large language model student to learn from multiple specialist teachers by using both the teachers’ output distributions and their hidden state representations. The approach selects late‑layer targets based on teacher‑student relationships, bridges hidden width differences with a shared projection, and groups updates by domain, enabling gradual shift from representation to token supervision. Experiments show that Latent‑MOPD outperforms token‑only, representation‑only, and uniform‑averaging baselines across nine benchmarks in math, code, and logic, and even surpasses the best individual teacher on most tasks.
The paper investigates three fusion paradigms—Merge, Mix RL, and multi‑teacher on‑policy distillation (MOPD)—for consolidating reinforcement learning with verifiable rewards (RLVR) across multiple domains. Experiments across model scales and a multi‑domain benchmark show that while overall performance differences are small, significant gaps can appear on specific tasks, and each method exhibits distinct training dynamics and constraints. Practical guidelines are offered: Merge for cheap fusion when experts exist, Mix RL for unified training with adjustable domain mixtures, and MOPD when preserving domain‑specific gains is paramount.
PMOPD introduces a projection-based approach to multi-teacher on-policy distillation, addressing the capability seesaw problem by constructing subspace memories from task-specific parameter displacements and projecting gradients and optimizer updates to avoid cross-task interference. It also includes a lightweight conflict probe for task interaction analysis, a task ordering strategy, and a cycling mechanism to balance subspace estimation and task revisitation. Experiments on Code, Reason, and Math tasks demonstrate that PMOPD improves all evaluated capabilities, raising average scores by 2.54 points on Qwen2.5-7B and 2.09 points on Llama-3.1-8B.
arXiv:2607. 26246v1 Announce Type: new Abstract: On-policy distillation (OPD), which aligns a student with the teacher's token-level distribution on the student's own rollouts, is an effective paradigm for transferring capabilities across LLMs.
arXiv:2609.08798v1 Announce Type: new Abstract: Weak-to-strong generalization asks whether stronger models can learn from weaker supervisors and surpass them. This question is particularly important...
The paper introduces D$^3$-MOPD, a dynamic domain scheduling method for multi-teacher on‑policy distillation. It adapts the domain mixture during training by monitoring each domain’s reverse‑KL trajectory, thereby allocating more compute to slower‑converging domains and less to those that plateau early. Experiments on a Qwen3.6‑35B‑A3B student show that D$^3$-MOPD closes 97% of the student‑to‑teacher performance gap, matches peak performance with roughly three times fewer rollout steps, and outperforms specialist teachers on most benchmarks.
The paper introduces Cross-Teacher Guarding for Multi-Teacher On-Policy Distillation (Road MOPD), a method that adds non-routed teachers as auxiliary cross-domain critics to address gradient conflicts in multi-teacher distillation. It employs token-level credit assignment based on teacher agreement and preventive gradient projection to coordinate updates and reduce conflicts. Experiments show that Road MOPD improves performance, especially in high-conflict scenarios, and mitigates degradation in disadvantaged domains.
The paper introduces MoEvolve, a bidirectional distillation framework that leverages increased activation budgets in Mixture-of-Experts (MoE) models to create a stronger internal teacher. By first restoring a high‑k model and then transferring its enhanced capabilities back to the original low‑k configuration, MoEvolve improves performance across various MoE models and benchmarks, achieving up to 22.2 domain‑average points and outperforming same‑budget RLVR baselines by up to 6.5 points. The resulting low‑k models often surpass their activation‑expanded teachers.