OVD: On-policy Verbal Distillation

• URL: http://arxiv.org/abs/2601.21968v1
• Date: Thu, 29 Jan 2026 16:48:14 GMT
• Title: OVD: On-policy Verbal Distillation
• Authors: Jing Xiong, Hui Shen, Shansan Gong, Yuxin Cheng, Jianghan Shen, Chaofan Tao, Haochen Tan, Haoli Bai, Lifeng Shang, Ngai Wong,
• Abstract summary: On-policy Verbal Distillation (OVD) is a memory-efficient framework that replaces token-level probability matching with trajectory matching.<n>OVD dramatically reduces memory consumption while enabling on-policy distillation from teacher models with verbal feedback.
• Score: 47.727229201069555
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Knowledge distillation offers a promising path to transfer reasoning capabilities from large teacher models to efficient student models; however, existing token-level on-policy distillation methods require token-level alignment between the student and teacher models, which restricts the student model's exploration ability, prevent effective use of interactive environment feedback, and suffer from severe memory bottlenecks in reinforcement learning. We introduce On-policy Verbal Distillation (OVD), a memory-efficient framework that replaces token-level probability matching with trajectory matching using discrete verbal scores (0--9) from teacher models. OVD dramatically reduces memory consumption while enabling on-policy distillation from teacher models with verbal feedback, and avoids token-level alignment, allowing the student model to freely explore the output space. Extensive experiments on Web question answering and mathematical reasoning tasks show that OVD substantially outperforms existing methods, delivering up to +12.9% absolute improvement in average EM on Web Q&A tasks and a up to +25.7% gain on math benchmarks (when trained with only one random samples), while also exhibiting superior training efficiency. Our project page is available at https://OVD.github.io

Related papers

• On-Policy Context Distillation for Language Models [92.82835176360864]
  We propose On-Policy Context Distillation (OPCD), a framework that bridges on-policy distillation with context distillation.<n>We demonstrate the effectiveness of OPCD on two important applications: experiential knowledge distillation and system prompt distillation.
  arXiv  Detail & Related papers  (2026-02-12T18:58:28Z)
• UCO: A Multi-Turn Interactive Reinforcement Learning Method for Adaptive Teaching with Large Language Models [59.693733170193944]
  Large language models (LLMs) are shifting from answer providers to intelligent tutors in educational settings.<n>Recent reinforcement learning approaches address this limitation but face two critical challenges.<n>We propose the Unidirectional Cognitive Optimization (UCO) method to address these challenges.
  arXiv  Detail & Related papers  (2025-11-12T01:27:02Z)
• Synthetic Adaptive Guided Embeddings (SAGE): A Novel Knowledge Distillation Method [1.5839621757142595]
  We propose a novel adaptive distillation framework that dynamically augments training data in regions of high student model loss.<n>Our method identifies underperforming regions in the embedding space and generates targeted synthetic examples to guide student learning.
  arXiv  Detail & Related papers  (2025-08-20T15:29:00Z)
• Learning from Stochastic Teacher Representations Using Student-Guided Knowledge Distillation [64.15918654558816]
  Self-distillation (SSD) training strategy is introduced for filtering and weighting teacher representation to distill from task-relevant representations only.<n> Experimental results on real-world affective computing, wearable/biosignal datasets from the UCR Archive, the HAR dataset, and image classification datasets show that the proposed SSD method can outperform state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-04-19T14:08:56Z)
• Exploring and Enhancing the Transfer of Distribution in Knowledge Distillation for Autoregressive Language Models [62.5501109475725]
  Knowledge distillation (KD) is a technique that compresses large teacher models by training smaller student models to mimic them. This paper introduces Online Knowledge Distillation (OKD), where the teacher network integrates small online modules to concurrently train with the student model. OKD achieves or exceeds the performance of leading methods in various model architectures and sizes, reducing training time by up to fourfold.
  arXiv  Detail & Related papers  (2024-09-19T07:05:26Z)
• Augmentation-Free Dense Contrastive Knowledge Distillation for Efficient Semantic Segmentation [16.957139277317005]
  Augmentation-free Dense Contrastive Knowledge Distillation (Af-DCD) is a new contrastive distillation learning paradigm. Af-DCD trains compact and accurate deep neural networks for semantic segmentation applications.
  arXiv  Detail & Related papers  (2023-12-07T09:37:28Z)
• Improved knowledge distillation by utilizing backward pass knowledge in neural networks [17.437510399431606]
  Knowledge distillation (KD) is one of the prominent techniques for model compression. In this work, we generate new auxiliary training samples based on extracting knowledge from the backward pass of the teacher. We show how this technique can be used successfully in applications of natural language processing (NLP) and language understanding.
  arXiv  Detail & Related papers  (2023-01-27T22:07:38Z)
• EmbedDistill: A Geometric Knowledge Distillation for Information Retrieval [83.79667141681418]
  Large neural models (such as Transformers) achieve state-of-the-art performance for information retrieval (IR) We propose a novel distillation approach that leverages the relative geometry among queries and documents learned by the large teacher model. We show that our approach successfully distills from both dual-encoder (DE) and cross-encoder (CE) teacher models to 1/10th size asymmetric students that can retain 95-97% of the teacher performance.
  arXiv  Detail & Related papers  (2023-01-27T22:04:37Z)
• Compressing Visual-linguistic Model via Knowledge Distillation [43.73998154661652]
  We study knowledge distillation to compress a transformer-based large visual-linguistic model into a small model. We show that our proposed distillation significantly improves the performance of small VL models on image captioning and visual question answering tasks.
  arXiv  Detail & Related papers  (2021-04-05T18:02:17Z)
• Generative Adversarial Simulator [2.3986080077861787]
  We introduce a simulator-free approach to knowledge distillation in the context of reinforcement learning. A key challenge is having the student learn the multiplicity of cases that correspond to a given action. This is the first demonstration of simulator-free knowledge distillation between a teacher and a student policy.
  arXiv  Detail & Related papers  (2020-11-23T15:31:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.