Probing to Refine: Reinforcement Distillation of LLMs via Explanatory Inversion

• URL: http://arxiv.org/abs/2603.19266v1
• Date: Thu, 26 Feb 2026 23:01:46 GMT
• Title: Probing to Refine: Reinforcement Distillation of LLMs via Explanatory Inversion
• Abstract summary: We introduce a novel framework that moves beyond simple mimicry to instill a deeper conceptual understanding.<n>underlinetextitFirst, to address pattern memorization, Explanatory Inversion (EI) generates targeted explanatory probes''<n>underlinetextitSecond, to improve generalization, Explanatory GRPO (texttEXGRPO) uses a reinforcement learning algorithm with a novel Dialogue Structure Utility Bonus.
• Score: 84.20493238687187
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Distilling robust reasoning capabilities from large language models (LLMs) into smaller, computationally efficient student models remains an unresolved challenge. Despite recent advances, distilled models frequently suffer from superficial pattern memorization and subpar generalization. To overcome these limitations, we introduce a novel distillation framework that moves beyond simple mimicry to instill a deeper conceptual understanding. Our framework features two key innovations. \underline{\textit{First}}, to address pattern memorization, Explanatory Inversion (EI) generates targeted ``explanatory probes'' that compel the student to articulate the underlying logic behind an answer, rather than just memorizing it. \underline{\textit{Second}}, to improve generalization, Explanatory GRPO (\texttt{EXGRPO}) uses a reinforcement learning algorithm with a novel Dialogue Structure Utility Bonus, which explicitly rewards the student for maintaining a coherent reasoning process across these probes. Extensive evaluations on 12 datasets demonstrate significant improvements. Using Gemma-7b as the student model, our method yields an average \textbf{20.39\%} increase over zero-shot performance and a \textbf{6.02\%} improvement over the state-of-the-art distillation baselines. Moreover, models distilled with our method show remarkable training efficiency (e.g., surpassing vanilla fine-tuning with \textbf{10-25\%} training data) and strong generalization to out-of-distribution tasks. Implementation is released at https://github.com/Zhen-Tan-dmml/ExGRPO.git.

Related papers

• Rethinking Reward Supervision: Rubric-Conditioned Self-Distillation [60.55792673956761]
  We propose textbfRubric-Conditioned Self-Distillation, a framework that incorporates rubrics as structured, fine-grained feedback for on-policy self-distillation.<n>We evaluate on a diverse suite of science reasoning benchmarks and results show that rubric-conditioned self-distillation effectively converts rubric-level criteria into token-level guidance.
  arXiv  Detail & Related papers  (2026-06-17T17:54:04Z)
• DeltaPrompts: Escaping the Zero-Delta Trap in Multimodal Distillation [49.98710755440242]
  Distillation enables compact Vision-Language Models (VLMs) to obtain strong reasoning capabilities.<n>We reveal a critical inefficiency in this approach: up to 69% of the prompts in standard chart / document reasoning datasets are effectively zero-delta.<n>We propose a staged synthesis pipeline that repurposes existing datasets as seeds, actively targeting student failure modes to produce better prompts.
  arXiv  Detail & Related papers  (2026-05-15T02:04:12Z)
• Curriculum Learning for Efficient Chain-of-Thought Distillation via Structure-Aware Masking and GRPO [24.91321958525287]
  Distilling Chain-of-Thought (CoT) reasoning from large language models into compact student models presents a fundamental challenge.<n>Existing approaches either compress reasoning into single-step, losing the interpretability that makes CoT valuable.<n>We present a three-stage curriculum learning framework that addresses this capacity mismatch through progressive skill acquisition.
  arXiv  Detail & Related papers  (2026-02-05T05:27:11Z)
• R^3: Replay, Reflection, and Ranking Rewards for LLM Reinforcement Learning [32.16683059021539]
  Large reasoning models (LRMs) aim to solve diverse and complex problems through structured reasoning.<n>Recent advances in group-based policy optimization methods have shown promise in enabling stable advantage estimation without reliance on process-level annotations.<n>We propose a reinforcement learning mechanism named emphtextbfR3 that along three directions: (1) a emphcross-context underlinetextbfReplay strategy that maintains the intra-group advantage, (2) an emphin-context self-underlinetextbfReflection mechanism
  arXiv  Detail & Related papers  (2026-01-27T13:55:34Z)
• Enhancing Reasoning Capabilities in SLMs with Reward Guided Dataset Distillation [0.0]
  We propose AdvDistill, a reward-guided dataset distillation framework.<n>We utilise multiple generations (responses) from a teacher for each prompt and assign rewards based on rule-based verifiers.<n>These varying and normally distributed rewards serve as weights when training student models.
  arXiv  Detail & Related papers  (2025-06-25T20:07:47Z)
• TL;DR: Too Long, Do Re-weighting for Efficient LLM Reasoning Compression [55.37723860832064]
  We propose a dynamic ratio-based training pipeline that does not rely on sophisticated data annotations.<n>We validate our approach across models on DeepSeek-R1-Distill-7B and DeepSeek-R1-Distill-14B and on a diverse set of benchmarks with varying difficulty levels.
  arXiv  Detail & Related papers  (2025-06-03T09:23:41Z)
• FDBPL: Faster Distillation-Based Prompt Learning for Region-Aware Vision-Language Models Adaptation [17.51747913191231]
  We propose large textbfFaster large textbfDistillation-large textbfBased large textbfPrompt large textbfLL (textbfFDBPL)<n>It addresses issues by sharing soft supervision contexts across multiple training stages and implementing accelerated I/O. Comprehensive evaluations across 11 datasets demonstrate superior performance in base-to-new generalization, cross-dataset transfer, and robustness tests, achieving $2.2times$ faster training speed.
  arXiv  Detail & Related papers  (2025-05-23T15:57:16Z)
• AceReason-Nemotron: Advancing Math and Code Reasoning through Reinforcement Learning [50.02117478165099]
  We show that large-scale reinforcement learning can significantly enhance the reasoning capabilities of strong, small- and mid-sized models.<n>We propose a simple yet effective approach: first training on math-only prompts, then on code-only prompts.
  arXiv  Detail & Related papers  (2025-05-22T08:50:47Z)
• Compile Scene Graphs with Reinforcement Learning [69.36723767339001]
  Next-token prediction is the fundamental principle for training large language models (LLMs)<n>We introduce R1-SGG, a multimodal LLM (M-LLM) initially trained via supervised fine-tuning (SFT) on the scene graph dataset.<n>We design a set of graph-centric rewards, including three recall-based variants -- Hard Recall, Hard Recall+Relax, and Soft Recall.
  arXiv  Detail & Related papers  (2025-04-18T10:46:22Z)
• Improving In-Context Learning with Reasoning Distillation [25.377625891065236]
  Language models rely on semantic priors to perform in-context learning.<n>We propose ReDis, a reasoning distillation technique designed to improve the inductive reasoning capabilities of language models.
  arXiv  Detail & Related papers  (2025-04-14T18:59:10Z)
• Distill Not Only Data but Also Rewards: Can Smaller Language Models Surpass Larger Ones? [58.80794196076336]
  Distilling large language models (LLMs) typically involves transferring the teacher model's responses through supervised fine-tuning (SFT)<n>We propose a novel distillation pipeline that transfers both responses and rewards.<n>Our method generates pseudo-rewards through a self-supervised mechanism that leverages the inherent structure of both teacher and student responses.
  arXiv  Detail & Related papers  (2025-02-26T20:50:11Z)
• Dynamic Context Pruning for Efficient and Interpretable Autoregressive Transformers [29.319666323947708]
  We present a novel approach that dynamically prunes contextual information while preserving the model's expressiveness. Our method employs a learnable mechanism that determines which uninformative tokens can be dropped from the context. Our reference implementation achieves up to $2times$ increase in inference throughput and even greater memory savings.
  arXiv  Detail & Related papers  (2023-05-25T07:39:41Z)

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.