Teacher top-K distillation drops the low-probability <tool call> token from the response teacher's support, creating a one-sided gradient that causally raises tool over-calling.
hub Canonical reference
Yoshua Bengio, Jérôme Louradour, Ronan Collobert, and Jason Weston
Canonical reference. 100% of citing Pith papers cite this work as background.
33
Pith papers citing it
Background
100% of classified citations
abstract
Knowledge distillation (KD) is widely used for compressing a teacher model to reduce its inference cost and memory footprint, by training a smaller student model. However, current KD methods for auto-regressive sequence models suffer from distribution mismatch between output sequences seen during training and those generated by the student during inference. To address this issue, we introduce Generalized Knowledge Distillation (GKD). Instead of solely relying on a fixed set of output sequences, GKD trains the student on its self-generated output sequences by leveraging feedback from the teacher on such sequences. Unlike supervised KD approaches, GKD also offers the flexibility to employ alternative loss functions between the student and teacher, which can be useful when the student lacks the expressivity to mimic the teacher's distribution. Furthermore, GKD facilitates the seamless integration of distillation with RL fine-tuning (RLHF). We demonstrate the efficacy of GKD for distilling auto-regressive language models on summarization, translation, and arithmetic reasoning tasks, and task-agnostic distillation for instruction-tuning.
hub tools
citation-role summary
background 5
citation-polarity summary
roles
background 5polarities
background 5representative citing papers
Purified OPSD subtracts a reference-only teacher's signal from standard OPSD supervision and applies PMI to create a cleaner distillation target, yielding gains on long-CoT models while preserving epistemic behavior.
PrincipalBench exposes a sharp split in frontier LLMs between selective and over-refusing behavior on multi-party loyalty, with prompt scaffolding and KL distillation reducing harm rates but only along an existing leak/over-refusal trade-off.
RGSD distills rubric-conditioned teacher distributions into base policies token-by-token, matching GRPO rubric satisfaction on Qwen models with one rollout and zero verifier calls.
OPD updates occupy a relaxed off-principal regime and rapidly lock into a low-dimensional subspace that is functionally sufficient for its performance, distinct from SFT and RLVR trajectories.
vOPD stabilizes on-policy distillation gradients by subtracting a closed-form per-token negative reverse KL baseline as a detached control variate, preserving unbiasedness while lowering variance and matching expensive full-vocabulary methods.
New RPS and AGS metrics show within-family distilled LLM agents have 5.9 pp higher tool-use graph similarity than cross-family pairs, with some models exceeding their teachers.
TurnOPD improves on-policy distillation for long-horizon agents by adaptively budgeting rollout depth and progressively shifting KL loss from token-level to turn-balanced weighting, achieving up to 2.29x faster training with better accuracy on ALFWorld, WebShop, and Multi-Hop Search.
Transferring a weak model’s RL-induced log-ratio policy shift on a strong student’s own rollouts raises AIME accuracy more cheaply than imitating the weak teacher or running matched-step RL on the student.
PHF distills token-to-token transition directions and trajectory geometry in hidden states during on-policy self-distillation, reporting 1.5-2.2 point gains on Average@12 for Qwen3-1.7B/4B/8B over reproduced OPSD baseline under a 100-step schedule.
SGCD reshapes GRPO token advantages via detached sibling-contrast credit from an external LLM, improving AppWorld and airline tool-use scores while keeping policy gradient as the actor update.
Visual-SDPO distills visual feedback from rendered code outputs into a student policy via grounded credit weighting and GRPO, yielding over 10-point gains on chart/UI/slide benchmarks.
RAFT improves domain accuracy by 23.2% over standard SFT while recovering 18.2% and 10.2% relative performance on MS-Bench and IFEval through refined supervision and trajectory-preserving distillation.
FA-OPD co-trains a flow-matching teacher and MLP student via adversarial dual on-policy distillation, improving robustness over baselines on six robot benchmarks with noisy or limited demonstrations.
A multi-teacher distillation framework that packs 50 effect LoRAs and fast sampling into a single adapter while aiming to avoid concept interference.
Position-Weighted On-Policy Self-Distillation (PW-OPSD) weights later tokens more heavily after a diagnostic shows position predicts teacher reliability better than entropy, yielding +1.0 and +1.1 Avg@12 gains on AIME 2024/2025.
RESD turns failure trajectories into token-level supervision via retrospective reflections and a persistent global playbook, enabling faster improvement than standard self-distillation or GRPO with only one rollout per prompt.
A two-axis taxonomy of student entropy and teacher-student divergence identifies informative tokens in on-policy distillation, allowing near-full performance with 10-50% of tokens.
A pruning technique called Reasoning-Aware Compression (RAC) jointly reconstructs input and chain-of-thought activations to preserve reasoning performance better than standard methods when compressing models like DeepSeek-R1.
SCoRe uses multi-turn online RL with regularization on self-generated traces to improve LLM self-correction, achieving 15.6% and 9.1% gains on MATH and HumanEval for Gemini models.
ASVD compresses LLMs by 10-30% and KV caches by 50% via activation-aware SVD that absorbs outliers into transformed weights and calibrates per-layer sensitivity.
MiniLLM distills large language models into smaller ones via reverse KL divergence and on-policy optimization, yielding higher-quality responses with lower exposure bias than standard KD baselines.
TREK uses verified teacher proposals to expand a student model's exploration support before standard GRPO refinement, improving performance on hard math and agentic tasks.
V-Zero trains MLLMs for visual reasoning without answer labels by gating on-policy distillation trajectories using contrastive evidence from relevant versus negative image crops.
citing papers explorer
-
When Top-K Misses the Decision: Tool-Call Drift in Multi-Teacher On-Policy Distillation
Teacher top-K distillation drops the low-probability <tool call> token from the response teacher's support, creating a one-sided gradient that causally raises tool over-calling.
-
Purified OPSD: On-Policy Self-Distillation Without Losing How to Think
Purified OPSD subtracts a reference-only teacher's signal from standard OPSD supervision and applies PMI to create a cleaner distillation target, yielding gains on long-CoT models while preserving epistemic behavior.
-
Whose Side Is Your Agent On? Multi-Party Principal Loyalty in LLM Agents
PrincipalBench exposes a sharp split in frontier LLMs between selective and over-refusing behavior on multi-party loyalty, with prompt scaffolding and KL distillation reducing harm rates but only along an existing leak/over-refusal trade-off.
-
Rubric-Guided Self-Distillation: Post-Training Without Rubric Verifiers
RGSD distills rubric-conditioned teacher distributions into base policies token-by-token, matching GRPO rubric satisfaction on Qwen models with one rollout and zero verifier calls.
-
On the Geometry of On-Policy Distillation
OPD updates occupy a relaxed off-principal regime and rapidly lock into a low-dimensional subspace that is functionally sufficient for its performance, distinct from SFT and RLVR trajectories.
-
KL for a KL: On-Policy Distillation with Control Variate Baseline
vOPD stabilizes on-policy distillation gradients by subtracting a closed-form per-token negative reverse KL baseline as a detached control variate, preserving unbiasedness while lowering variance and matching expensive full-vocabulary methods.
-
When Agents Look the Same: Quantifying Distillation-Induced Similarity in Tool-Use Behaviors
New RPS and AGS metrics show within-family distilled LLM agents have 5.9 pp higher tool-use graph similarity than cross-family pairs, with some models exceeding their teachers.
-
TurnOPD: Making On-Policy Distillation Turn-Aware for Efficient Long-Horizon Agent Training
TurnOPD improves on-policy distillation for long-horizon agents by adaptively budgeting rollout depth and progressively shifting KL loss from token-level to turn-balanced weighting, achieving up to 2.29x faster training with better accuracy on ALFWorld, WebShop, and Multi-Hop Search.
-
Weak-to-Strong Generalization via Direct On-Policy Distillation
Transferring a weak model’s RL-induced log-ratio policy shift on a strong student’s own rollouts raises AIME accuracy more cheaply than imitating the weak teacher or running matched-step RL on the student.
-
PHF: Privileged Hidden Flow for On-Policy Self-Distillation
PHF distills token-to-token transition directions and trajectory geometry in hidden states during on-policy self-distillation, reporting 1.5-2.2 point gains on Average@12 for Qwen3-1.7B/4B/8B over reproduced OPSD baseline under a 100-step schedule.
-
Keep Policy Gradient in Charge: Sibling-Guided Credit Distillation for Long-Horizon Tool-Use Agents
SGCD reshapes GRPO token advantages via detached sibling-contrast credit from an external LLM, improving AppWorld and airline tool-use scores while keeping policy gradient as the actor update.
-
Self-Distillation Policy Optimization via Visual Feedback: Bridging Code and Visual Artifacts
Visual-SDPO distills visual feedback from rendered code outputs into a student policy via grounded credit weighting and GRPO, yielding over 10-point gains on chart/UI/slide benchmarks.
-
RAFT: Data Refinement and Adaptive Distillation for Domain Fine-Tuning with Alleviated Forgetting
RAFT improves domain accuracy by 23.2% over standard SFT while recovering 18.2% and 10.2% relative performance on MS-Bench and IFEval through refined supervision and trajectory-preserving distillation.
-
Adversarial Dual On-Policy Distillation from Expressive Teacher
FA-OPD co-trains a flow-matching teacher and MLP student via adversarial dual on-policy distillation, improving robustness over baselines on six robot benchmarks with noisy or limited demonstrations.
-
CollectionLoRA: Collecting 50 Effects in 1 LoRA via Multi-Teacher On-Policy Distillation
A multi-teacher distillation framework that packs 50 effect LoRAs and fast sampling into a single adapter while aiming to avoid concept interference.
-
When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning
Position-Weighted On-Policy Self-Distillation (PW-OPSD) weights later tokens more heavily after a diagnostic shows position predicts teacher reliability better than entropy, yielding +1.0 and +1.1 Avg@12 gains on AIME 2024/2025.
-
Learning with Rare Success but Rich Feedback via Reflection-Enhanced Self-Distillation
RESD turns failure trajectories into token-level supervision via retrospective reflections and a persistent global playbook, enabling faster improvement than standard self-distillation or GRPO with only one rollout per prompt.
-
TIP: Token Importance in On-Policy Distillation
A two-axis taxonomy of student entropy and teacher-student divergence identifies informative tokens in on-policy distillation, allowing near-full performance with 10-50% of tokens.
-
Reasoning Models Can be Accurately Pruned Via Chain-of-Thought Reconstruction
A pruning technique called Reasoning-Aware Compression (RAC) jointly reconstructs input and chain-of-thought activations to preserve reasoning performance better than standard methods when compressing models like DeepSeek-R1.
-
Training Language Models to Self-Correct via Reinforcement Learning
SCoRe uses multi-turn online RL with regularization on self-generated traces to improve LLM self-correction, achieving 15.6% and 9.1% gains on MATH and HumanEval for Gemini models.
-
ASVD: Activation-aware Singular Value Decomposition for Compressing Large Language Models
ASVD compresses LLMs by 10-30% and KV caches by 50% via activation-aware SVD that absorbs outliers into transformed weights and calibrates per-layer sensitivity.
-
MiniLLM: On-Policy Distillation of Large Language Models
MiniLLM distills large language models into smaller ones via reverse KL divergence and on-policy optimization, yielding higher-quality responses with lower exposure bias than standard KD baselines.
-
TREK: Distill to Explore, Reinforce to Refine
TREK uses verified teacher proposals to expand a student model's exploration support before standard GRPO refinement, improving performance on hard math and agentic tasks.
-
V-Zero: Answer-Label-Free On-Policy Distillation with Contrastive Evidence Gating for Fine-Grained Visual Reasoning
V-Zero trains MLLMs for visual reasoning without answer labels by gating on-policy distillation trajectories using contrastive evidence from relevant versus negative image crops.
-
Scaling Laws for Task-Specific LLM Distillation
Empirical scaling laws for task-specific LLM distillation in quantitative finance indicate that chain-of-thought supervision recovers general knowledge lost during iterative pruning while in-domain performance degrades predictably.
-
DenseSteer: Steering Small Language Models towards Dense Math Reasoning
DenseSteer is an inference-time steering framework that improves small LLMs' accuracy on math reasoning by modulating representations toward dense reasoning patterns with fewer but higher-density steps.
-
Learning to Foresee: Unveiling the Unlocking Efficiency of On-Policy Distillation
On-policy distillation gains efficiency from early foresight in module allocation and update directions, which the proposed EffOPD method exploits for 3x faster training with comparable performance.
-
Near-Policy: Accelerating On-Policy Distillation via Asynchronous Generation and Selective Packing
NPD accelerates on-policy distillation 8.1 times faster than baselines by using asynchronous SFT with Δ-IFD filtering, outperforming standard SFT and enabling a 1B model to achieve 68.73% SOTA score.
-
When Should the Teacher Move? Temporal Coupling and Stability in Self On-Policy Distillation
Isolation periods between teacher updates stabilize self on-policy distillation, and a consolidation-gated refresh rule eliminates collapse across four tasks without per-task retuning.
-
Token-Operations-Oriented Inference Optimization Techniques for Large Models
A survey of large-model inference optimization, organized as a four-layer 'token-operations' taxonomy: multi-model fusion, model optimization, compute-model fusion, and compute-network-model fusion.
-
A Survey on Efficient Inference for Large Language Models
The paper surveys techniques to speed up and reduce the resource needs of LLM inference, organized by data-level, model-level, and system-level changes, with comparative experiments on representative methods.
-
A Brief Overview: On-Policy Self-Distillation In Large Language Models
This overview paper explains the conceptual foundations and design principles of On-Policy Self-Distillation for large language models from a beginner's perspective.
- Constitutional On-Policy Safe Distillation