SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.07079 · ICML · 2026

Entropy-Aware On-Policy Distillation of Language Models

Yi Zhou, Kimin Lee, Dennis Wei, Nathalie Baracaldo, Taywon Min, Yongjin Yang, Woogyeol Jin, Swanand Ravindra Kadhe

arXiv · PDF · Open in the Atlas

Code that ran

10 of 10 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 9 failures of our instrument; 0 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2603.07079.json · badge: /api/ran/2603.07079.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
WLS04/EOPD canonical (classifier label)
classifier basis not recorded on the link
10 ran · 0 no recorded run
FunctionStatusWhere it lives
compute_ce_dpo_loss_rm Ran
ran; our contract draft was wrong, not the code
WLS04/EOPD/recipe/prime/prime_core_algos.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from LeapLabTHU/OPDVR, is pointer only (licence: NONE) · get_code("f220595adfc5e83c")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from LeapLabTHU/OPDVR, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

compute_detach_dpo_loss_rm Ran
ran; our contract draft was wrong, not the code
WLS04/EOPD/recipe/prime/prime_core_algos.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from LeapLabTHU/OPDVR, is pointer only (licence: NONE) · get_code("4d91e4383716e42c")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from LeapLabTHU/OPDVR, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

compute_online_dpo_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
WLS04/EOPD/recipe/spin/core_algos.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from LeapLabTHU/OPDVR, is pointer only (licence: NONE) · get_code("f19825983a523003")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from LeapLabTHU/OPDVR, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

compute_onlinedpo_pref Ran
ran; our contract draft was wrong, not the code
WLS04/EOPD/recipe/spin/core_algos.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from LeapLabTHU/OPDVR, is pointer only (licence: NONE) · get_code("41b6e81dc3ecba79")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from LeapLabTHU/OPDVR, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

compute_score_data_source Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
WLS04/EOPD/recipe/open_math_reasoning/compute_score.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from LeapLabTHU/OPDVR, is pointer only (licence: NONE) · get_code("05a86ccb0b3876dd")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from LeapLabTHU/OPDVR, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

generate_config_from_args Ran
ran; our contract draft was wrong, not the code
WLS04/EOPD/verl/model_merger/base_model_merger.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from LeapLabTHU/OPDVR, is pointer only (licence: NONE) · get_code("8896f4e2bafe8abc")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from LeapLabTHU/OPDVR, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_dynamic_pipeline_shards Ran
ran, honoured the contract we drafted
WLS04/EOPD/verl/model_merger/megatron_model_merger.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from LeapLabTHU/OPDVR, is pointer only (licence: NONE) · get_code("94209884dcb2308e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from LeapLabTHU/OPDVR, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_kl_controller Ran
ran; our contract draft was wrong, not the code
WLS04/EOPD/recipe/spin/core_algos.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from LeapLabTHU/OPDVR, is pointer only (licence: NONE) · get_code("757d109d90121142")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from LeapLabTHU/OPDVR, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

normalize Ran
ran; our contract draft was wrong, not the code
WLS04/EOPD/recipe/gkd/megatron_kl_loss.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from LeapLabTHU/OPDVR, is pointer only (licence: NONE) · get_code("11c8e99168eeef2e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from LeapLabTHU/OPDVR, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

reward_func Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
WLS04/EOPD/recipe/r1/reward_score.py
code served (permissive licence) · get_code("8a84f8cc921357e2")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Mock. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from unittest.mock import Mock, patch

torch.manual_seed(0)

# Mock the imported modules to avoid import errors
mock_math_reward = Mock()
mock_math_reward.compute_score = Mock(return_value=0.85)

mock_gpqa = Mock()
mock_gpqa.compute_score = Mock(return_value=0.92)

mock_livecodebench = Mock()
mock_livecodebench.compute_score = Mock(return_value=0.78)

# Patch the imports at module level
import sys
sys.modules['recipe'] = Mock()
sys.modules['recipe.r1'] = Mock()
sys.modules['recipe.r1.tasks'] = Mock()
sys.modules['recipe.r1.tasks.math_reward'] = mock_math_reward
sys.modules['recipe.r1.tasks.gpqa'] = mock_gpqa
sys.modules['recipe.r1.tasks.livecodebench'] = mock_livecodebench

ARGS = {
    'data_source': 'Maxwell-Jia/AIME_2024',
    'solution_str': 'The answer is 42',
    'ground_truth': '42',
    'extra_info': None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
recipe, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Mock — a Mock output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 10 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 9 the same month, 10 later) and 0 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
generate_config_from_args 57 other papers
18 dated earlier, 6 the same month, 30 later, 3 undated
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training
2509.25787 (2025-09)
Self-Evolving Vision-Language Models for Image Quality Assessment via Voting and Ranking
2510.08630 (2025-10)
ExPO-HM: Learning to Explain-then-Detect for Hateful Meme Detection
and 54 more not named here
get_dynamic_pipeline_shards 52 other papers
16 dated earlier, 6 the same month, 28 later, 2 undated
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training
2509.25787 (2025-09)
Self-Evolving Vision-Language Models for Image Quality Assessment via Voting and Ranking
2510.08630 (2025-10)
ExPO-HM: Learning to Explain-then-Detect for Hateful Meme Detection
and 49 more not named here
compute_ce_dpo_loss_rm 46 other papers
19 dated earlier, 6 the same month, 18 later, 3 undated
2409.19256 (2024-09)
HybridFlow: A Flexible and Efficient RLHF Framework
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training
2504.16084 (2025-04)
TTRL: Test-Time Reinforcement Learning
and 43 more not named here
compute_detach_dpo_loss_rm 44 other papers
17 dated earlier, 7 the same month, 17 later, 3 undated
2409.19256 (2024-09)
HybridFlow: A Flexible and Efficient RLHF Framework
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training
2504.16084 (2025-04)
TTRL: Test-Time Reinforcement Learning
and 41 more not named here
compute_online_dpo_loss 38 other papers
13 dated earlier, 5 the same month, 17 later, 3 undated
2409.19256 (2024-09)
HybridFlow: A Flexible and Efficient RLHF Framework
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training
2504.16084 (2025-04)
TTRL: Test-Time Reinforcement Learning
and 35 more not named here
get_kl_controller 38 other papers
13 dated earlier, 5 the same month, 17 later, 3 undated
2409.19256 (2024-09)
HybridFlow: A Flexible and Efficient RLHF Framework
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training
2504.16084 (2025-04)
TTRL: Test-Time Reinforcement Learning
and 35 more not named here
compute_onlinedpo_pref 30 other papers
7 dated earlier, 5 the same month, 15 later, 3 undated
2409.19256 (2024-09)
HybridFlow: A Flexible and Efficient RLHF Framework
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training
2504.16084 (2025-04)
TTRL: Test-Time Reinforcement Learning
and 27 more not named here
reward_func 15 other papers
4 dated earlier, 1 the same month, 9 later, 1 undated
2410.02992 (2024-10)
Learning to Better Search with Language Models via Guided Reinforced Self-Training
2601.17687 (2026-01)
Agentic reinforcement learning empowers next-generation chemical language models for molecular design and synthesis
2601.21468 (2026-01)
MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoning
and 12 more not named here
compute_score_data_source 13 other papers
2 dated earlier, 2 the same month, 8 later, 1 undated
2601.21468 (2026-01)
MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoning
2602.17550 (2026-02)
MASPO: Unifying Gradient Utilization, Probability Mass, and Signal Reliability for Robust and Sample-Efficient LLM Reasoning
2603.02938 (2026-03)
Beyond One-Size-Fits-All: Adaptive Subgraph Denoising for Zero-Shot Graph Learning with Large Language Models
and 10 more not named here
normalize 6 other papers
1 dated earlier, 4 later, 1 undated
2602.17550 (2026-02)
MASPO: Unifying Gradient Utilization, Probability Mass, and Signal Reliability for Robust and Sample-Efficient LLM Reasoning
2606.06021 (2026-06)
OPRD: On-Policy Representation Distillation
2606.19236 (2026-06)
STARE: Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy Stability
and 3 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

On-policy distillation is a promising approach for transferring knowledge between language models, where a student learns from dense token-level signals along its own trajectories. The standard objective is reverse KL divergence, which encourages the student to match the teacher's highconfidence predictions. However, we show that the mode-seeking property of reverse KL reduces generation diversity and yields unstable learning signals when the teacher distribution has high entropy. To address this, we introduce Entropy-Aware On-Policy Distillation (EOPD), which augments the reverse KL objective with forward KL on tokens where the teacher distribution has high entropy. This captures the full range of plausible outputs at uncertain steps while retaining precise imitation elsewhere, balancing mode-seeking precision with mode-covering robustness without sacrificing on-policy training efficiency. Experiments show that our method maintains generation diversity (sustained token-level entropy) and improves student-teacher alignment (lower forward KL on high-entropy tokens). Across six math reasoning benchmarks, this yields Pass@8 accuracy gains of +1.37 for Qwen3-0.6B-Base, +2.39 for Qwen3-1.7B-Base, and +5.05 for Qwen3-4B-Base compared to baseline on-policy distillation methods. These results demonstrate that accounting for teacher uncertainty is essential for maintaining diversity and achieving effective knowledge transfer. Our code is publicly available at https://github.com/WLS04/EOPD.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2603.07079")
get_code_for_paper("2603.07079")
have("2603.07079")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2603.07079.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2603.07079.svg)](https://syntology.ai/paper/2603.07079)

Connect an agent — have() is free.