SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.09233 · 2026

GIFT: Reconciling Post-Training Objectives via Finite-Temperature Gibbs Initialization

Wentao Zhang, Xiaochen Ma, Haoze Sun, Peng Pei, Lexiang Tang, Zhengyang Zhao, Lu Ma, Chengyu Shen, Yizhen Jiang, Zimo Meng

arXiv · PDF · Open in the Atlas

Code that ran

7 of 16 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; 3 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 3 failures of our instrument; 9 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/2601.09233.json · badge: /api/ran/2601.09233.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
zzy1127/GIFT canonical (classifier label)
classifier basis: census
7 ran · 9 no recorded run
FunctionStatusWhere it lives
compute_ce_dpo_loss_rm Ran
ran; our contract draft was wrong, not the code
zzy1127/GIFT/verl/recipe/prime/prime_core_algos.py
pointer only (licence: NONE); 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: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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
zzy1127/GIFT/verl/recipe/prime/prime_core_algos.py
pointer only (licence: NONE); 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: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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.

extract_boxed_content Ran
ran; our contract draft was wrong, not the code
zzy1127/GIFT/utils/polymath/run_eval.py
pointer only (licence: NONE); the stored copy get_code serves, from QwenLM/PolyMath, is pointer only (licence: NONE) · get_code("0db9401e2d8d130e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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.

extract_step Ran
ran, honoured the contract we drafted
zzy1127/GIFT/verl/verl/trainer/fsdp_gift_trainer.py
pointer only (licence: NONE) · get_code("216e28c040173a61")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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.

is_digit Ran
ran, violated the contract we drafted
zzy1127/GIFT/utils/polymath/scripts.py
pointer only (licence: NONE); the stored copy get_code serves, from QwenLM/Qwen2.5-Math, is pointer only (licence: NONE) · get_code("fb37d3c84b18e368")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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.

is_digit Ran
ran, honoured the contract we drafted
zzy1127/GIFT/utils/openmathinst_utils.py
pointer only (licence: NONE) · get_code("2a180d3156997924")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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.

parse_digits Ran
ran, honoured the contract we drafted
zzy1127/GIFT/utils/polymath/scripts.py
pointer only (licence: NONE); the stored copy get_code serves, from QwenLM/Qwen2.5-Math, is pointer only (licence: NONE) · get_code("7d4b7f8c3875b126")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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.

check_conversation_format Not yet run
unverified (harvested, no recorded run)
zzy1127/GIFT/utils/data_utils.py
pointer only (licence: NONE) · get_code("875194b0e06d05e1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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. No invocation recorded.

math_equal_ray Not yet run
unverified (harvested, no recorded run)
zzy1127/GIFT/utils/reward_utils/my_reward_func.py
pointer only (licence: NONE) · get_code("dadd2ed3c9c1e9fe")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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. No invocation recorded.

math_equal_ray Not yet run
unverified (harvested, no recorded run)
zzy1127/GIFT/utils/reward_utils/reward_func.py
pointer only (licence: NONE) · get_code("7aaa2eb334380beb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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. No invocation recorded.

normalize Not yet run
unverified (harvested, no recorded run)
zzy1127/GIFT/utils/openmathinst_utils.py
pointer only (licence: NONE) · get_code("6b11d915b45fae0d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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. No invocation recorded.

normalize_answer_string Not yet run
unverified (harvested, no recorded run)
zzy1127/GIFT/utils/openmathinst_utils.py
pointer only (licence: NONE) · get_code("11d097e9be003804")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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. No invocation recorded.

numeric_equal Not yet run
unverified (harvested, no recorded run)
zzy1127/GIFT/utils/polymath/scripts.py
pointer only (licence: NONE) · get_code("a929aa929946ee74")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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. No invocation recorded.

pm_judge Not yet run
unverified (harvested, no recorded run)
zzy1127/GIFT/utils/polymath/judge.py
pointer only (licence: NONE) · get_code("f30ed7799f976908")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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. No invocation recorded.

read_json Not yet run
unverified (harvested, no recorded run)
zzy1127/GIFT/utils/data_utils.py
pointer only (licence: NONE) · get_code("dd9ca8cdc6aa690d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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. No invocation recorded.

read_jsonl Not yet run
unverified (harvested, no recorded run)
zzy1127/GIFT/utils/data_utils.py
pointer only (licence: NONE) · get_code("82f198e67affc0bb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zzy1127/GIFT 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. No invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 12 are also present in other papers whose code we hold (12 with at least one paper dated earlier, 4 the same month, 6 later) and 4 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
compute_ce_dpo_loss_rm 46 other papers
11 dated earlier, 4 the same month, 28 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
9 dated earlier, 4 the same month, 28 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
extract_step 19 other papers
12 dated earlier, 6 later, 1 undated
2410.21236 (2024-10)
Flaming-hot Initiation with Regular Execution Sampling for Large Language Models
2025.findings-emnlp.1015 (2025 (year only))
MT-R1-Zero: Advancing LLM-based Machine Translation via R1-Zero-like Reinforcement Learning
2502.14768 (2025-02)
Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning
and 16 more not named here
is_digit 18 other papers
14 dated earlier, 1 the same month, 2 later, 1 undated
2405.14365 (2024-05)
JiuZhang3.0: Efficiently Improving Mathematical Reasoning by Training Small Data Synthesis Models
2410.07985 (2024-10)
Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models
2502.06703 (2025-02)
Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling
and 15 more not named here
parse_digits 18 other papers
14 dated earlier, 1 the same month, 2 later, 1 undated
2405.14365 (2024-05)
JiuZhang3.0: Efficiently Improving Mathematical Reasoning by Training Small Data Synthesis Models
2410.07985 (2024-10)
Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models
2502.06703 (2025-02)
Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling
and 15 more not named here
is_digit 7 other papers
6 dated earlier, 1 later
2404.02078 (2024-04)
Advancing LLM Reasoning Generalists with Preference Trees
2412.01981 (2024-12)
Free Process Rewards without Process Labels
2502.01456 (2025-02)
Process Reinforcement through Implicit Rewards
and 4 more not named here
check_conversation_format 1 other paper
1 dated earlier
2505.18822 (2025-05)
AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting
extract_boxed_content 1 other paper
1 dated earlier
2504.18428 (2025-04)
PolyMath: Evaluating Mathematical Reasoning in Multilingual Contexts
normalize 1 other paper
1 dated earlier
2505.18822 (2025-05)
AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting
normalize_answer_string 1 other paper
1 dated earlier
2505.18822 (2025-05)
AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting
read_json 1 other paper
1 dated earlier
2505.18822 (2025-05)
AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting
read_jsonl 1 other paper
1 dated earlier
2505.18822 (2025-05)
AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting

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

The prevailing post-training paradigm for Large Reasoning Models (LRMs)-Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL)-suffers from an intrinsic optimization mismatch: the rigid supervision inherent in SFT induces distributional collapse, thereby exhausting the exploration space necessary for subsequent RL. In this paper, we reformulate SFT to reconcile post-training objectives and propose Gibbs Initialization with Finite Temperature (GIFT). We characterize standard SFT as a degenerate zero-temperature limit that suppresses base priors. Conversely, GIFT incorporates supervision as a finitetemperature energy potential, establishing a distributional bridge that promotes objective consistency throughout the post-training pipeline. Our experiments demonstrate that GIFT significantly outperforms standard SFT and other competitive baselines when utilized for RL initialization, providing a mathematically principled pathway to preserve exploration and align the two post-training stages. Our code is available at https://github.com/zzy1127/GIFT.

For agents

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

get_harvested_code_for_paper("2601.09233")
get_code_for_paper("2601.09233")
have("2601.09233")

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

curl https://syntology.ai/api/ran/2601.09233.json

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

[![Syntology run record](https://syntology.ai/api/ran/2601.09233.svg)](https://syntology.ai/paper/2601.09233)

Connect an agent — have() is free.