Home › Code › get_kl_controller

get_kl_controller

Syntologyentry name in harvested coderead from the graph 2026-09-24

get_kl_controller appears in the code Syntology harvested for 53 papers, as 6 distinct code bodies found in 53 places (a place is one code body under one paper). At least one of them ran in 46 of the papers; 0 of the code bodies carry a behaviour fingerprint.

What this page is not. Routines are grouped here by the exact string of their function or class name. Nothing asserts that two samples named get_kl_controller do the same thing, share code, or are comparable; the name is a string, not an identity. Behaviour outputs (what a fingerprinted sample returned on the shared battery) are not in this export and are not shown here; the graph at syntology.ai holds them. "Ran" means executed on a synthesized fixture, not that the code is correct or reproduces a paper.

Samples Syntology

Syntology ran 3 of the 6 distinct code bodies named get_kl_controller; 3 are unverified. One tile per status, in the site's fixed vocabulary, each code body counted once:

0ran · honoured contract
0ran · violated contract
0ran · our draft was wrong
0ran · fixture could not drive it
3ran
3unverified
0fingerprinted

Licence is a property of each copy, so it is counted per place: 18 of the 53 places are pointer only (Syntology does not serve that copy's text). This site shows no code text for any sample; every row below links to the file in its repository where the record names one.

“Ran” means the sample executed on a synthesized input; it does not mean the output is correct. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code, and those samples did run. The ran count above is every status except unverified, the same rule as each paper page.

Papers

53 papers shown of 53, newest first; 53 places in the table. A paper with no recorded date is placed by the month its arXiv id encodes, shown in the Date column as YYYY-MM (from id). One row per place: a paper whose repository defines the name more than once appears more than once, and the same code body held for several papers appears once under each, with the same status. Titles and dates are the archive's archive 2025-07-28 for papers in the archive, and the graph's for 39 papers added by Syntology; 4 papers have no page here and are shown by arXiv id only. Status and fingerprint are Syntology's record of each code body; licence is recorded for each place. The File cell ends with the code body's code_sha256, Syntology's identity for that exact code: an agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

PaperDateFileStatus SyntologyLicence
On-policy Distillation with Verifiable Reward added by Syntology 2026-08 (from id) LeapLabTHU/OPDVR/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
Mismatch Matters: On-Policy Distillation Beyond Token Agreement added by Syntology 2026-08 (from id) yzc-666/TIDE/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
Toward Plasticity-Preserving KL Regularization for Capability Retention in LLM Reinforcement Learning added by Syntology 2026-08 (from id) Lumina04/CoKL/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
ARMOR: Stabilizing On-Policy LLM RL with Off-Policy Anchor Samples added by Syntology 2026-07 (from id) Hesse73/ARMOR/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
OPID: On-Policy Skill Distillation for Agentic Reinforcement Learning added by Syntology 2026-06 (from id) jinyangwu/OPID/recipe/spin/core_algos.py 757d109d90121142 ran MIT (permissive)
OPERA: Aligning Open-Ended Reasoning via Objective Perplexity-based Reinforcement Learning added by Syntology 2026-06 (from id) pangpang-xuan/OPERA/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
ExTra: Exploratory Trajectory Optimization for Language Model Reinforcement Learning added by Syntology 2026-06 (from id) allen4747/extra/verl/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
STARE: Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy Stability added by Syntology 2026-06 (from id) hp-luo/STARE/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
How Does Reasoning Flow? Tracing Attention-Induced Information Flow for Targeted RL in LLMs added by Syntology 2026-06 (from id) Jiayi-Pan/TinyZero/verl/trainer/ppo/core_algos.py 45d08b3300b3da88 ran Apache-2.0 (permissive)
SAW: Stage-Aware Dynamic Weighting for Multi-Objective Reinforcement Learning in Large Language Models added by Syntology 2026-06 (from id) Zhaolutuan/SAW/verl/trainer/ppo/core_algos.py 45d08b3300b3da88 ran Apache-2.0 (permissive)
OPRD: On-Policy Representation Distillation added by Syntology 2026-06 (from id) ShenzhiYang2000/OPRD/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy Distillation added by Syntology 2026-06 (from id) YuYingLi0/FiRe-OPD/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
When Self-Belief Misleads: Active Label Acquisition for Reinforcement Learning with Verifiable Rewards added by Syntology 2026-05 (from id) Lumina04/CARE/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
Learning to Foresee: Unveiling the Unlocking Efficiency of On-Policy Distillation added by Syntology 2026-05 (from id) caiyuchen-ustc/EffOPD/EffOPD/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
SEIF: Self-Evolving Reinforcement Learning for Instruction Following added by Syntology 2026-05 (from id) Rainier-rq1/SEIF/verl/trainer/core_algos.py 33a3cccef610855c ran Apache-2.0 (permissive)
When Less is Enough: Efficient Inference via Collaborative Reasoning added by Syntology 2026-05 (from id) fairytale9/llm_bottleneck/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
HEALing Entropy Collapse: Enhancing Exploration in Few-Shot RLVR via Hybrid-Domain Entropy Dynamics Alignment added by Syntology 2026-04 (from id) XMUDeepLIT/HEAL/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
CoSearch: Joint Training of Reasoning and Document Ranking via Reinforcement Learning for Agentic Search added by Syntology 2026-04 (from id) snap-research/CoSearch/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
Seeing Isn't Believing: Mitigating Belief Inertia via Active Intervention in Embodied Agents added by Syntology 2026-04 (from id) WangHanLinHenry/EVU/verl-agent/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
Can LLMs Learn to Reason Robustly under Noisy Supervision? added by Syntology 2026-04 (from id) ShenzhiYang2000/OLR/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
Experience is the Best Teacher: Motivating Effective Exploration in Reinforcement Learning for LLMs added by Syntology 2026-03 (from id) sikelifei/HeRL/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time added by Syntology 2026-03 (from id) Jasper-Yan/SCRL/verl/recipe/spin/core_algos.py 757d109d90121142 ran MIT (permissive)
REAL: Regression-Aware Reinforcement Learning for LLM-as-a-Judge added by Syntology 2026-03 (from id) YasminZhang/REAL/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
On Information Self-Locking in Reinforcement Learning for Active Reasoning of LLM agents added by Syntology 2026-03 (from id) unimpor/T3/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
Entropy-Aware On-Policy Distillation of Language Models added by Syntology 2026-03 (from id) WLS04/EOPD/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
Bootstrapping Exploration with Group-Level Natural Language Feedback in Reinforcement Learning added by Syntology 2026-03 (from id) LuckyyySTA/GOLF/golf/verl/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
SaFeR-ToolKit: Structured Reasoning via Virtual Tool Calling for Multimodal Safety added by Syntology 2026-03 (from id) hiyouga/EasyR1/verl/trainer/core_algos.py 7633fd09bcf8972a unverified Apache-2.0 (permissive)
MASPO: Unifying Gradient Utilization, Probability Mass, and Signal Reliability for Robust and Sample-Efficient LLM Reasoning added by Syntology 2026-02 (from id) FlyTune/MASPO-RL/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
Beyond Static Pipelines: Learning Dynamic Workflows for Text-to-SQL added by Syntology 2026-02 (from id) Satissss/SquRL/verl/trainer/ppo/core_algos.py 45d08b3300b3da88 ran Apache-2.0 (permissive)
Echo: Towards Advanced Audio Comprehension via Audio-Interleaved Reasoning added by Syntology 2026-02 (from id) wdqqdw/Echo/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
Agent-Omit: Adaptive Context Omission for Efficient LLM Agents added by Syntology 2026-02 (from id) usail-hkust/Agent-Omit/AgentOmit-RL/verl/agent_trainer/ppo/core_algos.py 45d08b3300b3da88 ran no licence file found · pointer only
MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoning added by Syntology 2026-01 (from id) meituan/MemOCR/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
Agentic reinforcement learning empowers next-generation chemical language models for molecular design and synthesis added by Syntology 2026-01 (from id) HowardLi1984/ChemCraft/chemcraft_rl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
RelayLLM: Efficient Reasoning via Collaborative Decoding added by Syntology 2026-01 (from id) Chengsong-Huang/RelayLLM/RL_stage/verl/trainer/core_algos.py 7633fd09bcf8972a unverified no licence file found · pointer only
AgentOCR: Reimagining Agent History via Optical Self-Compression added by Syntology 2026-01 (from id) langfengQ/AgentOCR/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
AT 2 PO: Agentic Turn-based Policy Optimization via Tree Search added by Syntology 2026-01 (from id) zzfoutofspace/ATPO/ATPO/verl_atpo/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
Ophiuchus: Incentivizing Tool-augmented "Think with Images" for Joint Medical Segmentation, Understanding and Reasoning added by Syntology 2025-12 (from id) SII-zyj/Ophiuchus/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
Mitigating the Safety Alignment Tax with Null-Space Constrained Policy Optimization added by Syntology 2025-12 (from id) ivanniu/NSPO/verl/recipe/spin/core_algos.py 757d109d90121142 ran no licence file found · pointer only
Multi-Agent Tool-Integrated Policy Optimization added by Syntology 2025-10 (from id) mzf666/MATPO/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
Scaling RL to Long Videos 10 Jul 2025 NVlabs/Long-RL/verl/trainer/core_algos.py 7633fd09bcf8972a unverified Apache-2.0 (permissive)
Reinforcement Fine-Tuning Naturally Mitigates Forgetting in Continual Post-Training 7 Jul 2025 zhhvvv/rft_vs_sft/EasyR1/verl/trainer/core_algos.py 33a3cccef610855c ran no licence file found · pointer only
SoundMind: RL-Incentivized Logic Reasoning for Audio-Language Models 15 Jun 2025 xid32/SoundMind/recipe/spin/core_algos.py 757d109d90121142 ran MIT (permissive)
Reinforcement Learning Tuning for VideoLLMs: Reward Design and Data Efficiency 2 Jun 2025 appletea233/temporal-r1/verl/trainer/core_algos.py d61a07d67d82ec38 unverified Apache-2.0 (permissive)
ARPO:End-to-End Policy Optimization for GUI Agents with Experience Replay 22 May 2025 dvlab-research/arpo/verl/trainer/core_algos.py 33a3cccef610855c ran Apache-2.0 (permissive)
GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoning 16 May 2025 yueliu1999/guardreasoner-vl/train/EasyR1/verl/trainer/core_algos.py b5973cf2aefac349 unverified MIT (permissive)
TTRL: Test-Time Reinforcement Learning 22 Apr 2025 prime-rl/ttrl/verl/recipe/spin/core_algos.py 757d109d90121142 ran MIT (permissive)
OpenVLThinker: An Early Exploration to Complex Vision-Language Reasoning via Iterative Self-Improvement 21 Mar 2025 yihedeng9/openvlthinker/EasyR1/verl/trainer/core_algos.py 7633fd09bcf8972a unverified Apache-2.0 (permissive)
Guided Stream of Search: Learning to Better Search with Language Models via Optimal Path Guidance 3 Oct 2024 snu-mllab/guided-rest/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
HybridFlow: A Flexible and Efficient RLHF Framework 28 Sep 2024 du-nlp-lab/lengthreward/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
arXiv:openreview_v70fTOqer2 liziniu/KnapsackRL/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
arXiv:openreview_lbaBsu0CaY Tunanzzz/Meerkat-VL/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)
arXiv:openreview_Y1PXB8HBV7 youwyouw/mirl-main/verl/trainer/core_algos.py 7633fd09bcf8972a unverified Apache-2.0 (permissive)
arXiv:openreview_Tkdgg5uqNK ZhijianZhou/Disppo/recipe/spin/core_algos.py 757d109d90121142 ran Apache-2.0 (permissive)

This site shows no code text; each File cell links to the file on GitHub at the repository's current default branch, which may have changed since the harvest. "Pointer only" means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence cell for the reason. Per-sample records for a paper are on its paper page under "Code Syntology ran".

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections